13.6.1 Mechanisms of macrovascular disease in diabetes

The virtual epidemic of diabetes that has appeared over the last couple of decades has highlighted the influence of Western lifestyles and obesity on the development of glucose intolerance and associated cardiovascular disease. Two important hypotheses need consideration in contemplating the strong clinical links that exist between diabetes and cardiovascular disease.

One of the great advances in understanding in the past 20 years has been the observation that insulin resistance is associated with inflammatory and atherothrombotic risk factor clustering to provide a risk ‘mirror’ for the changes observed in the vulnerable atheromatous plaque. This brings together the thrifty and the common soil hypotheses and indicates that physiological fluctuations in weight and insulin resistance seen in relation to variation in food availability become pathological with chronic exposure leading to both type 2 diabetes and cardiovascular disease. As insulin resistance cycles to type 2 diabetes, hyperglycaemia has further detrimental effects on vascular disease through the generation of reactive oxygen species, glycation of longlasting proteins, and direct effects of glucose. Epidemiological studies demonstrate a marked increase in vascular outcomes as individuals move from euglycaemic insulin resistance to type 2 diabetes to reflect this increased risk. Finally, the development of microvascular renal disease amplifies vascular risk further and the combination of hyperglycaemia and renal disease provides a common pathway for increased cardiovascular risk in both type 1 and type 2 diabetes.

Ischaemic macrovascular disease in subjects with diabetes tends to be widespread and diffuse rather than localized and in the coronary arteries is associated with both proximal and distal disease. This combination of disease characteristics is at least partially responsible for the generally poor vascular outcomes noted in patients with diabetes, with higher mortality rates post acute coronary syndromes (ACS), increased restenosis after intervention and poorer outcomes in relation to heart failure. The development of arterial plaques is a complex process involving early endothelial cell damage and dysfunction, foam cell and fatty streak formation, and plaque vulnerability and rupture, ultimately leading to thrombus formation and vascular occlusion. Prominent players in this progression include (the endothelium) endothelial cells, macrophages, platelets, and the fluid phases of both coagulation and inflammation. Evidence now indicates that all these components are affected adversely by insulin resistance, hyperglycaemia, and reactive oxygen species to provide biochemical support for the strong association between diabetes and cardiovascular disease. As one can reasonably consider the stable plaque as relatively benign, much attention has focused on plaque characteristics that predispose to lack of stability and plaque rupture, and particularly, the effect of diabetes on these processes. In general, plaque rupture occurs in plaques with relatively larger lipid content, increased macrophage infiltration and decreased smooth muscle cells. Studies in arterial lesions from subjects with diabetes indicate that these lesions have a bigger lipid core and increased macrophage content and intraplaque-thrombus compared with the nondiabetic population. Furthermore, patients with diabetes who have unstable angina are reported to have relatively higher thrombus content in the affected coronary artery than controls to emphasize the importance of coagulation processes in the final phenotype of ACS. Studies of advanced glycation end products (AGEs) in plaques from patients with diabetes have not supported an equivalent role for AGEs and, by extension, glycaemic control in this process. In summary, histological studies of coronary artery disease indicate that this is an inflammatory atherothrombotic disease, and that all aspects of this process are accentuated in the presence of diabetes. Understanding of the molecular basis of these processes will clarify the pathological basis of disease and support the development of rational therapeutic approaches.

One of the most interesting developments in recent years has been the change in understanding of the fat cell and its potential role in the development of both diabetes and cardiovascular disease. Initially thought of as a simple storage cell for triglyceride and nonesterified fatty acids, in recent years awareness has grown that the adipocyte is an endocrine organ with the capacity to influence insulin resistance, appetite, inflammatory responses, blood flow, and thrombosis. The adipocyte has the capacity to mount inflammatory responses as a reaction to infection and it appears that in the presence of obesity, these mechanisms become dysregulated and contribute to the development of low-grade systemic inflammation. Under these circumstances the ‘fat-filled’ adipocyte secretes increased amounts of interleukin (IL)-6, complement C3, tumour necrosis factor α (TNFα), the fibrinolytic inhibitor, plasminogen activator inhibitor-1 (PAI-1), and components of the renin–angiotensin system. In addition, levels of adiponectin, an adipocyte specific chemokine which increases insulin sensitivity, are suppressed and leptin produced by the adipocyte is increased. Some evidence indicates that these events are mediated by interactions between the adipocyte and resident macrophages, mediated by free fatty acid activation of the monocyte/macrophage. The development of obesity has been associated with a marked increase in resident macrophages and the adipocyte/macrophage interactions probably account for the change in phenotypic expression of the fat cell mass in obese subjects. The importance of these observations in relation to the development of insulin resistance and type 2 diabetes is that all of the changes in protein expression could be seen as contributing to the development of systemic insulin resistance, and thereby provide a potential mechanism by which chronic obesity could lead to the development of diabetes and the associated cardiovascular risk cluster which occurs with insulin resistance. A more recent development has been the recognition of the potential role of ectopic fat in all of these processes. The development of obesity is associated with increased visceral fat deposition around organs and tissues, particularly the liver, and also in the heart itself and around blood vessels. Fat deposition can alter organ function, not only by its physical presence but also through the effects of adipokines on cells. These observations provide a mechanism by which fat tissue can affect function locally, and contribute to the development of cardiovascular disease, while simultaneously promoting insulin resistance and type 2 diabetes.

Pathological studies have demonstrated a defined series of changes in the vessel wall during atherogenesis. These changes involve multiple cell types including: macrophages, vascular smooth muscle cells, fibroblasts, platelets, and endothelial cells. The endothelium is uniquely placed to act as a key regulator of vascular homeostasis and there is now compelling evidence that very early in the atherogenic process before the onset of morphological changes a subtle change occurs in endothelial cell phenotype. This change is characterized by a shift to an unfavourable imbalance between the release of antiatherosclerotic and proatherosclerotic signalling molecules by the endothelium. This change in phenotype is now commonly termed as endothelial dysfunction. Arguably the most crucial feature of endothelial dysfunction and certainly the most extensively studied is a decline in the bioavailability of the antiatherosclerotic signalling molecule nitric oxide. Nitric oxide is generated by a family of nitric oxide synthases (NOSs) from l-arginine in a reaction that requires oxygen, NADPH and the essential cofactors tetrahydrobiopterin (BH4), FAD, and FMN. Nitric oxide production and bioavailability are regulated/dysregulated at transcriptional and post-transcriptional levels. The key determinant of nitric oxide bioavailability is probably the balance between nitric oxide and reactive oxygen species. Multiple characteristics of the type 2 diabetes phenotype have been shown to reduce nitric oxide bioavailability including: hyperglycaemia, hypertension, dyslipidaemia, increased free fatty acids, systemic inflammation, and reduced adiponectin. Numerous studies have now established reduced nitric oxide bioavailability as a hallmark of type 2 diabetes.

The bioavailability of nitric oxide represents a key marker in vascular health. Reduced nitric oxide bioavailability potentially impacts on a number of different cellular components of the atherothrombotic plaque and the different stages of atherogenesis.

Consistent with nitric oxide being a key antiatherosclerotic molecule, longitudinal studies have shown that impaired nitric oxide-dependent vasodilatation can predict future cardiac events and the development of coronary artery atherosclerosis.

The earliest morphological features of atheroma formation consist of the development of fatty streaks in association with macrophage infiltration, extracellular lipid deposition, and the formation of lipid rich foam cells. Progression of this process in later life is characterized by the development of larger lipid pools, fibrous connective tissue, and thrombus formation. Macrophage migration plays a critical role in this process. Proliferation of macrophages within the early lesion sets up a cycle of proinflammatory processes that include increased secretion of IL-1β and TNFα, increased proinflammatory gene expression and secretion of growth factors which amplify inflammatory responses and enhance smooth muscle cell proliferation and atheroma formation. A number of mechanisms relevant to diabetes and insulin resistance have been implicated in increased macrophage activation and foam cell formation. Evidence indicates that PPARγ plays a critical role in macrophage biology by increasing ox-low-density lipoprotein (oxLDL) degradation, suppression of proinflammatory responses, including metalloproteinases and increasing macrophage CD36 expression with reduced expression of macrophage scavenger receptor, class A (SR-A). Overall these observations support the view that low PPARγ activity promotes foam cell and early atheroma formation, processes that are reversed by therapeutic PPARγ activators. Additionally, there is evidence to indicate that glucose itself acts to further enhance macrophage activation in the presence of macrophage activating factors, and that AGEs of LDL enhance macrophage activation through interactions with the macrophage toll 4 receptor. Macrophages have a functioning surface insulin receptor which, when activated by insulin, sets in chain the classic pathway of insulin signalling seen in cell types classically associated with glucose metabolism. Insulin, in an insulin sensitive setting, has a variety of effects on the macrophage to reduce reactive oxygen species production, enhance glucose metabolism and reduce macrophage apoptosis. Conversely, the development of macrophage insulin resistance is broadly associated with increased CD36 and SR-A expression, increased uptake of atherogenic lipoproteins, reduced phagocytosis, formation of atheromatous foam cells and fatty streaks and the development of complex plaques that morphologically would be associated with advanced unstable plaques prone to rupture and thrombus formation.

Early atheroma formation is characterized by increased numbers of smooth muscle cells and the development of an intra-intimal necrotic lipid core associated with varying degrees of inflammatory changes that include monocyte/macrophage infiltration, and foam cell formation. Stable plaques tend to have a high fibrous collagen content with less lipid core and more inflammatory changes. One of the major areas of interest in the prevention and management of ACS has been identification of the factors that convert the stable plaque to a vulnerable and ultimately unstable phenotype. Characteristics that predispose to a vulnerable plaque include an increased volume of necrotic core, enhanced inflammatory changes, thinning of the fibrous cap, and thrombosis. It appears that two distinct but related mechanisms predispose to ACS: plaque rupture with thrombus formation accounting for the majority of sudden coronary deaths, and thrombus on a plaque erosion accounting for the remainder. There is evidence to indicate that erosion occurs with a higher frequency in diabetic subjects. In subjects with diabetes, a whole array of metabolic changes are seen which increase the risk of conversion of a stable plaque into a vulnerable plaque prone to rupture. These include increased expression of adhesion molecules and chemokines which attract and bind the monocyte/macrophage, and enhance monocyte/macrophage migration. Other inflammatory cells such as T cells and mast cells also appear to have a role, although the effects of diabetes are less well established compared to the monocyte/macrophage.

In insulin-resistant states, decreased macrophage insulin signalling seems to have an important role in the development of a vulnerable plaque as outlined above. The final steps which lead to plaque rupture remain to be determined. It is widely held that matrix degradation plays an important role through the release of matrix-degrading metalloproteinases from resident macrophages. Evidence indicates that mast cells and inflammatory cytokines can promote this process. An alternative mechanism could be envisaged by a reduction in the generation of atheromatous matrix as vulnerable plaques have diminished collagen and other matrix components. The effects of insulin resistance and diabetes on some of these latter mechanisms remain to be established, although the matrix-degrading enzymes MMP1 and MMP9 are increased by hyperglycaemia and MMP9 and tissue inhibitors of MMPs in plasma were higher in diabetic than nondiabetic subjects with ACS. Although these observations are not unequivocal proof for a role of MMPs in the increased risk of ACS in diabetes, they do suggest a potential mechanism that warrants further investigation.

Rupture of an atheromatous plaque is accompanied by the release of procoagulant material, platelet activation and activation of the fluid phase of coagulation. In most circumstances this process resolves without clinical sequelae, equally however, it can lead to the formation of an occlusive platelet-rich fibrin mesh with tissue damage and death. As mentioned, atheromatous plaques from subjects with diabetes have increased intraplaque thrombosis, whereas diabetes patients with unstable angina have increased intracoronary thrombus. An enormous volume of evidence has documented changes in platelet function and in the fluid phase of coagulation in diabetes. However, two further groups of studies further emphasize the importance of platelet/coagulation mechanisms in relation to diabetes. Studies show thrombus formation is enhanced in blood from diabetes patients and particularly so in relation to fluctuations in glycaemic control. In addition, clinical studies of platelet inhibition using either IIb/IIIa inhibitors or thiopyridines in the setting of ACS show particular benefit in diabetes patients.

The principal mechanisms that maintain the platelet in a quiescent, nonthrombotic phase are related to endothelial–platelet interactions and local platelet regulatory mechanisms. As described earlier in more detail, hyperglycaemia and endothelial insulin resistance both interfere with endothelial nitric oxide production, the former through generation of reactive oxygen species, which additionally activate NF-κB and increases TNFα and IL-6 production, proteins involved in both the development of insulin resistance and atheromatous vascular disease. Decreased nitric oxide generation creates a vasoconstrictive vascular phenotype and makes the platelet more susceptible to activation. Association studies indicate that poor glycaemic control enhances platelet activation directly. It has been proposed that diminished lipid peroxidation may account for this finding. Other metabolic abnormalities associated with insulin resistance and type 2 diabetes, including raised LDL and triglyceride, generation of reactive oxygen species, and a low-grade inflammatory response have been shown to have direct effects on platelet activation. There is additionally evidence that increased leptin and suppression of adiponectin levels enhance platelet activation. Activated platelets release platelet microparticles, which bind to surface antigens and which have potent proinflammatory properties. Raised levels of circulating microparticles have been reported in type 2 diabetic patients in relation to the presence of vascular complications. The activated platelet also expresses CD40L, IL-1β, and other inflammatory mediators, to provide further links between inflammation and thrombosis.

Insulin resistance is associated with alterations in components of the fluid phase of the coagulation and fibrinolytic pathways, the most marked and consistent of which is elevated levels of the fibrinolytic inhibitor, PAI-1. Evidence indicates that the fat-filled adipocyte secretes PAI-1, and that this is mediated by TNFα. Levels of PAI-1 are elevated in the nondiabetic, insulin-resistant, first-degree relatives of type 2 diabetes patients and lowered by strategies (weight loss, metformin, thiazolidinediones) that ameliorate insulin resistance in humans. Biochemically, PAI-1 binds to tissue plasminogen activator (tPA) in the circulation to inhibit tPA induced plasmin generation from plasminogen. This interaction is predominantly in place to prevent the generation of circulating plasmin which would increase fibrinogen degradation and increase bleeding risk as occurs in intravascular coagulation. Although there are mechanisms in place to prevent PAI-1/tPA interactions on the clot surface, in practice, low levels of PAI have been associated with increased bleeding risk and high levels of PAI-1 have been related to the development of venous thrombosis and recurrent myocardial infarction. In addition to these changes in PAI-1, increased levels of coagulation factors such as factors VII and XII and fibrinogen are also seen in insulin-resistant diabetic populations and, although formal clinical proof is not forthcoming, biochemically, this combination of abnormalities would be expected to promote clot formation and inhibit clot lysis, thereby promoting more extensive occlusive disease.

The development of hyperglycaemia further complicates thrombotic risk, partly through direct effects of glucose and partly through post-translational modifications to coagulation proteins. Dominant among these is the effect of glycation on fibrinogen, which leads to the formation of a fibrin clot that has a denser structure with thinner fibres—a structure associated with increased cardiovascular risk. Molecular studies indicate that this structure is associated with increased binding of plasmin inhibitor, decreased plasmin generation, and slower clot lysis in comparison to fibrin(ogen) from nondiabetic subjects.

The development of ischaemic arterial disease is a complex process involving multiple cell types including, the vascular endothelium, smooth muscle cells, macrophages and platelets. Additionally, the fluid phases of the thrombotic and inflammatory processes have an important role in promoting atheroma formation, plaque rupture, and occlusive thrombus. The complexity of cardiovascular disease is mirrored by the metabolic complexity associated with the development of diabetes itself, where varying combinations of insulin resistance, hyperglycaemia, protein glycation, and reactive oxygen species production stimulate the cellular and biochemical processes that lead to arterial disease. In this manner, the biology of diabetes and cardiovascular disease brings these disorders together as one condition to confirm the epidemiological and trial data sets which themselves support the concept of the common soil hypothesis. While individuals with type 1 diabetes may reach this destination through long standing hyperglycaemia, associated renal microvascular disease and reactive oxygen species production with a lesser impact of insulin resistance, type 2 diabetes has the additional burden of longstanding insulin resistance predating the development of B-cell function with an accompanying phenotypic shift in the biochemical profile stimulating further vascular risk.

In the past 20 years a portfolio of clinical trials in type 2 diabetes has provided us with the opportunity to successfully modify vascular risk in our patients. Unequivocally, the use of angiotensin-converting enzyme inhibitors and statins has improved vascular outcomes and driven mortality levels in relation to cardiovascular disease to new low levels. Equally, where the use of these agents has provided clarity, the management of hyperglycaemia has progressively created more doubt. From the cessation of UGDP, due to concerns that sulphonylureas were associated with increased vascular risk, through to the reportedly deleterious effects of metformin/sulphonylureas combination in the United Kingdom Prospective Diabetes Study and the recent concerns about the use of rosiglitazone, it has been difficult to find an oral antidiabetic agent associated with consistent cardiovascular benefits. Additionally, data presented recently from two studies of the effects of improving glycaemic control on cardiovascular outcomes reported tight control associated with worse outcomes (ACCORD) or no effect (ADVANCE). However, in type 1 diabetes, the relationship between improved glycaemic control and cardiovascular outcomes is a little stronger with evidence from the Diabetes Control and Complications Trial of a legacy effect of good control leading to improved long-term cardiovascular outcomes. In type 2 diabetes, there exists layers of complexity which seem to contribute to cardiovascular risk over and above the effects of glycaemic control that generate a plethora of metabolic abnormalities. These include alterations in the cellular and fluid phases of inflammatory atherothrombotic pathways that precisely mirror alterations in the vessel wall and ultimately promote the sequence of atheroma formation through to plaque rupture and occlusive thrombus formation. A further understanding of the molecular mechanisms underpinning these processes will provide opportunities to develop new approaches to the management of these complex conditions.