Diabetes Mellitus-Associated Atherosclerosis
- 66 Downloads
While diabetes mellitus is most often associated with hypertension, dyslipidemia, and obesity, these factors do not fully account for the increased burden of cardiovascular disease in patients with the disease. This strengthens the need for comprehensive studies investigating the underlying mechanisms mediating diabetic cardiovascular disease and, more specifically, diabetes-associated atherosclerosis. In addition to the recognized metabolic abnormalities associated with diabetes mellitus, upregulation of putative pathological pathways such as advanced glycation end products, the renin-angiotensin system, oxidative stress, and increased expression of growth factors and cytokines have been shown to play a causal role in atherosclerotic plaque formation and may explain the increased risk of macrovascular complications. This review discusses the methods used to assess the development of atherosclerosis in the clinic as well as addressing novel biomarkers of atherosclerosis, such as low-density lipoprotein receptor-1. Experimental models of diabetes-associated atherosclerosis are discussed, such as the streptozocin-induced diabetic apolipoprotein E knockout mouse. Results of major clinical trials with inhibitors of putative atherosclerotic pathways are presented. Other topics covered include the role of HMG-CoA reductase inhibitors and fibric acid derivatives with respect to their lipid-altering ability, as well as their emerging pleiotropic anti-atherogenic actions; the effect of inhibiting the renin-angiotensin system by either ACE inhibition or angiotensin II receptor antagonism; the effect of glycemic control and, in particular, the promising role of thiazolidinediones with respect to their direct anti-atherogenic actions; and newly emerging mediators of diabetes-associated atherosclerosis, such as advanced glycation end products, vascular endothelial growth factor and platelet-derived growth factor. Overall, this review aims to highlight the observation that various pathways, both independently and in concert, appear to contribute toward the pathology of diabetes-associated atherosclerosis. Furthermore, it reflects the need for combination therapy to combat this disease.
The incidence of diabetes mellitus (hereafter referred to as diabetes) continues to increase. Four percent (135 million people) of the worldwide population experienced diabetes in 1995. This is predicted to rise to 5.4% (300 million) of the population by 2025. Furthermore, it is developing countries that will be afflicted with much of the burden, with an estimated rise in cases of 170%, compared with 42% in developed countries, between 1995 and 2025. The associated economic burden must also be considered, particularly costs associated with complications, primarily vascular in origin, arising from diabetes.
Diabetes is associated with an increased risk of cardiovascular disease, such that a person with diabetes has a risk of myocardial infarction (MI) as high as that of a non-diabetic person with a previous MI. In fact, cardiovascular disease accounts for >50% of all deaths in the diabetic population. The large-scale study, MRFIT (Multiple Risk Factor Intervention Trial) reported that after a 12-year follow-up, patients with diabetes had a 3-fold increase in the risk of cardiovascular disease compared with patients without diabetes. Similar results were found in the Framingham study, with diabetic patients having a 2- to 3-fold increase in cardiovascular risk. Similar findings have been reported for those patients with type 1 diabetes. A study of 23 000 patients with type 1 diabetes (part of the Diabetes UK Cohort) demonstrated that this population has an increased risk of cardiovascular and cerebrovascular disease. The prevalence of coronary artery calcification, as measured by electron beam tomography, in a type 1 diabetic population with an average age of 20 years was 15.1% in men and 6.3% in women, compared with 8.8% and 1.2%, respectively, in a study of young adults without diabetes. Furthermore, the EDIC (Epidemiology of Diabetes Interventions and Complications) study demonstrated that carotid intimal-medial thickening (IMT) as measured by B-mode ultrasonography, a surrogate marker of cardiovascular risk, was increased in those with type 1 diabetes compared with control subjects.
This article reviews the underlying mechanisms mediating diabetic cardiovascular disease, more specifically, diabetes-associated atherosclerosis.
1. Anatomy and Mechanisms of Atherosclerosis
The various stages of atherosclerosis are arbitrarily divided into three phases: the fatty streak; the intermediate or fibrofatty lesion; and the fibrous plaque or advanced complicated lesion of atherosclerosis (reviewed by Ross).
The fatty streak consists of an intimal collection of lipid-filled, monocyte-derived macrophages in association with T-lymphocytes (T cells). The accumulation of lipid-filled macrophages, known as foam cells, represents the bulk of the lesion and give the lesion a yellow discoloration when viewed en face. These fatty streaks can be found throughout the vascular tree but are more commonly found in areas of blood-flow changes such as back currents at branches, bifurcations, and curves in the vessels.
The fibrofatty lesion consists of layers of lipid-filled macrophages and T cells alternating with layers of smooth muscle cells surrounded by a poorly developed connective tissue matrix of collagen fibrils, elastic fibers, and proteoglycans.
The fibrous plaque is characterized by a dense cap of fibrous connective tissue that contains numerous smooth muscle cells surrounded by dense layers of connective tissue matrix. The smooth muscle cells take on an unusual appearance; they occupy slit-like or lacunar-like spaces. The fibrous cap may contain macrophages and T cells. The cap covers a deeper layer of macrophages with a core of lipid and necrotic material.
The response to injury model originally formulated in 1973 states that various risk factors, such as those associated with diabetes, may induce some form of endothelial dysfunction. Dysfunctional changes in the endothelium may result in alterations in permeability, adhesive characteristics, and growth factors. These characteristics ultimately lead to the migration and infiltration by monocyte-derived macrophages, as described above in fatty streaks. Through a process of co-stimulation and remodeling, the formation of connective tissue occurs, leading to the development of the fibrous plaque covering a lipid core, particularly in diabetic hyperlipidemic patients.
This inflammatory fibroproliferative response is seen as a protective mechanism; however, in disease states such as diabetes, this response becomes excessive and starts to destroy tissue as well as creating a hyperplastic lesion.
1.1 Animal Models of Diabetes-Associated Atherosclerosis
Other animal models include hamsters and minipigs. Twenty-six weeks after the induction of diabetes with streptozocin, hamsters had hyperglycemia, hyperlipidemia, and fibrous plaques, as well as plaques containing calcium deposits or cholesterol clefts. In addition, deposits of APOE and advanced glycation end products (AGEs) were identified. Pigs are a good physiologic model, as their cardiovascular system has similarities to that of a human. Thus, Chinese Guizhou minipigs fed a high-fat, high-sucrose diet have recently been generated as a model of accelerated atherosclerosis. After 6 months, these animals exhibited insulin resistance, mild diabetes, hypertrophic adipocytes, fat deposits in the liver, loss of pancreatic β-cells, and aortic fatty streak lesions.
1.2 Markers of Atherosclerosis
Animal studies are able to directly quantitate the effect of pharmacologic agents on aortic plaque progression using techniques such as sudan IV staining and subsequent en face analysis or serial sectioning of the aortic sinus. By contrast, clinical studies are limited to utilizing surrogate markers of vascular function as a measure of plaque progression. These include functional techniques such as flow-mediated dilation,[28,29] IMT, systemic arterial compliance, and pulse wave velocity, which have all been validated in the clinic as surrogate markers of coronary artery disease.
More recently, a number of plasma biomarkers have emerged. C-reactive protein (CRP) is one of the more common proinflammatory markers, although there is now controversy as to whether CRP is a marker of disease or does in fact play a role in plaque progression. Other emerging markers include lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1), which is expressed in endothelial cells, smooth muscle cells, and macrophages. In addition to binding oxidized low-density lipoprotein (oxLDL), LOX-1 can bind apoptotic cells, activated platelets, and AGEs. LOX-1 expression is increased in disease states such as hypercholesterolemia, hypertension, and diabetes and is induced by angiotensin II and endothelin-1. Matrix metalloproteinase (MMP)-9 or gelatinase B is an enzyme capable of proteolytically degrading type IV collagen involved in basement membrane formation. MMP-9 has been shown to be expressed in human atherosclerotic plaques, particularly at the shoulder regions of the plaque and around foam cells, with circulating levels of this enzyme increased in patients with type 2 diabetes and coronary artery disease. Protease-activated receptors (PARs) are a group of 7-transmembrane-domain, G-protein-coupled receptors that link tissue injury to responses such as inflammation and tissue repair. PAR-1, PAR-3, and PAR-4 are activated by thrombin, and PAR-2 is activated by trypsin and tissue factor. PARs are expressed in endothelial cells, smooth muscle cells, and platelets, and their expression is increased in human atherosclerotic lesions. Activation of PARs results in an increase in leukocyte rolling, adhesion and recruitment, and platelet adhesion (see Szmitko et al. for a review of emerging markers of cardiovascular disease).
2. Potential for Pharmacological Interventions
The lipid profile of a diabetic patient typically includes low levels of high-density lipoprotein (HDL) cholesterol, normal, but increasingly modified (glycated or oxidized), low-density lipoprotein (LDL) cholesterol levels, and high levels of triglyceride-rich lipoproteins. There is also an associated shift in LDL make-up to a small, dense composition that renders the particles more atherogenic. The UKPDS (United Kingdom Prospective Diabetes Study) demonstrated that an increased concentration of LDL cholesterol is significantly correlated with coronary artery disease risk in diabetic patients, as had been seen in the general population of the Framingham study. Furthermore, a decreased concentration of HDL cholesterol was recognized to be a significant coronary artery disease risk factor. In addition to its role in reverse cholesterol transport, HDL per se has been shown to have an antioxidant effect, as paraoxonase, which is often associated with HDL, can reduce potential interactions between lipid peroxidases and LDL. HDL is also thought to have anti-inflammatory properties, reducing E-selectin and vascular cellular adhesion molecule (VCAM)-1, which are modulators of leukocyte recruitment and implicated in the progression of atherosclerosis. Indeed, dyslipidemia is a modifiable risk for patients with diabetes and must be aggressively treated. Therapies aimed at improving the dyslipidemic profile include statins and fibrates.
2.1.1 HMG-CoA Reductase Inhibitors
HMG-CoA reductase inhibitors (statins) inhibit the rate-limiting step in hepatic cholesterol production. This results in a reduction in total cholesterol, primarily through a reduction in LDL cholesterol. Statins currently used in the clinic include atorvastatin, simvastatin, pravastatin, fluvastatin, and rosuvastatin. These agents vary in their potency and subsequent pleiotropic actions.
In addition to their effects on the lipid profile, statins are now recognized to reduce atherosclerosis by a variety of mechanisms independent of their lipid-lowering properties.
Endothelial nitric oxide synthase (eNOS) has been shown to improve endothelial function and to be regulated by statins via various pathways. Statins upregulate eNOS expression and prevent its downregulation by oxLDL. Simvastatin has been shown to increase eNOS stabilization in the absence of changes in eNOS gene transcription. In addition, statins have been shown to reduce caveolin-1 levels. Caveolin is directly regulated by LDL levels; thus, high LDL levels increase caveolin abundance and, in turn, interact with eNOS to reduce nitric oxide production.
Statins also influence the development of atherosclerotic plaques. These agents have been shown to reduce monocyte chemotaxis and neutrophil-endothelial interactions at cholesterol-lowering doses; modulate smooth muscle cell apoptosis, migration and proliferation, angiogenesis; increase fibrinolytic activity; and influence plaque stability. A study of patients receiving pravastatin 3 months prior to carotid endarterectomy demonstrated that therapy altered plaque composition by reducing lipids, lipid oxidation, inflammation, MMP-2, and cell death and increasing collagen content, implicating an effect on plaque stabilization, albeit in a non-diabetic context. Statins have also been shown to reduce a number of inflammatory markers such as tumor necrosis factor-α (TNFα) and CRP. In addition, statins are also thought to act as antioxidants, reducing lipoprotein oxidation. Specifically, in a study of hypercholesterolemic patients, lovastatin or fluvastatin therapy reduced the susceptibility of LDL to oxidation over 24 weeks. These findings are supported by associated clinical studies showing an improvement in endothelial-dependent forearm blood flow in the absence of LDL lowering. Another study demonstrated improvements in both basal and stimulated endothelial function as early as 1 month after commencing statin therapy and this was independent of changes in serum cholesterol levels. Interestingly, statins have also been shown to upregulate peroxisome proliferator-activated receptor (PPAR)-α.
Studies in animal models have confirmed the anti-atherogenic effects of statins. When compared with control mice, apoE-/- mice fed simvastatin 100 mg/kg daily for 6 weeks had significantly reduced atherosclerosis, as measured by aortic cholesterol content, in the absence of changes in plasma lipids. However, in another study, simvastatin administration to 30-week-old apoE-/- mice resulted in an increase in serum lipid levels associated with increased atherosclerotic plaques in the innominate/brachiocephalic artery. However, staining revealed that the frequency of intraplaque hemorrhage was reduced by almost 50% as was the frequency of calcification, demonstrating a plaque-stabilizing effect of statins. Further supporting the pleiotropic effect of statins, a Dutch group treated high-cholesterol-fed apoE*3-Leiden mice with rosuvastatin and fed other apoE*3-Leiden mice a low cholesterol diet so that their plasma cholesterol levels were matched. Despite having cholesterol levels similar to the untreated mice, the rosuvastatin-treated mice had significantly lower cross-sectional lesion area, lesion size, lesion number, and monocyte adherence and an attenuation in the levels of the inflammatory markers, monocyte chemoattractant protein-1 (MCP-1) and TNFα, thus demonstrating the independent anti-inflammatory and anti-atherogenic effects of statins.
The Heart Protection Study included a cohort of 5963 people with type 2 diabetes who were randomly allocated to receive either simvastatin 40 mg/day or placebo. The study investigated the effect of substantially lowering LDL cholesterol levels by statin therapy on vascular mortality and morbidity in a diabetic population regardless of baseline levels. The study demonstrated that statin therapy in a diabetic population reduces the risk of heart attack, stroke, and revascularization. Simvastatin treatment resulted in a 1 mmol/L reduction in LDL cholesterol level and reduced the risk of major vascular events by ≈25%, to a level similar to that seen in non-diabetic patients. Compared with non-diabetic participants in the study, diabetic patients were older, had significantly lower mean total and LDL cholesterol, and had a higher mean triglyceride concentration.
A post hoc subgroup analysis of 4S (Scandinavian Simvastatin Survival Study) examined the effect of simvastatin in 202 diabetic and 4242 non-diabetic patients with either previous MI or angina pectoris and a total cholesterol between 5.5 and 8.0 mmol/L. After a median of 5.4 years follow-up, treatment was associated with a relative risk of 0.57 (p = 0.087 vs placebo) for total mortality, 0.45 (p = 0.002) for major coronary heart disease (CHD) events, and 0.63 (p = 0.018) for any atherosclerotic event in diabetic patients compared with 0.71 (p = 0.001), 0.68 (p < 0.0001), and 0.74 (p < 0.0001) in non-diabetic patients.
Of the nearly 20 000 hypertensive patients (with ≥3 other risk factors) involved in the ASCOT (Anglo-Scandinavian Cardiac Outcomes Trial), just over 50% of those with a non-fasting cholesterol concentration of ≤6.5 mmol/L went on to participate in the lipid-lowering arm of the study (ASCOT-LLA). About 25% of this group was diabetic. Participants were randomized to receive either atorvastatin 10 mg/day or placebo and were to be followed for 5 years, with the primary endpoint being non-fatal MI and fatal CHD. The study was concluded after 3.3 years because of the excellent results achieved in the statin-treated group. Specifically, treated patients had a significant reduction in fatal and non-fatal stroke (89 vs 121; p = 0.024), total cardiovascular events (389 vs 489; p = 0.0005), and total coronary events (178 vs 247; p = 0.0005) at the end of the study and a reduction in total cholesterol of 1.30 and 1.1 mmol/L after 1 and 3 years, respectively. A similar reduction in CHD events was observed in the diabetic group; however, this did not reach statistical significance, as the study was not powered to make such a comparison in the smaller cohort of diabetic subjects. However, there was a significant odds reduction in all cardiovascular events in the diabetic subgroup similar to that observed amongst the other hypertensive patients studied.
The LIPID (Long-term Intervention with Pravastatin in Ischaemic Disease) study investigated the effect of pravastatin 40 mg/day for 6 years in a group of 9014 patients, of whom 1077 had diabetes and 940 had impaired fasting glucose (IFG) levels. In placebo recipients, the risk of a major CHD event was 61% higher in diabetic patients and 23% higher in patients with IFG than in patients with normal fasting glucose. Statin therapy decreased the risk of a major CHD event in the diabetic group from 23.4% to 19.6% (relative risk reduction [RRR] 19%; p = 0.11). This was not statistically different to the reductions in the other groups. Pravastatin reduced the risk of any cardiovascular event in the diabetic group (RRR 21%; p < 0.008) and in the IFG group (RRR 26%; p = 0.003). Pravastatin significantly reduced the risk of stroke (RRR 39%; p = 0.02) in the diabetic group but not in the IFG group (RRR 42%; p = 0.09). Finally, pravastatin reduced major CHD events to 1 per 18 people in diabetic patients and to 1 per 23 in patients with IFG.
A subgroup analysis of the CARE (Cholesterol and Recurrent Events) trial allowed the investigation of the effect of lowering lipid levels in patients with diabetes with CHD and normal cholesterol levels. Patients received either pravastatin 40 mg/day or placebo and were followed up after 5 years. Of the original cohort for the CARE study, 14.1% of patients had type 2 diabetes; they were older, more obese, and more hypertensive than non-diabetic patients. Pravastatin reduced total and LDL cholesterol and triglycerides and increased HDL cholesterol to a similar extent in both diabetic and non-diabetic patients. There was a higher rate of events in the diabetic group than in the non-diabetic group, including CHD death or non-fatal MI (20% vs 12%; p < 0.001), coronary artery bypass graft surgery, percutaneous transluminal coronary angioplasty, and stroke; this was independent of age and sex. Pravastatin reduced the absolute risk of coronary events by 8.1% and 5.2% and the relative risk by 25% (p = 0.05) and 23% (p < 0.001) in diabetic and non-diabetic patients, respectively. In addition, patients with IFG demonstrated an increased risk of CHD, which was attenuated with pravastatin.
The recently concluded CARDS (Collaborative Atorvastatin Diabetes Study) was terminated 2-years short of its planned duration. A cohort of people aged between 40 and 75 years with type 2 diabetes, no documented history of cardiovascular disease, and at least one other risk factor were randomized to receive either atorvastatin 10 mg/day or placebo. The primary endpoint was time to first occurrence of an acute CHD event, coronary revascularization, or stroke. Patients were followed for an average of 3.9 years. Atorvastatin treatment was associated with a 37% reduction in cardiovascular events. More specifically, atorvastatin decreased the incidence of acute CHD events by 36%, coronary revascularization by 31%, stroke by 48%, and death by 27%.
2.1.2 Fibric Acid Derivatives
Fibric acid derivatives (fibrates), including gemfibrozil and fenofibrate, are currently used for the treatment of dyslipidemia. These agents increase β-oxidation and are PPARα agonists. PPARα is a nuclear receptor, which is highly expressed in metabolic tissues such as the heart, liver, and skeletal muscle. Fibrates increase the transcription of APOA1 and APOA2, which results in higher HDL cholesterol levels. These drugs increase efflux of cholesterol from macrophages through the upregulation of the ATP-binding cassette transporter A1 (ABCA1) and increase hepatic uptake of HDL cholesterol via the upregulation of the scavenger receptor CLA-1/SR-B1. In addition, they regulate triglyceride levels by increasing levels of lipoprotein lipase, an enzyme responsible for the hydrolysis of triglycerides, and decrease APOC3, which inhibits the lipoprotein lipase-mediated breakdown of triglycerides. Furthermore, fibrates modulate LDL cholesterol by shifting the balance from small and dense to large and buoyant particles which render them less atherogenic, as well as increasing their clearance. Fibrates may secondarily mediate cardioprotective effects via an increase in HDL cholesterol levels.
In addition to the recognized effects of fibrates on lipid metabolism, these drugs are now recognized to have direct anti-atherogenic effects. As well as being expressed in metabolic tissues, PPARα has also been shown to be expressed in endothelial cells, smooth muscle cells, and macrophages of the vessel wall. PPARα agonists have been shown to mediate effects on atherosclerosis, inhibiting numerous stages of plaque development including recruitment, adhesion, foam-cell formation, reverse cholesterol transport, migration, and thrombogenicity.[87,88] PPARα agonists have also been shown to mediate anti-inflammatory effects, reducing such markers as CRP, cyclo-oxygenase-2 (COX-2), and interleukin-6 (IL-6). It has recently been suggested that these effects occur via a transrepression pathway, altering the signaling of such transcription factors as nuclear factor-B (NF-κB) and activator protein-1 (AP-1). Furthermore, fibrates have been shown to reduce reactive oxygen species, in particular via the downregulation of various subunits of NAD(P)H oxidase.
Treatment with fenofibrate in apoE-/- mice fed a Western diet resulted in decreased total and esterified cholesterol content in the descending aorta, which was associated with a reduction in MCP-1 (monocyte chemoattractant protein-1) gene expression, although no change in ABCA1 expression was noted. In human APOA1 transgenic apoE-/- mice, fenofibrate increased human APOA1 plasma levels and concomitantly reduced plaque area in the aortic sinus; however, these studies were not performed in the setting of diabetes.
Our group has recently demonstrated that the fibrate gemfibrozil markedly reduces atherosclerosis in streptozoc-in-induced diabetic apoE-/- mice. This was observed in association with a reduction in the diabetes-induced increase in NAD(P)H oxidase subunit expression as well as an attenuation in expression of MMP-2, MMP-9, and MCP-1.
A study of type 2 diabetic patients demonstrated that after 3 months of treatment with ciprofibrate, flow-mediated dilation was significantly increased as was the plasma HDL cholesterol level. Treatment also significantly reduced LDL cholesterol levels and postprandial oxidative stress. Further evidence to support the view that PPARα agonists exert anti-atherosclerotic effects independently of their effects on lipids comes from a study investigating the effect of L162V polymorphism in the PPARα gene. Allele carriers of this mutation had a reduced progression of atherosclerosis and those homozygous for the intron 7 allele had an increased risk of ischemic heart disease; this was observed in the absence of changes in lipid concentrations. Screening for polymorphisms of the PPARα gene revealed mutations that were associated with alterations in lipid profile in diabetic patients but not in healthy people, further linking dyslipidemia and diabetes.
As early as 1971, fibrates were used in clinical trials for the treatment of ischemic heart disease. In addition to their ability to lower cholesterol, there are a number of clinical trials investigating the effect of fibrates on CHD risk. The DAIS (Diabetes Atherosclerosis Intervention Study) assessed the effect of normalizing lipoprotein levels on CHD risk in patients with type 2 diabetes. Four hundred and eighteen men and women with good glycemic control and mild lipoprotein abnormalities were randomized to receive either micronized fenofibrate 200 mg/day or placebo. Treatment was associated with a significant reduction in total cholesterol, LDL cholesterol, and triglycerides and a significant increase in HDL cholesterol. Furthermore, the fenofibrate group had a significantly smaller increase in percentage diameter stenosis and a smaller decrease in minimum lumen diameter in stenotic arteries. Although not adequately powered, the study also showed a reduction in the number of cardiovascular events; 38 in fenofibrate-treated patients compared with 50 in the placebo group.
The results of the FIELD (Fenofibrate Intervention and Event Lowering in Diabetes) trial, a study investigating the effect of fenofibrate on cardiovascular morbidity and mortality in >9000 people with type 2 diabetes, were released in late 2005. The FIELD study investigators reported a non-significant 11% reduction in the incidence of coronary events in the fenofibrate group (non-fatal MI and coronary death; p = 0.16). Interestingly, fenofibrate was associated with a significant reduction in non-fatal MI (p = 0.010) and a non-significant increase in coronary heart disease mortality (p = 0.22). Fenofibrate significantly reduced the incidence of total cardiovascular events (p = 0.035), which included a 21% reduction in revascularization procedures. Fenofibrate was also associated with a marked reduction in progression to albuminuria (p = 0.002). It was speculated that the high incidence of patients commencing statin therapy in the placebo group may have confounded the results.
The SENDCAP (St. Mary’s, Ealing, Northwick Park Diabetes Cardiovascular Disease Prevention) placebo-controlled trial of 164 patients with type 2 diabetes demonstrated no significant effect of bezafibrate on the progression of carotid and femoral artery atherosclerosis after 3 years of treatment. However, there was a reduction in the combined incidence of Minnesota-coded probable ischemic change on resting ECG and documented MI. This was associated with a reduction in total cholesterol and triglycerides and an elevation in HDL cholesterol.
Other studies have investigated subsets of diabetic populations. Highlighting the importance of HDL cholesterol levels, the VAHIT (Veterans Affairs High Density Lipoprotein Intervention Trial) was designed to determine the effect of increasing HDL cholesterol levels and lowering triglycerides on cardiovascular events in a population with CHD and low HDL cholesterol but with low-to-normal LDL cholesterol levels. Patients were treated with gemfibrozil 1200 mg/day and followed for 5 years. Gemfibrozil was more effective in diabetic patients at reducing the risk and incidence of CHD death (41%; p = 0.02) and stroke (40%; p = 0.046). There was no significant effect on these parameters in the non-diabetic group. Gemfibrozil was equally effective at reducing non-fatal MI in diabetic and non-diabetic patients.
The BIP (Bezafibrate Infarct Prevention) study was designed to investigate bezafibrate as a secondary prevention strategy in a population with previous CHD and low HDL cholesterol. Bezafibrate had no significant effect on primary endpoints and an effect on the secondary endpoints in only the patients with normal fasting insulin levels. However, it must be noted that only 11% of people in that study were diabetic patients.
Of the 4081 patients with high LDL cholesterol levels recruited for the Helsinki Heart Study, 135 had type 2 diabetes. They had lower HDL cholesterol levels, higher triglycerides, and a greater body mass index than non-diabetic subjects. Those receiving gemfibrozil had a reduced incidence of CHD (3.4%) compared with the placebo group (10.5%), a non-significant lowering of 7.1%. By contrast, treatment only reduced the incidence of CHD in non-diabetic patients by 1.2%, demonstrating the effectiveness of fibrates in a diabetic population.
Statins versus Fibrates
A comparison was made between 6 weeks’ treatment with fenofibrate or atorvastatin treatment in a crossover study of patients with type 2 diabetes and mixed hyperlipoproteinuria. Both treatments reduced total cholesterol and increased HDL cholesterol to similar levels. However, atorvastatin was more effective at lowering LDL cholesterol, larger more buoyant particles, intermediate dense particles, and small dense particles, whereas fenofibrate only significantly lowered small dense LDL particles. Fenofibrate lowered triglyceride levels and E-selectin levels; whereas only atorvastatin significantly lowered VCAM-1 levels. Neither treatment altered intercellular adhesion molecule-1 (ICAM-1) levels. The same group also investigated, in a similar cohort, the effects of these treatments on lipoprotein subfraction distribution and hemorheologic parameters. Following 6 weeks of treatment with either atorvastatin 10 mg/day or fenofibrate 200 mg/day in a crossover study, atorvastatin decreased all LDL subfractions, reducing LDL cholesterol by 29%. Fenofibrate was more effective at reducing triglycerides (−39%) and shifted the LDL balance from small dense to intermediate less dense particles. The drugs were equally effective at raising HDL cholesterol. In addition, fenofibrate altered hemorheologic parameters, decreasing fibrinogen concentration which was associated with a decrease in plasma viscosity and improved red-cell aggregation.
Finally, a study compared the effect of simvastatin 20 mg/day and fenofibrate 200 mg/day on endothelial function in type 2 diabetic patients with dyslipidemia in a crossover 3-month treatment study with a 2-month washout. Simvastatin decreased total and LDL cholesterol, and was associated with increases in ascorbic acid, plasminogen activator inhibitor-1 (PAI-1), von Willebrand factor, and vascular endothelial growth factor (VEGF) and a decrease in glutathione. Fenofibrate decreased triglycerides, and was associated with a decrease in malondialdehyde and an increase in PAI-1 and P-selectin, further demonstrating the different mechanisms by which these drugs act.
Whether a fibrate or statin should be used for the treatment of diabetic dyslipidemia primarily depends on the lipid profile of the individual patient. Statins appear more effective at lowering high LDL cholesterol levels whereas fibrates appear more effective at increasing HDL cholesterol. Thus, the individual with normal LDL cholesterol but low HDL cholesterol would be most suited to fibrate therapy as the initial drug therapy.
2.2 Renin-Angiotensin System
2.2.1 ACE Inhibitors
ACE inhibitors block the conversion of angiotensin I to angiotensin II, as well as inhibiting the breakdown of bradykinin, a potent vasodilator. This results in the inhibition of various angiotensin II-mediated effects (table I), thus reducing oxidative stress, cell adhesion, growth, and migration and increasing nitric oxide. In addition, both ACE inhibitors and ARBs have been shown to lower the formation of AGEs, which have been shown to play a role in the development of diabetic complications (section 2.4).
ACE inhibitors have been shown to be effective in reducing atherosclerosis in a number of models of atherosclerosis including Watanabe heritable hyperlipidemic rabbits, cholesterol-fed Cynomolgus monkeys, cholesterol-fed rabbits, and Pittman-Moore minipigs. ACE inhibitors also play a role in the prevention of diabetes-associated atherosclerosis. A study of diabetic apoE-/- mice demonstrated that treatment with the ACE inhibitor perindopril for 20 weeks attenuated plaque area to levels seen in non-diabetic mice. This was associated with reductions in ACE, connective tissue growth factor (CTGF), and VCAM-1 expression.
Small studies have shown the benefit of ACE inhibitors in reducing VCAM-1 levels and improving endothelial function in both type 1 and type 2 diabetic patients with either borderline hypertension or normotension.[132,133] Larger scale trials such as the HOPE (Heart Outcomes Prevention Evaluation) study have demonstrated impressive findings with ACE inhibition in a diabetic population. The HOPE study included a subgroup of 3577 people with diabetes aged ≥55 years who had ≥1 other cardiovascular risk factor or had a previous cardiovascular event. Treatment with ramipril over 4.5 years was associated with a risk reduction of 25% (p = 0.0004) for the combined primary outcome of MI, stroke, and death. This significant reduction was interpreted by the investigators to be far greater than that attributable to BP reduction. Ramipril was also associated with a 30% reduction in new-onset diabetes in a subgroup with cardiovascular risk factors.
The ALLHAT (Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial) was a study of 42 448 hypertensive individuals aged >55 years of whom ≈36% (15 297) had diabetes. The study was designed to compare the effect of a calcium channel antagonist or an ACE inhibitor to a diuretic, with the primary outcome being combined fatal CHD or non-fatal MI. Mean follow-up was 4.9 years and no significance was found between treatments on primary outcome. Similarly, in the group with diabetes, lisinopril was not advantageous over chlortalidone for most cardiovascular-disease outcomes.
Studies performed in an exclusively diabetic population include FACET (Fosinopril versus Amlodipine Cardiovascular Events Trial), a randomized trial of patients with hypertension and type 2 diabetes. Fosinopril was associated with a significant reduction in cardiovascular events (acute MI, stroke, and hospitalized angina) compared with amlodipine (14 of 189 vs 27 of 191; p = 0.03). The ABCD (Appropriate Blood Pressure Control in Diabetes) study investigated the effect of nisoldipine or enalapril on cardiovascular outcomes in 470 patients with diabetes and hypertension. Whilst both drugs had a similar effect on BP, enalapril was associated with a lower incidence of non-fatal MI compared with nisoldipine (5 vs 22; p = 0.001).
Studies that included a diabetic subgroup include CAPPP (Captopril Prevention Project), which investigated the effect of captopril compared with conventional therapy on cardiovascular morbidity and mortality in patients with hypertension. Captopril was more effective in the diabetic cohort, reducing the primary endpoint (rate of fatal and non-fatal MI, stroke, and other cardiovascular deaths), acute MI, and all cardiovascular events by 40%, 65%, and 33%, respectively. Captopril also reduced the incidence of new-onset diabetes compared with conventional treatment.
The EUROPA (EURopean trial On reduction of cardiac events with Perindopril in stable coronary Artery disease) study included 13 655 patients, 12% of whom had diabetes. Perindopril was associated with a 20% RRR (p = 0.0003) in cardiovascular death, MI, or cardiac arrest, compared with placebo. This benefit was consistent in all predefined subgroups including diabetes, which saw a 12.6% incidence of primary events in the perindopril-treated group compared with 15.5% in the placebo group.
Finally, the PROGRESS (Perindopril Protection Against Recurrent Stroke Study) trial assessed the effect of perindopril on stroke after a median of 3.9 years in 6105 patients with prior stroke or transient ischemic attack, 761 of whom had diabetes. Diabetes increased the risk of stroke by 35% and the risk reduction of stroke associated with perindopril was 38% in the diabetic group and 28% in the non-diabetic group. Although not a significant difference between the groups, perindopril was at least as effective in reducing stroke in the diabetic group as it was in the non-diabetic group.
There are also a number of retrospective analyses of the effects of ACE inhibition in diabetic cohorts. The effect of trandolapril on morbidity and mortality in diabetic patients with left ventricular dysfunction after acute MI has been reviewed in a retrospective analysis of the TRACE (Trandolapril Cardiac Evaluation) study. Trandolapril was associated with a relative risk of death from any cause of 0.64 in the diabetic group compared with 0.82 in the non-diabetic group. Furthermore, in the diabetic cohort, trandolapril significantly reduced the risk of progression to severe heart failure, with no significant effect demonstrated in the non-diabetic group.
The GISSI-3 (Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico) study investigated the effect of randomization to 6 weeks of treatment with either lisinopril or nitroglycerin, with treatment commencing within 24 hours of an acute MI. A retrospective analysis of the data demonstrated that lisinopril was associated with a decrease in mortality at 6 weeks (37 ± 12 lives saved per 1000 treated patients); this was significantly higher than in the non-diabetic population (p < 0.025). Furthermore, the survival benefit was mostly maintained at 6 months despite treatment withdrawal at 6 weeks. These studies emphasize the role of ACE inhibition in reducing cardiovascular risk, particularly in a diabetic population, as well as the added benefit in non-diabetic patients of reducing the risk of progression to overt diabetes.
In addition, there are a number of studies that have assessed the effect of ACE inhibitors specifically on atherosclerosis, as measured by either quantitative coronary angiography (QAC) or B-mode carotid ultrasound, albeit in a non-diabetic population. A subgroup analysis of 477 people with CHD enrolled in QUIET (Quinapril Ischaemic Event Trial) demonstrated no significant effect of quinapril treatment (20 mg/day) on QAC compared with placebo after 3 years. It has been suggested that this may be because of the low dose used in the study and the fact that patients were at a relatively low risk of cardiovascular events. Similarly, SCAT (Simvastatin/Enalapril Coronary Atherosclerosis Trial) found a neutral angiographic effect, as measured by QAC, of enalapril (5 mg/day) compared with placebo amongst 460 normocholesterolemic patients with coronary atherosclerosis after 4 years. The randomized, placebo-controlled trial PART2 (Prevention of Atherosclerosis with Ramipril Trial) demonstrated no significant effect, as measured by B-mode ultrasound, of ACE inhibition (5 or 10 mg/day) after 4 years in a cohort of 617 patients with a mean age of 60 years. However, a substudy of the HOPE trial, the SECURE (Study to Evaluate Carotid Ultrasound Changes in Patients Treated with Ramipril and Vitamin E) trial, involved analysis after 4.5 years of a cohort of 732 patients with vascular disease and a mean age of 65 years. Ramipril (10 mg/day) was associated with a significantly reduced progression slope of mean maximum carotid IMT compared with placebo (0.0217 vs 0.0137 mm/year; p = 0.033), whereas there was no significant effect of vitamin E. Finally, ACE inhibition with enalapril has also been shown to slow IMT progression in the common carotid arteries of patients with type 2 diabetes.
2.2.2 Angiotensin Receptor Antagonists
The effects of angiotensin II are mediated by two plasma membrane receptors, angiotensin II type 1 (AT1) and type 2 (AT2), with receptor binding leading to a cascade of downstream effects. AT2 is important for embryogenesis, although other effects mediated via this receptor are still being defined. Most of the recognized actions of angiotensin II are mediated via the AT1 receptor and conventional ARBs used for the treatment of hypertension specifically block the AT1 receptor subtype. Thus, ARBs inhibit the downstream effects of angiotensin II (table I), thereby exerting antihypertensive and cardioprotective effects. AT1-receptor expression in the vasculature has been shown to be upregulated in disease states such as hypercholesterolemia and diabetes.
ARBs have been shown to effectively reduce atherosclerosis in a variety of animal models of atherosclerosis including high-fat fed apoE-/- mice, New Zealand White rabbits, and high fat-fed monkeys. In a model of diabetic atherosclerosis, streptozocin-treated apoE-/- mice were administered either irbesartan or amlodipine to attain a similar level of BP control. Interestingly, while amlodipine had no effect on atherosclerosis, irbesartan significantly attenuated plaque deposition, highlighting the direct vascular effects of ARBs on atherosclerosis independent of their BP lowering capabilities. Irbesartan treatment was also associated with reductions in collagen content, cellular proliferation, macrophage infiltration, expression of the AT1 receptor, and expression of various mediators of atherosclerosis including platelet-derived growth factor (PDGF), MCP-1, and VCAM-1.
Cardiovascular morbidity and mortality in patients with diabetes were evaluated in the LIFE (Losartan Intervention For Endpoint reduction in hypertension) study in a cohort of patients with a mean age of 67 years, diabetes, hypertension, and signs of left-ventricular hypertrophy, as diagnosed by ECG-based criteria. BP fell to a similar level in those treated with losartan and in those treated with the β-blocker, atenolol. However, losartan was associated with a lower incidence of primary endpoints (cardiovascular death, stroke, or MI; p = 0.031), cardiovascular death (p = 0.028), and mortality from all causes (p = 0.002). Furthermore, losartan was also associated with a 25% reduction in incidence of new-onset diabetes.
The CHARM-preserved (Candesartan in Heart Failure Assessment of Mortality and Morbidity and Preserved Left-Ventricular Ejection Fraction) trial was a study of 3023 people with congestive heart failure and a left ventricular ejection fraction of >40% that investigated the addition of an ARB to current treatment. Although, ≈28% of these patients had diabetes, no subgroup analysis was conducted. While there was no significant difference in the primary outcome (cardiovascular death or unplanned admission to hospital for the management of worsening congestive heart failure) between the candesartan and placebo groups, fewer patients in the candesartan group were admitted to hospital for congestive heart failure (p = 0.017). Furthermore, those treated with candesartan were 40% less likely to develop new-onset diabetes.
The VALUE (Valsartan Antihypertensive Long-Term Use Evaluation) study treated hypertensive patients at high cardiovascular risk with either valsartan or amlodipine. This study also demonstrated that valsartan treatment significantly reduced the incidence of new-onset diabetes (p < 0.0001); this is similar to the incidence seen with an ACE inhibitor in studies such as HOPE and CAPP.
There have also been a number of studies in patients with type 2 diabetes and hypertension that have included cardiovascular outcomes as secondary endpoints. In the IDNT (Irbesartan Diabetic Nephropathy Trial), irbesartan treatment (300 mg/day), after a follow-up of 2.6 years, was associated with no effect on a secondary cardiovascular composite endpoint. However, irbesartan treatment was associated with a 23% lower incidence of hospitalization for heart failure. In the RENAAL (Reduction in Endpoints with the Angiotensin Antagonist Losartan) study of 1513 patients, losartan treatment (50–100mg once daily) was associated with no effect on a composite of morbidity and mortality from cardiovascular causes; however, losartan treatment significantly reduced hospitalization for heart failure by 32% compared with placebo (p = 0.005).
The ONTARGET (ONgoing Telmisartan Alone and in combination with Ramipril Global Endpoint Trial) is an ongoing trial investigating the effect of the ARB telmisartan and the ACE inhibitor ramipril as monotherapies or as a combination, on cardiovascular risk in 23 400 subjects, including a significant proportion with diabetes. A secondary endpoint will be new onset of diabetes.
Our understanding of the mechanisms by which the blockade of the renin-angiotensin system by either ACE inhibitors or ARBs occurs is expanding. It is becoming apparent that these agents have a wide variety of effects in addition to their powerful antihypertensive actions. Further studies will allow us to substantiate the role of renin-angiotensin system blockade in the treatment of diabetes and cardiovascular disease.
2.3 Glycemic Control
Studies investigating the effect of tight glycemic control in patients with both type 1 and type 2 diabetes have observed reductions in the incidence and severity of microvascular complications such as nephropathy and retinopathy. However, the association between glycemic control and macrovascular disease is less well understood. Fasting plasma glucose and glycosylated hemoglobin (HbA1c) levels are associated in epidemiological studies with the risk of cardiovascular disease. Thus, it has been generally assumed that intensified glycemic control would be associated with a reduction in cardiovascular risk.
The DCCT (Diabetes Control and Complications Trial) investigated whether tightly controlling diabetes with intensive therapy was superior in reducing the incidence and severity of diabetes-associated complications to conventional therapy in 1441 patients with type 1 diabetes. Patients had a mean age of ≈27 years with no retinopathy (primary prevention cohort) or mild retinopathy (secondary prevention cohort). Intensive therapy involved insulin administered either by an external pump or via ≥3 daily injections to maintain an average HbA1c of 7.2%, while conventional therapy involved one to two daily insulin injections to maintain a mean HbA1c of 9% over the study duration of 6.5 years. At the conclusion of the study, intensive treatment was associated with a 76% reduction in retinopathy compared with conventional treatment in the primary cohort. When both groups were combined, intensive treatment was associated with a 60% reduction in neuropathy, a 39% reduction in microalbuminuria, a 54% reduction in albuminuria, and a 60% reduction in clinical neuropathy. Although there was a significant reduction in hypercholesterolemia as assessed by serum LDL cholesterol levels (p = 0.02), when all major cardiovascular and peripheral vascular disease events were combined, however, intensive therapy was associated with a non-significant 41% reduction in the risk for macrovascular disease. It is thought that this lack of a statistically positive result may be due to the relatively young age of the patient cohort, which had a low prevalence of cardiovascular disease.
The EDIC study is a long-term follow-up of 1229 of 1441 patients recruited for the DCCT. The mean age of patients at a recent analysis was ≈35 years. Patients were examined for carotid IMT progression (measured by B-mode ultrasonography) as a surrogate marker of atherosclerosis. Patients were assessed 1 year and 6 years after the commencement of the EDIC study. After 1 year, there was no significant difference in IMT between age-matched controls and diabetic patients. After 6 years however, IMT was higher in the diabetic group than in the group of age-matched controls. Furthermore, common carotid IMT progression was significantly less in the intensively treated group compared with the conventionally treated group, with a difference of 0.013mm (p = 0.01) after adjustment for sex, age, ultrasonography equipment used, and the year-1 results. Combined carotid/internal artery IMT progression was also less in the intensive-treatment group compared with the conventional-treatment group. The association between mean HbA1c during the DCCT and carotid IMT at 6 years remained significant after adjusting for age, sex, ultrasonography equipment used, and the year-1 results. Importantly, during the study all patients received therapy from their own doctors with intensive therapy recommended for everyone. Mean HbA1c remained different between the groups for the first 4 years of the EDIC study but after that time point no significant difference in glycemic control could be detected. It was stated that the difference in HbA1c during the DCCT accounted for 96% of the difference in IMT observed after 6 years in EDIC. These results were followed up by a recent small study of 59 patients with type 1 diabetes and a mean age of 44 years who were randomized to either intensified conventional insulin treatment or standard insulin treatment and have been followed for >10 years. Intensive treatment was associated with significantly lower HbA1c (mean 7.1%; p <0.0001), less stiff arteries (p = 0.009), thinner intima-media in the left common carotid artery, and better endothelial function (p = 0.028).
Studies of patients with type 2 diabetes include the Veterans Affairs Diabetes Feasibility Trial of 153 men, which examined the effects of standard or intensified glycemic control. While the investigators demonstrated a difference in HbA1c of >2% (p < 0.001) between the two groups over an average of 27 months, there was no significant difference in the incidence of new cardiovascular events (p = 0.1) and no difference in total and cardiovascular mortality. However, it is likely that this was a markedly underpowered study with an inadequate duration of follow-up. The UKPDS commenced in 1977 with the intent to investigate whether intensive glycemic control reduced the incidence of microvascular or macrovascular complications in patients with recently diagnosed type 2 diabetes. Patients receiving intensive therapy achieved a mean HbA1c of 7.0% compared with 7.9% for those receiving conventional therapy over 10 years. For every 1% reduction in updated mean HbA1c there was an associated reduction in risk of 21% for any endpoint related to diabetes (p < 0.0001), 21% for death related to diabetes (p < 0.0001), 14% for MI (p < 0.0001), and 37% for microvascular complications (p < 0.0001). The risk reduction associated with improved glycemic control for microvascular endpoints and amputation or death from peripheral vascular disease was greater than that noted for MI, stroke, and heart failure. A sub-analysis of the UKPDS investigated the effect of intensive blood glucose control with either a sulphonylurea or insulin compared with conventional treatment with diet. Over 10 years, HbA1c levels were 7.0% in the intensively treated group compared with 7.9% in the conventionally treated group. While intensive therapy was associated with a 12% risk reduction for any diabetes-related endpoint (p = 0.029), the reduction in diabetes-related death and all-cause mortality was not significant. Furthermore, the reduction in risk for any diabetes-related endpoint was primarily attributable to a reduction in microvascular risk. The reduction in MI almost reached significance (p = 0.052). Another sub-analysis of the UKPDS investigated the role of intensive treatment with metformin in overweight patients with type 2 diabetes. Treatment was associated with significant risk reductions for any diabetes-related endpoint, diabetes-related death, and all-cause mortality. Thus, while intensive glycemic control may account for some reduction in cardiovascular risk, it appears that hyperglycemia per se is not wholly responsible for the macrovascular complications observed in both type 1 and type 2 diabetes. Other factors such as AGEs (section 2.4), lipids, and hemodynamic factors such as hypertension (section 2.2) may play an important role.
2.3.1 Insulin-Providing Agents
Hyperinsulinemia, a hallmark of type 2 diabetes, is associated with endothelial dysfunction, which is a recognized risk factor for coronary artery disease. Indeed, it has been suggested that the progression of hyperinsulinemia to type 2 diabetes is thought to parallel the progression of endothelial dysfunction to atherosclerosis. However, in those people unable to produce insulin, such as those with type 1 diabetes or those in the later stages of type 2 diabetes, insulin therapy is required. The UKPDS demonstrated that insulin therapy was associated with a reduction in microvascular events and an almost significant reduction in incidence of MI (p = 0.052). Furthermore, the results from the EDIC study confirm the positive effects of insulin therapy on vascular endpoints in diabetes. Insulin mediates its effects through two pathways. The first is via the phosphatidylinositol 3-kinase pathway, which mediates the effects of insulin on glucose transport in skeletal muscle as well as its effects on nitric oxide. Insulin stimulates the production of nitric oxide, an endothelium-derived relaxing factor, which in turn has been shown to reduce smooth muscle cell migration and proliferation. Nitric oxide has also been shown to reduce the expression of E-selectin, VCAM-1, and ICAM-1, involved in leukocyte recruitment and adhesion, as well as reducing the pro-inflammatory cytokines TNF-α and MCP-1. In addition, nitric oxide regulates smooth muscle proliferation, vasoconstriction, and oxidation of LDL cholesterol. The second pathway involves activation of the mitogen-activated protein kinase signaling pathway by insulin, which can lead to pro-atherogenic actions. These effects include promoting vascular smooth muscle cell growth and migration, activation of growth factors, and subsequent downstream effects that can contribute to atherosclerosis and restenosis. Thus, it remains controversial as to whether insulin per se is anti- or pro-atherogenic.
Sulphonylureas stimulate insulin secretion by closing ATP-sensitive potassium channels in pancreatic β-cells. Some of these agents specifically block the sulphonylurea receptor (SUR)-1 in β -cells, whereas other agents are non-selective and mediate effects on both SUR-1 and SUR-2 in cardiac and skeletal muscle. Gliclazide is β-cell specific and reversible, whereas glibenclamide and repaglinide mediate effects on cardiac and smooth muscle channels, in addition to β-cells, and their effects are only slowly reversible. This has led to the suggestion that these agents could induce myocardial damage, particularly in the setting of myocardial ischemia. In the UKPDS, treatment with a sulphonylurea was associated with no improvement in macrovascular disease risk but it also had no adverse effect on cardiovascular outcomes. Gliclazide, in addition to its effects on insulin, has been demonstrated to have antioxidant properties as a result of its aminoazabicyclo-octane ring, which is grafted onto the sulphonylurea. In a study of 30 patients with type 2 diabetes, gliclazide treatment, when compared with glibenclamide, was associated with a significant reduction in lipid peroxidases and a significant increases in red blood cell superoxide dismutase and platelet reactivity to collagen, thus highlighting the free-scavenging ability of this agent after just 3 months of treatment. Another short-term study in patients with type 2 diabetes demonstrated that treatment with gliclazide was associated with normalization in the level of plasma lipid peroxidases and monocyte adhesion. The investigators complemented this finding with in vitro studies, demonstrating that gliclazide reduces oxLDL and AGE-induced monocyte adhesion in endothelial cells, the latter via an inhibition of NF-κB. It remains to be determined if these effects of certain sulphonylureas ultimately translate to superior end-organ, in particular, cardiovascular protection.
2.3.2 Insulin-Sensitizing Agents
Metformin is a dimethylbiguanide compound that improves insulin sensitivity by enhancing insulin-mediated suppression of hepatic glucose production and improving insulin-stimulated glucose disposal, resulting in increased glycogen formation and glucose oxidation. Metformin also increases the uptake of glucose in adipose tissue and its subsequent oxidation as well as increasing lipogenesis and inhibiting lipolysis. Furthermore, metformin increases the binding ability of insulin to its receptor, increasing phosphorylation and tyrosine kinase activity and, thus, leading to enhanced glucose transporter-4 (GLUT-4) translocation and activity. As mentioned previously, metformin treatment is associated with a reduction in cardiovascular morbidity and mortality. In the UKPDS, metformin therapy was associated with a risk reduction of 32% for any diabetes-related endpoint (p = 0.002), 42% for diabetes-related death (p = 0.017), 36% for all-cause mortality (p = 0.011), 39% for MI (p = 0.01), and 41% for stroke (p = 0.032). Thus, metformin has been interpreted as mediating beneficial effects on the vasculature beyond its direct antihyperglycemic actions.
In a placebo-controlled study investigating the effect of 12 weeks of metformin treatment (500mg twice daily) in patients with diet-controlled type 2 diabetes, active therapy was associated with a significant improvement (p = 0.0027) in endothelium-dependent dilation, as assessed by forearm plethysmography. Consistent with this finding was a study in patients with impaired glucose tolerance that demonstrated that metformin therapy was associated with significant reductions in soluble ICAM-1, soluble VCAM-1, and von Willebrand factor levels. Reductions in CRP and PAI-1 have also been associated with metformin treatment in patients with type 2 diabetes. Metformin is also thought to mediate actions on the AGE pathway. Methylglyoxal, a precursor to AGEs and considered to mediate actions itself, is increased in the plasma of patients with type 2 diabetes and has been shown to be dose-dependently reduced by metformin therapy. It is thought that metformin reacts with intermediates such as methylglyoxal and glyoxal in a similar way to the effects of a known AGE inhibitor, pimagedine, thus preventing the formation of AGEs, thereby reducing the downstream pro-atherogenic effects of these chemical moieties (see section 2.4.2). Finally, antioxidant properties have been attributed to metformin. This agent has been shown to be a mild inhibitor of respiratory chain complex 1, as well as activating adenosine monophosphate kinase and glucose-6-phosphate dehydrogenase, and reducing hydroxy free radicals. Studies have also associated metformin treatment with an increase in activity of the free radical scavengers, CuZn superoxide dismutase, catalase, and glutathione peroxidase in the erythrocytes of patients with type 2 diabetes.
Thiazolidinediones (TZDs), specifically rosiglitazone and pioglitazone, are currently used in the clinic for the treatment of type 2 diabetes as insulin sensitizing agents. TZDs activate PPARγ, thus inducing transcription of target genes. In addition, TZDs have recently been shown to interfere with other nuclear transcription factors such as NF-κB and AP-1, and it is thought that this may be the major mechanism by which PPARγ agonists exert many of their anti-inflammatory effects. PPARγ is expressed most abundantly in adipocytes as well as in skeletal muscle. Indeed, these are the primary sites where these drugs exert their insulin sensitizing actions. TZDs act upon adiopocytes to induce differentiation; this increases the ability of these cells to take up fatty acids, thus sparing other metabolically active tissue such as skeletal muscle and the liver. Adipose tissue is now recognized as a metabolically active tissue itself; producing pro-atherosclerotic cytokines such as TNF-α, leptin, and resistin. By increasing adipocyte differentiation, TZDs reduce the production of these adipokines, thus indirectly exerting anti-atherogenic actions. TZDs also mediate effects on skeletal muscle to improve glucose uptake via increases in GLUT-4 and phosphatidyl 3-kinase. In addition, TZDs have been shown to rejuvenate pancreatic β-cells and improve their function. Thus, TZDs act to improve insulin and glycemic control and in turn may delay progression to diabetes complications through these actions.
In addition to their effects on insulin and glycemic control, more recent studies have suggested that TZDs may have direct anti-atherogenic effects independent of their metabolic effects. PPARγ has been shown to be expressed in the vessel wall in endothelial cells, smooth muscle cells, and macrophages. PPARγ agonists have been shown to directly regulate many of the processes involved in the formation of atherosclerotic plaques, such as monocyte recruitment, migration, foam-cell formation,[198,199] reverse cholesterol transport,[78,200] and plaque stability. TZDs have also been demonstrated to regulate the renin-angiotensin system, AGEs, and oxidative stress. In vivo studies have shown that troglitazone and rosiglitazone reduce atherosclerosis in LDLR-/-[205,206] and apoE-/- mice; however, these changes were confounded by concurrent changes in glucose and insulin levels. Therein lays the difficulty in determining whether the anti-atherogenic effects of TZDs are dependent on their well characterized improvements in glucose and lipid metabolism or if they are direct independent effects. In a study of streptozocin-induced diabetic apoE-/- mice, rosiglitazone reduced plaque area with no change in glucose levels, although insulin levels were not reported. Results from our own laboratory have demonstrated a reduction in atherosclerosis with rosiglitazone in streptozocin-diabetic apoE-/- mice. These vascular benefits were observed in the absence of changes in glucose or insulin levels. Moreover, these changes were observed in association with a reduction in superoxide production, macrophage infiltration, and expression of the AT1 receptor.
There are a small number of clinical studies supporting the aforementioned cellular and animal studies. A study of patients with diabetes demonstrated that 8 weeks of rosiglitazone treatment increased HDL cholesterol levels (predominantly HDL2 levels), increased LDL cholesterol levels (although this was associated with a shift from atherogenic small dense LDL cholesterol to large buoyant LDL cholesterol), decreased free fatty acids by 20%, and had no effect on triglyceride levels. A further 16 weeks’ treatment with atorvastatin, in addition to rosiglitazone, resulted in a further increase in HDL cholesterol levels (thought to be because of an increase in HDL3) and a reduction in apolipoprotein B, LDL cholesterol, and triglyceride levels. Rosiglitazone treatment in type 2 diabetic patients has also been associated with alterations in functional markers, such as a significant improvement in vasodilation in response to acetylcholine (using venous occlusion plethysmography); this effect was reversed by the nitric oxide inhibitor N-monomethyl-L-arginine-acetate. Furthermore, insulin sensitivity and blood-flow response to the PPAR agonist were correlated. Studies investigating the effect of TZDs on plasma biomarkers of coronary artery disease have also demonstrated positive results. Six months of treatment with rosiglitazone in patients with type 2 diabetes was associated with a 2-fold increase in plasma levels of the anti-inflammatory, anti-atherogenic cytokine, adiponectin; one may speculate this was because of the role of TZDs in adipocyte differentiation, as discussed previously. In another study of patients with type 2 diabetes, six months of treatment with rosiglitazone was associated with a reduction in plasma MMP-9 and CRP levels. Twelve weeks of troglitazone treatment was associated with a significant reduction in plasma levels of the soluble pro-inflammatory and pro-atherogenic ligand CD40 in diabetic patients with (34%) or without (29%) macrovascular complications, as well as in newly diagnosed patients (27%). Similar findings confirming a reduction in soluble CD40 ligand (28%) were observed in patients with CHD and type 2 diabetes. In a study of subjects with type 2 diabetes and CHD, rosiglitazone reduced MMP-2 (19%) after just 2 weeks of treatment; this effect was maintained for the full duration of the 12-week study. Concomitant changes in TNF-α and serum amyloid A were also observed. Interestingly, only those changes in serum amyloid A were correlated with glucose levels. In a group of Japanese patients with type 2 diabetes, pioglitazone significantly increased adiponectin and decreased CRP levels; pulse wave velocity was also decreased in patients in whom the drug had responded (>1% reduction in HbA1c) or not responded (<1% reduction in HbA1c). Changes in CRP levels and pulse wave velocity were independent of changes in glycemic control. In support of these studies in diabetic patients are studies that have been performed in non-diabetic patients with coronary artery disease. These studies observed reductions in functional markers such as flow-mediated dilation and carotid IMT and in surrogate markers such as CRP,[216,218] von Willebrand factor,[216,218] E-selectin, and P-selectin-positive platelets. This provides further evidence to support the additional direct effect of TZDs on vascular function.
The PROACTIVE (PROspective pioglitAzone Clinical Trial In macroVascular Events) trial, a study of >5000 people with type 2 diabetes, investigated the effect of pioglitazone, taken in addition to the patients’ current glucose-lowering drugs and other medications, on macrovascular morbidity and mortality. The study investigators demonstrated that pioglitazone non-significantly reduced the risk of the composite primary endpoint of all-cause mortality, non-fatal MI, stroke, acute coronary syndrome, leg amputation, and coronary or leg revascularization (p = 0.095). Significantly less patients reached the main secondary endpoint composite of all-cause mortality, non-fatal MI, and stroke in the pioglitazone-treated group compared with the placebo group (p = 0.027). It was concluded that pioglitazone improved cardiovascular outcomes in patients with type 2 diabetes at high cardiovascular risk and reduced the need to add insulin to glucose-lowering therapies compared with placebo. Further studies in patients with type 2 diabetes who are at a lower cardiovascular risk, or even in people without diabetes, would be of interest.
Indeed, there are a number of other large-scale clinical trials currently underway investigating the effect of rosiglitazone and pioglitazone on cardiovascular outcomes. These studies are being undertaken in patients with type 2 diabetes. In addition to measuring markers of diabetes progression, the ADOPT (A Diabetes Outcome Progression Trial) study will investigate progression of microalbuminuria, a known predictor of atherosclerosis, as well as surrogate cardiovascular markers. The BARI-2D (Bypass Angioplasty Revascularization Investigation 2 Diabetes) trial will compare the glycemic and cardiovascular benefits of peripherally acting agents, such as rosiglitazone or metformin, with centrally acting agents such as sulphonylureas or insulin. The investigators will also assess revascularization. The RECORD (Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes) study will evaluate the effect of rosiglitazone on cardiovascular disease progression, as well as comparing the combined effect of a sulphonylurea plus metformin with the combined effect of rosiglitazone plus either a sulphonylurea or metformin. The VADT (Veterans Affairs Diabetes Trial) will investigate the effect of rosiglitazone in combination with other agents on glycemic control and the prevention of cardiovascular events. The National Institute of Health sponsored ACCORD (Action to Control Cardiovascular Risk in Diabetes) study will assess the effect of rosiglitazone on HDL cholesterol, triglycerides, BP, and glycemic control. The PPAR (Pioglitazone Protects DM Patients against Re-Infarction) study will include obese hypertensive patients and will investigate whether PPARγ agonists can prevent late ischemic events after percutaneous coronary intervention and whether rosiglitazone reduces mortality, MI, and revascularization. Finally, the DREAM (Diabetes Reduction Assessment with Ramipril and Rosiglitazone Medications) trial is a study of patients with IFG or impaired glucose tolerance and will investigate the effect of ramipril and/or rosiglitazone on the incidence of diabetes. Twenty percent of patients will undergo annual carotid ultrasounds to assess the effects of the regimens on atherosclerosis.
TZDs appear to be a very promising therapy for the treatment of diabetic atherosclerosis. They address some of the metabolic disturbances observed in diabetes by acting as an insulin sensitizer as well as attenuating various other complications associated with diabetes including nephropathy and retinopathy. We must now await the results of these large clinical trials to determine the relative importance of TZDs in the treatment of diabetes-associated atherosclerosis.
2.4 Advanced Glycation End Products (AGEs)
As discussed in section 2.3, the results of the DCCT demonstrated no significant effect of glycemic control per se in modulating cardiovascular risk. However, in a recent evaluation as part of the follow-up of the EDIC study (a long-term follow-up of the same patients with type 1 diabetes from the DCCT), tight glycemic control was associated with a reduction in IMT. One possible and very plausible explanation for the observed delayed response relates to the advanced glycation pathway, now increasingly recognized to play a pivotal role in the development of diabetic complications. Indeed, a recent analysis of a subgroup of 216 patients from the DCCT demonstrated that intensive treatment was associated with reduced skin collagen glycation, glycoxidation, and crosslinking compared with conventional treatment. AGEs are formed by a browning reaction otherwise known as the non-enzymatic Maillard reaction. Briefly, the carbonyl group, an aldehyde, or ketone of the reducing sugar attaches to an amino group to reversibly form a Schiff base. This then undergoes re-arrangement to form an amadori product. Subsequent rearrangement, dehydration, and condensation leads to the irreversible formation of an AGE. These AGEs accumulate over time and their production is accelerated by high ambient glucose levels as well as oxidative stress, both of which are increased in diabetes.
Streptozocin-induced diabetes in LDLR-/- mice has been associated with an upregulation of AGEs in the vessel wall and increased circulating levels of serum AGEs. Similar evidence of vascular AGE accumulation has been reported in streptozocin-treated apoE-/- mice.[19,248] Experimental findings have been supported by clinical studies demonstrating that patients with type 2 diabetes have significantly increased serum levels of AGEs compared with non-diabetic controls; in particular, those patients with CHD had significantly higher levels than those without CHD. Similar results were found in a study by Kiuchi et al., in which serum AGE levels reflected the severity of coronary artery disease in those patients with type 2 diabetes, independent of other coronary risk factors. This study implicated a role for serum AGE levels as a marker of CHD in diabetic patients. Furthermore, in a study of human plaques obtained from carotid endarterectomy, samples from patients with type 2 diabetes demonstrated more immunoreactivity for RAGE and this correlated with plasma HbA1c levels.
The role of AGEs is further extended by studies investigating the role of dietary AGEs in the development of atherosclerosis. Lin et al. administered a high- or low- (5-fold lower) AGE diet to streptozocin-induced diabetic apoE-/- mice for 2 months. Mice receiving the low-AGE diet had a >50% reduction in plaque deposition, with a similar reduction in serum AGE levels. Reductions in tissue AGEs, RAGE, tissue factor, VCAM-1, and MCP-1 were also observed in the low-AGE fed mice compared with the high-AGE fed mice. In a similarly designed study, AGEs were fed to non-diabetic rabbits. This was associated with a significant increase in AGE levels in sera and aortic tissue. This was observed in the setting of increased intimal proliferation, lipid deposition, ICAM-1, and VCAM-1 expression, compared with untreated controls, further demonstrating a role for AGEs in atherosclerosis. Similarly, a study in diabetic patients examined the effect of a standard diet compared with a low-AGE diet. Patients in the low-AGE diet arm had significantly lower levels of glycated LDL cholesterol and oxidized LDL cholesterol. Subsequent tissue culture studies investigated the effect of LDL cholesterol from these subjects on endothelial cells. LDL cholesterol from patients receiving a standard diet demonstrated extracellular signal-regulated protein kinase 1/2 phosphorylation, whereas it was absent in those receiving a low-AGE diet; this effect was associated with increased NF-κB activation. Furthermore, a significant increase in oxidative stress and soluble VCAM-1 production was also observed in response to LDL cholesterol obtained from diabetic patients receiving the higher AGE diet.
Soluble AGE Receptor
A truncated, soluble form of RAGE (sRAGE) has recently been investigated as an approach mechanism to reduce RAGE-mediated effects. Acting as a decoy, sRAGE binds AGEs, but since it only has an extracellular domain without any signaling domains, it is unable to exert any downstream effects of RAGE. The first study to utilize the streptozocin-treated apoE-/- mouse was a study investigating the role of sRAGE. Following induction of diabetes, a subgroup was treated daily with sRAGE for 6 weeks. This was associated with a dose-dependent attenuation of plaque area to a level seen in control mice. This benefit was seen despite persistent hyperglycemia and hyperlipidemia. In addition, LDL cholesterol from sRAGE-treated diabetic mice was less susceptible to oxidation. Furthermore, 6 weeks of sRAGE treatment in this model was associated with significant attenuation of RAGE, VCAM-1, and tissue factor expression, as well as reduced NF-κB activity, similar to that observed in control mice. Further studies have been undertaken in a model of advanced atherosclerosis, where, following induction of diabetes at 6 weeks of age, diabetic mice were administered sRAGE as a late intervention at 14 weeks of age. Administration of sRAGE was associated with a >4-fold decrease in plaque area at 20 weeks compared with diabetic mice treated with non-glycated protein. Furthermore, sRAGE retarded progression of atherosclerotic plaque development at 20 weeks so that the plaque area was not significantly different to that in diabetic mice at the 14-week timepoint. This attenuation of the plaque area was associated with suppression of diabetes-induced increases in RAGE, CD68, VCAM-1, COX-2, tissue factor, phospho-p38 MAP kinase, and MMP-9 activity. These findings further demonstrate the impact of RAGE in mediating downstream effects that lead to plaque development and, in particular, the pro-inflammatory changes seen in the diabetic vessel.
The first inhibitor of AGE formation was pimagedine. This drug is thought to act as a nucleophilic trap for carbonyl intermediates during AGE formation. Pimagedine has been shown to reduce atherosclerosis in cholesterol-fed rabbits, albeit in a non-diabetic context, and these effects were correlated to tissue AGE levels. At the highest dose, this drug was effective in the aortic arch as well as the thoracic and abdominal aorta. Similar results have been observed in diabetic apoE-/- mice, in which pimagedine attenuated plaque area by 40%. Concomitant changes in plasma AGE peptides and staining of the specific AGE carboxymethyllysine and RAGE in aortic tissue were observed. Pimagedine treatment was also associated with reductions in collagens I, III, and IV, prosclerotic growth factors (CTGF and transforming growth factor-β1 protein), α-smooth muscle actin immunostaining, and skin collagen AGE levels. Whilst pimagedine appears effective as an anti-atherosclerotic agent, its ultimate use in the clinic is limited because of concerns about its long-term toxicity.
Pyridoxamine, a vitamin B6 derivative, has been shown to reduce ALEs in addition to AGEs. Pyridoxamine is thought to react with early intermediates of ALE and AGE formation; it has been suggested that it may also work to break cross-links, although this has not yet been confirmed. Pyridoxamine has been shown to reduce complications of diabetes including nephropathy, retinopathy, and neuropathy in streptozocin-induced diabetic rats. Preliminary studies in diabetic apoE-/- mice have demonstrated a role for pyridoxamine in attenuating atherosclerosis. Pyridoxamine is currently in phase II clinical trials for the treatment of diabetic kidney disease; it will be interesting to observe its effect on diabetic patients with, or at risk of, atherosclerosis.
Another therapeutic strategy to target AGEs is the use of ‘cross-link breakers’ such as N-phenacylthiazolium bromide  and alagebrium chloride (ALT-711). These agents have great clinical potential; particularly in the setting of diabetes where preformed AGEs could be cleaved, reversing to some degree the associated vascular complications. It is considered that these agents may also be chelators of transition metals and antioxidants. A role for alagebrium chloride has been demonstrated in the attenuation of diabetic cardiomyopathy and nephropathy[264,265] in streptozocin-induced diabetic rats. Furthermore, treatment of streptozocin-induced diabetic rats with alagebrium chloride for 1–3 weeks resulted in the reversal of the diabetes-induced increase in large artery stiffness (as measured by systemic arterial compliance), aortic impedance, carotid artery compliance, and distensibility, which are all surrogate markers for cardiovascular disease. In addition, in a study in diabetic apoE-/- mice, alagebrium chloride attenuated atherosclerosis by ≥30%. This treatment also reduced the diabetes-associated increases in plasma AGE peptides, aortic carboxymethyllysine, and RAGE protein. Furthermore, a concomitant reduction in total collagen as well as collagens III and IV was observed.
Kass et al. have undertaken a study investigating the role of alagebrium chloride in hypertensive patients. Alagebrium chloride treatment was associated with a significant reduction in total arterial compliance, pulse pressure, and pulse wave velocity in the absence of changes in mean BP, cardiac output, or heart rate. Importantly, this study demonstrated that in the presence of antihypertensive therapy, alagebrium chloride still conferred additional benefit on various vascular parameters. Whilst this study did not specifically include diabetic patients, one would expect that results in patients with diabetes would be at least as good as, if not better than, those observed in this study. Indeed, the observation of this study and its clinical implications must be further evaluated, particularly in the setting of diabetes, which is associated with increased vascular stiffness and hypertension.
There are a number of clinical trials investigating the role of alagebrium chloride in the setting of cardiovascular dysfunction. These include: SPECTRA (Systolic Pressure Efficacy and Safety Trial of Alagebrium), evaluating the role of alagebrium chloride in systolic hypertension; PEDESTAL (Patients with Impaired Ejection Fraction and Diastolic Dysfunction: Efficacy and Safety Trial of Alagebrium), measuring the effects of alagebrium chloride on diastolic dysfunction and ventricular mass; SAPPHIRE (Systolic and Pulse Pressure Hemodynamic Improvement by Restoring Elasticity); SILVER (Systolic Hypertension Interaction with Left Ventricular Remodeling); and DIAMOND (Distensibility Improvement and Remodeling in Diastolic Heart Failure). While none of these directly assess CHD, many of them measure surrogate markers. Furthermore, many of these studies are being carried out in cohorts of patients with complications similar to those observed in the diabetic population, such as systolic hypertension and diastolic dysfunction. In addition, there is an ongoing phase II trial which commenced early in 2004, the John Hopkins Endothelial Study, investigating the effect of alagebrium chloride on endothelial function in older hypertensive patients, as assessed by vessel relaxation and biomarkers of endothelial function.
2.5 Growth Factors
2.5.1 Platelet-Derived Growth Factor
PDGF exists in at least five isoforms and mediates its effects by two receptors, PDGF-Rα and PDGF-Rβ. Receptor binding results in dimerization and autophosphorylation, activating downstream signaling pathways and mediating effects including cell growth, proliferation, chemotaxis, and differentiation. PDGF is a cytokine recognized to play a role in embryonic development. In addition, its upregulation is associated with disease states such as tumors, lung and kidney fibrosis, restenosis, and allograft arteriosclerosis. Furthermore, PDGF has been shown to be upregulated by a variety of factors involved in the diabetic milieu including angiotensin II, endothelin, inflammatory cytokines and AGEs. PDGF has been implicated in atherosclerosis and is thought to play a role in smooth muscle cell proliferation. Furthermore, blockade of PDGF-Rβ has been associated with a reduction in vascular smooth muscle cell accumulation in fibrous cap lesions of apoE-/- mice. Treatment with the PDGF receptor antagonist imatinib has been shown to reduce lesion area in diabetic apoE-/- mice and reduce MCP-1, VCAM, and CTGF expression. Studies in humans have demonstrated increased levels of PDGF-Rα and PDGF-Rβ messenger RNA in mononuclear cells of patients with hypercholesterolemia. Whether PDGF will ultimately be an appropriate target in human atherosclerosis in the absence or presence of diabetes remains to be confirmed.
2.5.2 Connective Tissue Growth Factor
Another cytokine, CTGF, has been shown to play a role in extracellular matrix accumulation, cell hypertrophy, mitogenesis, apoptosis, cellular adhesion, chemotaxis, cell transdifferentiation, and promotion of angiogenesis. CTGF has been found to be upregulated in the diabetic kidney. In addition, CTGF expression is increased in the diabetic myocardium and around the fibrous cap of complex lesions in diabetic apoE-/- mice. Various components of the diabetic milieu including hyperglycemia, reactive oxygen species, AGEs, and pro-inflammatory cytokines have be shown to contribute toward CTGF upregulation. The mechanism by which CTGF exerts its effects is yet to elucidated, particularly as a receptor specific to CTGF has not been convincingly demonstrated. It is likely that the biological effects of CTGF relate partly to its ability to interact with other growth factors. There is growing evidence for a role for CTGF in diabetic complications, with reductions in end-organ damage, such as nephropathy and atherosclerosis, being associated with a downregulation of CTGF gene and protein expression.[20,279] Further studies investigating the effect of blocking CTGF gene or protein expression in vivo are required to determine the potential role of CTGF inhibitors in the treatment of diabetes complications such as atherosclerosis.
2.5.3 Vascular Endothelial Growth Factor
VEGF is an angiogenic factor that plays a role in embryonic vasculogenesis and modulating the adult vasculature. In addition, it can promote monocyte and smooth muscle cell migration. VEGF is secreted by endothelial cells, smooth muscle cells, platelets, and tumor cells and its expression has been shown to be upregulated in the setting of hyperglycemia. VEGF mediates its effects through receptors Flt-1 and Flk-1, which are both autophosphorylating tyrosine kinase proteins with a high affinity for VEGF. VEGF has been linked to both diabetic microvascular and macrovascular disease. In diabetic retinopathy, VEGF inhibition has been associated with a reduction in retinal neovascularization and retinal vascular permeability. VEGF has also been shown to be upregulated in diabetic nephropathy and attenuated with treatments such as ARBs. The role of VEGF in diabetic atherosclerosis is less well defined. VEGF has been shown to be present in human coronary atherosclerotic plaques. The administration of recombinant VEGF accelerates plaque area in apoE/APOB100-deficient mice and high-fat fed New Zealand White rabbits.[288,289] Plasma VEGF levels have been shown to be significantly increased in patients with coronary as well as peripheral artery disease compared with controls. VEGF levels were also increased in patients with diabetes and atherosclerosis. In contrast, a recent study demonstrated that VEGF is upregulated in diabetic patients regardless of the extent of their vascular disease state and is correlated to HbA1c. Furthermore, aggressive treatment to reduce cardiovascular risk over 1 year was associated with a reduction in plasma VEGF levels in patients with diabetes (with or without overt cardiovascular disease). Further studies are required to fully delineate the role of VEGF in diabetic macrovascular disease.
2.5.4 Insulin-Like Growth Factor
Insulin-like growth factor-1 (IGF-1) is predominantly derived from the liver. It associates with multiple binding proteins in the circulation and tissue and primarily mediates its effects via the IGF-1 receptor, IGF-1R. Various components of the IGF-1 system are regulated by factors such as oxidative stress, lipoproteins, growth factors, and hemodynamic forces. In healthy people, serum IGF-1 levels are thought to peak in early adulthood then decrease with age. In diabetic individuals, serum IGF-1 levels are thought to decrease and almost diminish with age, and worse glycemic control has been associated with lower IGF-1 levels. In addition, genetically influenced lower IGF-1 levels are associated with an increased risk of MI in people with type 2 diabetes. IGF-1 has been shown to play a role in cell growth, differentiation, migration, and apoptosis and is differentially influenced by angiotensin, TNF-α, LDL cholesterol, and reactive oxygen species. IGF-1 stimulates vascular smooth muscle cell proliferation and migration, which can contribute toward atherosclerotic plaque development. However, IGF-1 is also antiapoptotic, an effect that can contribute toward plaque stabilization. IGF-1 has also been demonstrated to influence nitric oxide biology, increasing aortic eNOS expression. In addition, IGF-1 has been shown to improve insulin sensitivity in people with type 1 diabetes, which may contribute towards its cardioprotective effects. Thus, further studies are required to fully elucidate the role of IGF-1 in the setting of atherosclerosis and indeed in vessel biology, in particular in the setting of diabetes.
There are a number of effective agents currently available for the treatment of diabetic atherosclerosis. As we uncover further mechanisms by which these agents exert their effects we will be better able to assess their potential for the treatment of diabetic macrovascular complications. It appears likely that combination regimens will be the most effective method to combat diabetic atherosclerosis and its associated complications. It remains very important for future large-scale clinical trials to continue to further examine the cardiovascular benefits obtained from these new regimens, involving not only currently available agents but also drugs that are in preclinical and clinical development.
Dr Calkin is a recipient of a postgraduate scholarship from the National Heart Foundation of Australia. Dr Allen is a recipient of an RD Wright Fellowship Career Development Award from the National Health & Medical Research Council of Australia.
No sources of funding were used to assist in the preparation of this review. The authors have no conflicts of interest that are directly relevant to the content of this review.
- 41.Uusitupa MI, Niskanen LK, Siitonen O, et al. 5-year incidence of atherosclerotic vascular disease in relation to general risk factors, insulin level, and abnormalities in lipoprotein composition in non-insulin-dependent diabetic and nondiabetic subjects. Circulation 1990; 82: 27–36.PubMedCrossRefGoogle Scholar
- 58.Vincent L, Soria C, Mirshahi F, et al. Cerivastatin, an inhibitor of 3-hydrox-y-3-methylglutaryl coenzyme A reductase, inhibits endothelial cell proliferation induced by angiogenic factors in vitro and angiogenesis in in vivo models. Arterioscler Thromb Vasc Biol 2002; 22: 623–9.PubMedCrossRefGoogle Scholar
- 61.Kleemann R, Princen HM, Emeis JJ, et al. Rosuvastatin reduces atherosclerosis development beyond and independent of its plasma cholesterol-lowering effect in APOE*3-Leiden transgenic mice: evidence for antiinflammatory effects of rosuvastatin. Circulation 2003; 108: 1368–74.PubMedCrossRefGoogle Scholar
- 66.Inoue I, Goto S, Mizotani K, et al. Lipophilic HMG-CoA reductase inhibitor has an anti-inflammatory effect: reduction of MRNA levels for interleukin-1beta, interleukin-6, cyclooxygenase-2, and p22phox by regulation of peroxisome proliferator-activated receptor alpha (PPARalpha) in primary endothelial cells. Life Sci 2000; 67: 863–76.PubMedCrossRefGoogle Scholar
- 71.Sever PS, Dahlof B, Poulter NR, et al. Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-thanaverage cholesterol concentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial-Lipid Lowering Arm (ASCOT-LLA): a multicentre randomised controlled trial. Lancet 2003; 361: 1149–58.PubMedCrossRefGoogle Scholar
- 74.Goldberg RB, Mellies MJ, Sacks FM, et al. Cardiovascular events and their reduction with pravastatin in diabetic and glucose-intolerant myocardial infarction survivors with average cholesterol levels: subgroup analyses in the Cholesterol And Recurrent Events (CARE) trial. The Care Investigators. Circulation 1998; 98: 2513–9.Google Scholar
- 90.Inoue I, Goto S, Matsunaga T, et al. The ligands/activators for peroxisome proliferator-activated receptor alpha (PPARalpha) and PPARgamma increase Cu2+, Zn2+-superoxide dismutase and decrease p22phox message expressions in primary endothelial cells. Metabolism 2001; 50: 3–11.PubMedCrossRefGoogle Scholar
- 92.Cooper ME, Forbes JM, Smith CM, et al. Direct anti-atherogenic properties of PPAR alpha and PPAR gamma agonists. 41st Meeting of the European Association for the Study of Diabetes; 2005 Sep 10–15; Athens. Diabetologica 2005; 18 (S1): A1117.Google Scholar
- 96.Effect of fenofibrate on progression of coronary-artery disease in type 2 diabetes: the Diabetes Atherosclerosis Intervention Study, a randomised study. Lancet 2001; 357: 905–910.Google Scholar
- 99.Rubins HB, Robins SJ, Collins D, et al. Gemfibrozil for the secondary prevention of coronary heart disease in men with low levels of high-density lipoprotein cholesterol. Veterans Affairs High-Density Lipoprotein Cholesterol Intervention Trial Study Group. N Engl J Med 1999; 341: 410–8.PubMedCrossRefGoogle Scholar
- 103.Empen K, Frost RJ, Geiss HC, et al. Differential effects of fenofibrate versus atorvastatin on the concentrations of E-selectin and vascular cellular adhesion molecule-1 in patients with type 2 diabetes mellitus and mixed hyperlipoproteinemia: a randomized cross-over trial. Cardiovasc Diabetol 2003; 2: 17.PubMedCrossRefGoogle Scholar
- 108.Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group. BMJ 1998; 317: 703–713.Google Scholar
- 118.Nishimura H, Tsuji H, Masuda H, et al. Angiotensin II increases plasminogen activator inhibitor-1 and tissue factor mRNA expression without changing that of tissue type plasminogen activator or tissue factor pathway inhibitor in cultured rat aortic endothelial cells. Thromb Haemost 1997; 77: 1189–95.PubMedGoogle Scholar
- 134.Effects of ramipril on cardiovascular and microvascular outcomes in people with diabetes mellitus: results of the HOPE study and MICRO-HOPE substudy. Heart Outcomes Prevention Evaluation Study Investigators. Lancet 2000; 355: 253–259.Google Scholar
- 135.Major outcomes in high-risk hypertensive patients randomized to angiotensinconverting enzyme inhibitor or calcium channel blocker vs diuretic: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA 2002; 288: 2981–2997.Google Scholar
- 142.Gustafsson I, Torp-Pedersen C, Kober L, et al. Effect of the angiotensin-converting enzyme inhibitor trandolapril on mortality and morbidity in diabetic patients with left ventricular dysfunction after acute myocardial infarction. Trace Study Group. J Am Coll Cardiol 1999; 34: 83–9.CrossRefGoogle Scholar
- 146.MacMahon S, Sharpe N, Gamble G, et al. Randomized, placebo-controlled trial of the angiotensin-converting enzyme inhibitor, ramipril, in patients with coronary or other occlusive arterial disease. PART-2 Collaborative Research Group. Prevention of Atherosclerosis with Ramipril. J Am Coll Cardiol 2000; 36: 438–43.PubMedCrossRefGoogle Scholar
- 149.Nickenig G, Jung O, Strehlow K, et al. Hypercholesterolemia is associated with enhanced angiotensin ATI-receptor expression. Am J Physiol 1997; 272: H2701–7.Google Scholar
- 159.The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med 1993; 329: 977–986.Google Scholar
- 160.Ohkubo Y, Kishikawa H, Araki E, et al. Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective 6-year study. Diabetes Res Clin Pract 1995; 28: 103–17.PubMedCrossRefGoogle Scholar
- 164.Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998; 352: 854–865.Google Scholar
- 167.Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998; 352: 837–853.Google Scholar
- 182.Caballero AE, Delgado A, Aguilar-Salinas CA, et al. The differential effects of metformin on markers of endothelial activation and inflammation in subjects with impaired glucose tolerance: a placebo-controlled, randomized clinical trial. J Clin Endocrinol Metab 2004; 89: 3943–8.PubMedCrossRefGoogle Scholar
- 218.Sidhu JS, Cowan D, Kaski JC. The effects of rosiglitazone, a peroxisome proliferator-activated receptor-gamma agonist, on markers of endothelial cell activation, C-reactive protein, and fibrinogen levels in non-diabetic coronary artery disease patients. J Am Coll Cardiol 2003; 42: 1757–63.PubMedCrossRefGoogle Scholar
- 220.Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROACTIVE Study (PROspective pioglitAzone Clinical Trial In macro Vascular Events): a randomised controlled trial. Lancet 2005 Oct 8; 366(9493): 1279–89.PubMedCrossRefGoogle Scholar
- 223.Sobel BE, Frye R, Dette KM, et al. Burgeoning dilemmas in the management of diabetes and cardiovascular disease: rationale for the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) Trial [published erratum appears in Circulation 2003; 108 (4): 500]. Circulation 2003 Feb 4; 107(4): 636–42.PubMedCrossRefGoogle Scholar
- 226.ACCORD purpose: protocol abstract November 14, 2002 [online]. Available from URL: http://www.accordtrial.org/public/purpose.cfm [Accessed 2005 Dec 6].
- 227.Proglitazone protects DM patients against re-infarction (PPAR study) [online]. Available from URL: http://www.clinicaltrials.gov/ct/gui/show/NCT00212004 [Accessed 2005 Dec 6].
- 229.Monnier VM, Bautista O, Kenny D, et al. Skin collagen glycation, glycoxidation, and crosslinking are lower in subjects with long-term intensive versus conventional therapy of type 1 diabetes: relevance of glycated collagen products versus HbAlc as markers of diabetic complications. DCCT Skin Collagen Ancillary Study Group. Diabetes Control and Complications Trial. Diabetes 1999; 48: 870–80.Google Scholar
- 236.Schmidt AM, Hori O, Chen JX, et al. Advanced glycation endproducts interacting with their endothelial receptor induce expression of vascular cell adhesion molecule-1 (VCAM-1) in cultured human endothelial cells and in mice: a potential mechanism for the accelerated vasculopathy of diabetes. J Clin Invest 1995; 96: 1395–403.PubMedCrossRefGoogle Scholar
- 260.Jandeleit-Dahm K, Cooper M, Allen T. Advanced glycation inhibitors reduce atherosclerosis in diabetic ApoE-/- mice. 8th International Conference on the Maillard Reaction; 2004 Aug 28-Sep 1; Charleston (SC).Google Scholar
- 268.Alteon Inc. A.G.E crosslink breakers [online]. Available from URL: http://www.alteon.com/crossl.htm [Accessed 2005 May 1].
- 282.Natarajan R, Bai W, Lanting L, et al. Effects of high glucose on vascular endothelial growth factor expression in vascular smooth muscle cells. Am J Physiol 1997; 273: H2224–2231.Google Scholar