The worldwide prevalence of diabetes in 2010 is already at epidemic proportions with an estimated 284.6 million adults (aged 20–79 years), representing 6.4% of the global adult population.1 However, more worryingly, the prevalence has increased by 15% since 20072 and projections estimate that 438.4 million adults will have diabetes in 2030, a 54% increase on 2010 figures.1 This drastic rise in the prevalence of diabetes is mainly due to lifestyle-related issues (including diet, lack of exercise, and obesity) and it is recognised as becoming one of the leading causes of mortality and morbidity.
Common diabetes-associated complications include cardiovascular disease (CVD), stroke, neuropathy, renal impairment, retinopathy and blindness. More than one third of people born in the US in 2000 will develop diabetes2 and type 2 diabetes is increasingly being developed at a younger age.2 The increased incidence of diabetes and its associated complications, coupled with an earlier onset of the disease, will greatly increase incumbent healthcare costs. Hence, diabetes will undoubtedly become a global health catastrophe in the 21st century.
Diabetes and Cardiovascular Disease
Diabetes per se leads to a greatly increased risk of CVD and mortality rates are much higher in people with diabetes than those without this disease. Four-fifths of patients with diabetes die of cardiovascular complications4 and life expectancy is reduced by up to 10 years compared with that in persons without diabetes.5 The risk of a future myocardial infarction (MI) in diabetic patients with no history of coronary artery disease (CAD) is similar to that of non-diabetic subjects with known CAD6–8 and it has been calculated that developing diabetes confers a cardiovascular risk equivalent to ageing 15 years.9 Hence, treatment guidelines now recommend that diabetes should be handled as a coronary disease equivalent.3,10 CVD is the leading cause of mortality and one of the leading causes of disability worldwide11 and, like diabetes, there are contributing lifestyle-associated factors that are currently more prevalent in developed nations. In the US alone, more than 80 million adults have at least one type of CVD.12
A study utilising death certificate data for adults in North Dakota for the period 1992–6 identified CVD as the underlying cause of death in almost half of all deaths and diabetes in 15% of deaths.13 Furthermore, the age-adjusted relative risk (RR) of death was 2.6-fold higher in subjects with diabetes compared with non-diabetic subjects. A prospective study in Finland identified newly diagnosed patients with diabetes aged 45–64 years (n=133) and compared 15-year mortality with non-diabetic subjects within the same age group (n=144).14 Compared with their respective control groups, total mortality was approximately five-fold higher in subjects with diabetes in both male and female groups (age-adjusted odds ratio (OR) 5.0 and 5.2, respectively; p<0.001 in both cases). Corresponding increases in cardiovascular mortality for diabetic subjects were six-fold higher in males and 11-fold higher in female subjects (age-adjusted OR 6.2 and 11.2, respectively; p<0.001 in both cases). Additionally, hyperglycaemia or lipid abnormalities characteristic of diabetes at baseline, after five years or after 10 years, were predictive of cardiovascular mortality.
Glucose Perturbations in Cardiovascular Disease
The effect of glucose perturbations in patients with coronary artery disease (CAD) and without known diabetes has been assessed in studies of populations in Sweden (Glucose Abnormalities in Patients with Myocardial Infarction [GAMI] study, n=164),15 in China (China Heart Survey, n=2,263)16 and across 25 European countries (The Euro Heart Survey, n=1,920).17 Based on oral glucose tolerance test (OGTT) results, 35–37% of patients in these three studies were classified as having pre-diabetes and 18–31% as having type 2 diabetes mellitus (T2DM) (see Figure 1). The fact that only 34–45% of patients were classified as normoglycaemic is indicative of the high level of undiagnosed dysglycaemia among CAD patients in relatively developed countries.
These results also highlight the importance of using the OGTT as the method of diagnosing diabetes and, especially, pre-diabetes in patients with CAD. When Euro Heart Survey patients’ (with available data, n=1,867) glucose regulation status was classified according to the World Health Organization (WHO) criteria18 based on fasting plasma glucose (FPG) levels alone (as per the American Diabetes Association [ADA] criteria of 1997),19 81% were classified as having normal glucose levels (<6.1mmol/l), 15% as having impaired fasting glucose (IFG; 6.1–6.9mmol/l) and less than 5% as having diabetes (>7.0mmol/l). When the same subjects were assessed using WHO criteria for OGTT results only 47% were classified as having normal glucose regulation (NGR), 37% as having impaired glucose regulation (IGR: IFG or impaired glucose tolerance [IGT]) and 17% as having diabetes. Thus, 44% of patients identified as having normoglycaemia based on FPG alone would be reclassified as having either IGR or diabetes if OGTT results were used as the basis for classification (see Figure 2).
The importance of the OGTT for the early identification of dysglycaemia in patients without diagnosed diabetes is therefore an important aspect of the proper management of cardiovascular risk. The Diabetes Epidemiology: Collaborative Analysis of Diagnostic Criteria in Europe (DECODE) study assessed the risk of cardiovascular mortality in European subjects (aged ≥30 years, n=25,364) based on both FPG and OGTT results.20 Compared with patients who were classified as normoglycaemic according to ADA criteria for FPG,19 relative risk was increased by 21% in men (hazard ratio [HR]=1.21; 95% confidence interval [CI]: 1.05–1.41) with IGT and by 8% in women (HR=1.08; 95% CI: 0.70–1.66); the corresponding increase in risk in subjects with newly diagnosed diabetes was approximately 80% in both male and female subjects (HR=1.81; 95% CI: 1.49–2.20 and HR=1.79; 95% CI: 1.18– 2.69, respectively).
Using the WHO criteria to classify patients based on the OGTT results, relative risk was increased by 51% (HR=1.51; 1.32–1.72) in males with IGT (7.8–11.1mmol/l) and by 60% in women (HR=1.60; 1.22–2.10); the relative risk increased more than twofold in male (HR=2.02; 1.66–2.46) and 2.8-fold in female subjects (HR=2.77; 1.96–3.92) with newly diagnosed diabetes (>11.1mmol/l). Furthermore, if OGTT results were assessed for each FPG-based classification group, mortality increased with OGTT-based classification; e.g. among male subjects in the normoglycaemic group, those with post-load glucose >11.1mmol/l have a 79% increase in relative risk of death compared with those with post-load glucose <7.8mmol/l (HR=1.79; 1.22–2.62).
What Cellular Mechanisms Link Dysglycaemia with Cardiovascular Risk?
The normal cellular function of glucose oxidation is to supply nicotinamide adenine dinuleotide (NADH) equivalents for oxidative phosphorylation. Hyperglycaemia elicits increased glucose oxidation and the overproduction of NADH, disturbing the mitochondrial membrane potential and favouring the generation of superoxide by the mitochondrial electron-transport chain.
Overproduction of superoxide and other reactive oxygen species (ROS) results in increased levels of oxidative stress. In the absence of adequate compensatory responses from endogenous antioxidant systems, ROS may directly oxidise DNA, protein, and lipids, but also activate a variety of stress-sensitive intracellular signalling pathways with detrimental consequences (e.g. neovascularisation, proliferation of smooth muscle cells and cardiomyopathy (for review see Brownlee et al. 2001).21
At a physiological level, insulin resistance, resulting in impaired glucose tolerance, would lead to postprandial hyperglycaemia. Hyperglycaemia elicits cellular oxidative stress, resulting in endothelial dysfunction, cardiovascular disease, and diabetes resulting from β-cell impairment.22
The Changing Metabolic Status of the Population
A worrying trend has been the shift in ‘the normal’ metabolic status of the population. A 2008 study by Wilhelmsen et al. assessed changes in cardiovascular risk factors in cross-sectional population samples of 50- year-old men in Gothenberg examined at 10-year intervals between 1963 and 2003.23 Over this 40-year period decreases were observed in serum cholesterol levels (from 6.4 to 5.5mmol/l), blood pressure (138/91 to 135/85), and the prevalence of smoking (56 to 22%). However, during the same period, body mass index (BMI; 24.8 to 26.4 kg/m2), waist circumference (87 to 95cm), levels of triglycerides (1.3 to 1.7mmol/l), and prevalence of diabetes (3.6 to 6.6%) all increased. Conversely, the degree of leisure time physical activity decreased from 32% in 1963 to 24% in 2003. Although the rate of acute myocardial infarction (AMI) within the population reduced by more than 50% between 1975 and 2004, this may be attributed to improved medications for the treatment of hypertension and hyperlipidaemia. However, with the exception of the prevalence of smoking, all lifestyle-related cardiovascular risk factors within the population have deteriorated. This particular population has become more obese, less physically active and the prevalence of diabetes has increased by 83%.
This metabolic picture of the modern day myocardial infarction (MI) patient is supported by evidence from the GAMI trial.24 Compared with age- and sex-matched control subjects, MI patients without known diabetes had significantly higher triglyceride levels, FPG, post-load glucose levels and proinsulin levels, while levels of high-density lipoprotein cholesterol (HDL-C) were reduced (see Figure 3). Hence, during this period the stereotypical MI patient has ‘evolved’ from the relatively lean chain-smoker to a relatively sedentary, obese subject with the metabolic syndrome. Such patients are characterised by abdominal obesity, insulin resistance, dysglycaemia, diabetic dyslipidaemia and low-grade inflammation (increased levels of C-reactive protein, CRP).
Dysglycaemia and Cardiovascular Risk – The Perils of ‘Hidden’ Hyperglycaemia
A recent meta-analysis by the Emerging Risk Factors Collaboration assessed the impact of diabetes on the risk of vascular disease.25 This analysis included data for 698,782 people from 102 prospective trials, including 52,765 non-fatal or fatal vascular outcomes (8.49 person years at risk) with a median time to first outcome of 10.8 years.
Compared with non-diabetic subjects, the RR (adjusted for age, sex, BMI and systolic blood pressure) of coronary heart disease (CHD), coronary death, non-fatal MI, unclassified stroke, and other vascular deaths in subjects with diabetes was increased by 73–131%. Similarly, compared with non-diabetic subjects, the RR of CHD was elevated in patients with diabetes regardless of gender, age, smoking status, BMI tertile or systolic blood pressure tertile. However, the impact of diabetes on the RR of CHD was significantly higher in women compared with men (p<0.001), in non-smokers compared with smokers (p<0.001), in the youngest (40–59 years) versus oldest (≥70 years) age group (p<0.001), and in the lowest versus highest tertiles of BMI (p=0.00143) and systolic blood pressure (p<0.001). Thus the impact of dysglycaemia seems to be greater on traditionally lower-risk populations (e.g. women, non-smokers). Even among people with no known history of diabetes, FPG is non-linearly associated with risk of vascular disease.25
Evidence from the Euro Heart Survey also demonstrates the impact of diabetes on cardiovascular outcomes.26 This study classified subjects with CAD and available OGTT or FPG data, based on their glucometabolic state, and correlated this with one-year mortality. Subjects classified as having diabetes, either previously diagnosed (n=1,425) or newly diagnosed (n=452), both had greatly increased risk of one-year mortality (HR=2.4; 1.5–3.8 and HR=2.0; 1.1–3.6, respectively). However, IGR was not an independent predictor for one-year mortality in this study (HR 1.1, 95% CI 0.6–1.9). Taken together these studies confirm that patients with diabetes, newly diagnosed diabetes or pre-diabetes are at increased risk of mortality and cardiovascular events.
Multifactorial Approach to the Treatment of Diabetes
As we have outlined here, diabetes and CVD are very closely intertwined and, hence, the diagnosis and treatment of diabetes, pre-diabetes and CVD has been outlined in joint guidelines issued by the European Society of Cardiology and the European Association for the Study of Diabetes (ESC/EASD).27 These guidelines advocate a multifactorial approach addressing lifestyle, glycaemic control, and cardiovascular risk factors. The guidelines recommend structured education concerning lifestyle modification (e.g. diet, exercise, smoking cessation) and designate target levels for blood pressure (<130/80mmHg; <125/75mmHg in the case of renal impairment), glycosylated haemoglobin A1c (HbA1c; ≤6.5%), FPG (<6.0mmol/l), total cholesterol (<4.5mmol/l), low-density lipoprotein cholesterol (LDL-C) (<1.8mmol/l), HDL-C (male >1.0, female >1.2mmol/l) and triglycerides (<1.7mmol/l), in addition to other cardiometabolic risk factors.
Several trials have assessed the effect of an intensive multifactorial approach, utilising glycaemic control and the use of renin–angiotensin system blockers, aspirin, and lipid-lowering agents, in the management of diabetes and the associated cardiovascular risk. In the Steno-2 trial Danish subjects with T2DM and microalbuminuria (n=160) were randomly assigned to receive conventional multifactorial treatment or intensive, target-driven treatment, as currently recommended by the guidelines of the ADA28 and the ESC/EASD,27 for a mean period of 7.8 years.29
Although intensive treatment reduced the number of deaths already within the treatment period, the relatively small number of deaths precluded any definitive assessment of effects on mortality. Therefore, a pre-defined 5.5-year observational extension was included mainly to assess the primary end-point of time to death. 30 Although baseline levels were similar between groups, throughout the 7.8-year treatment period levels of HbA1c, systolic blood pressure, total cholesterol and LDL-C were consistently lower in the intensive treatment group than in those receiving conventional treatment. In contrast, at the end of the 5.5-year follow-up period, likely due to more intensive treatment among patients originally receiving conventional therapy, the differences in these risk factors between the groups had greatly narrowed. After 13.3 years of follow-up, intensive therapy was associated with a reduced risk of death from any cause (HR=0.54; 0.32–0.89; p=0.02), a reduced risk of death from cardiovascular causes (0.43; 0.19–0.94; p=0.04), and an absolute risk reduction of 29% for any cardiovascular event (HR=0.41; 0.25–0.67; p<0.001). This yielded a number needed to treat (NNT) for 13 years to avoid one event of five for any-cause death, eight for cardiovascular death, and three for any major cardiovascular event.
Therefore, in patients with type 2 diabetes, intensive intervention has sustained beneficial effects with respect to cardiovascular events, death from any cause and from cardiovascular causes. Another recent study assessed whether the poor prognosis of patients with diabetes and CAD was due to the underutilisation of evidence-based medicine (EBM, the combined use of β-blockers, renin–angiotensin system blockers, antiplatelet therapy and statins) or revascularisation procedures.31 The impact of EBM and revascularisation procedures on mortality and cardiovascular events was assessed in CAD patients with (n=2,063) and without (n=1,425) diabetes in the Euro Heart Survey on Diabetes and the Heart. EBM was used in only 44% of subjects with diabetes and 43% of those without diabetes, and revascularisation in 34 and 40% of patients, respectively. Compared with the impact of these two approaches in those without diabetes, both EBM (HR=0.37; 0.20–0.67, p=0.001) and revascularisation (HR=0.72; 0.39–1.32, p=0.275) improved one-year survival in patients with diabetes. The NNT to avoid one cardiovascular event using EBM was 32 and 141 in subjects with and without diabetes, respectively; corresponding NNTs for fatality were 24 and 1,826. The NNT to avoid one cardiovascular event using revascularisation was 14 and 41 in subjects with and without diabetes, respectively; corresponding NNTs for fatality were 34 and 105. This study highlights the utility of EBM and revascularisation to improve the prognosis of CAD patients with diabetes.
Although these clinical studies demonstrate the benefits of target-based, multifactorial management of diabetes and cardiovascular risk, achievement of risk-factor targets in the ‘real-world’ setting is disappointing. An analysis of the Euro Heart Survey demonstrated that actual levels of blood lipids, blood pressure, FPG and HbA1C were 23–57% above the targets for 1998–2003 and 32–92% above the current (2007) targets.32
Are Patients with T2DM Living Longer and Free of Complications?
Despite the high-risk population of the Diabetes Mellitus and Insulin Glucose Infusion in Acute Myocardial Infarction 2 (DIGAMI 2) trial – patients with T2DM and suspected acute MI – the overall study mortality was only 18.4%. This was considerably lower than the predicted mortality (22–23%). Data from a Swedish percutaneous coronary intervention (PCI) registry (n=57,708, including 10,857 with diabetes) indicate a two-year mortality of around 18% in a real-world, high-risk population ST-elevation MI (STEMI) and diabetes.33
Two recent trials, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial34 and the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial, 35 failed to show a statistically significant benefit from intensive glycaemic control to achieve an HbA1c level of <6.5%. However, in both ACCORD and ADVANCE, two-year mortality was well below expected levels. Such low event rates in clinical studies may be misleading and indicative that trial populations do not reflect the risk level of the real-world patient.
Although trial data have demonstrated the potential of target-driven, evidence-based medicine to improve outcomes in patients with diabetes, the implementation and execution of these regimens in clinical practice needs to improve. This may involve tailoring therapy to the individual patient to achieve multifactorial goals commensurate with their level of cardiovascular risk. Other approaches may involve methods to improve patient adherence and compliance with the prescribed therapeutic regimen.