have shown significant reductions in the incidence of cardiovascular events, including acute myocardial infarction and stroke [16]. Thus, the reduced mortality rates are not solely due to improved survival relating to treatment of complications but are, at least in part, due to the prevention of complications occurring. These improvements in cardiovascular risk have been more pronounced among adults with diabetes than in the general population. There was a 68% reduction in the rate of acute myocardial infarction among people with diabetes, with 141 events per 10,000 people in 1990 and 46 events per 10,000 people in 2010, whereas among people without diabetes, there was a 31% reduction, with 38 events per 10,000 people in 1990 and 26 events per 10,000 people in 2010. Diabetes still infers a greater risk of myocardial infarction, but the magnitude of the risk difference has reduced substantially. Reductions in rates of stroke were less substantial, such that myocardial infarction is now as common as stroke, having previously been significantly more common.
It is not possible from these large population-based studies to explain the determinants of the trends observed. Nevertheless, there are likely to be numerous contributing factors including population-wide risk factor trends, such as the decline in rates of cigarette smoking; improvements in the management of risk factors such as blood pressure and lipids in patients with diabetes; and advances in acute clinical care, such as improved access to and success of coronary revascularisation. In recent decades, numerous randomised control trials have shown the benefits of modifying different cardiovascular risk factors in high risk patients and there has been increasing use of medications such as statins, aspirin and renin-angiotensin system antagonists. Greater focus on improving systems of care and risk factor modification for patients with diabetes in particular may be responsible for the greater reductions seen in this population compared to the general population. It is also possible that more widespread screening for diabetes in asymptomatic individuals has led to there being an increased number of individuals in the diabetes population who have a short duration of disease and overall lower cardiovascular risk, although this is likely to be, at most, a minor contribution to the observed trends [16]. On the other hand, increased rates of screening could also have led to real improvements in cardiovascular disease prevention through early monitoring and intervention targeting glycaemia, blood pressure and lipids.
Glycaemia and Cardiovascular Risk
Cardiovascular disease has long been recognised as the most significant cause of morbidity and mortality in diabetes. Nevertheless, studies attempting to quantify the degree of increased risk inferred by diabetes or elucidate the exact nature of the relationship between different levels of glycaemia and cardiovascular outcomes have not always had consistent results.
Glycaemic thresholds for the diagnosis of diabetes were determined largely on the basis of the relationship between glycaemia and prevalent microvascular disease, particularly retinopathy [17]. Fasting plasma glucose, 2-h plasma glucose following an oral glucose load and haemoglobin A1c (HbA1c) all have narrow threshold ranges within which the prevalence of retinopathy begins to increase significantly [18]. On the other hand, the relationship between glycaemia and cardiovascular outcomes has been reported as a linear relationship with progressively lower risk even below diabetes thresholds [19]. A cohort analysis of the seminal United Kingdom Prospective Diabetes Study (UKPDS) found near linear relationships between HbA1c and incident cardiovascular complications among people with diabetes (Fig. 4), with no thresholds of glycaemia observed [20]. Every 1% lower level of mean HbA1c was associated with a 14% lower risk of myocardial infarction. The UKPDS enrolled patients with recently diagnosed diabetes.
Fig. 4. Association of HbA1c with mortality and incident cardiovascular complications among 4,585 people with type 2 diabetes from the UKPDS. Updated HbA1c is the mean of baseline and annual HbA1c levels during follow-up calculated for each participant. Data from Stratton et al. [20].
More recent data suggest that there may in fact be a glycaemic threshold, above which the risk of cardiovascular events begins to rise. Among people with established type 2 diabetes in the Action in Diabetes and Vascular disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial, risk of macrovascular disease and mortality began to rise within a threshold range of HbA1c around 6.5 and 7.0% [21]. Above the threshold there was a log-linear relationship between HbA1c and macrovascular complications such that for every 1% gain in mean HbA1c, there was a 38% higher risk of a macrovascular event. Below the threshold, there was no significant association between HbA1c levels and risk. Similarly, in a meta-analysis of individual-level data from 102 prospective studies with almost 700,000 participants, fasting plasma glucose was found to have a non-linear relationship with coronary heart disease and stroke (Fig. 5) [4]. Below 5.6 mmol/L there were no material associations between fasting plasma glucose and vascular risk among people with no known history of diabetes. Understanding the relationship between glycaemia and cardiovascular risk is a key component of determining which people may benefit from targeted preventive interventions.
Fig. 5. Hazard ratios (HRs) for coronary artery disease by mean baseline fasting plasma glucose (FPG) concentration in patients without known diabetes. Patients grouped into 0.5 mmol/L intervals (e.g., HR calculated for patients with FPG 5.5–6.0 mmol/L and plotted against their mean FPG). With permission from Emerging Risk Factors Collaboration [4].
Is Diabetes a Risk Equivalent to Previous Coronary Disease?
A widely cited Finnish study published in 1998 found that people with diabetes and no history of myocardial infarction had as high a risk of myocardial infarction as people with a history of prior myocardial infarction but not diabetes (7-year incidences of 20.2 and 18.8% respectively) [22]. The view that diabetes is a risk equivalent to established coronary vascular disease for incident cardiac events is still often espoused today [10]. However, it is probably not true in most populations. A meta-analysis of 13 studies involving more than 45,000 patients found the risk of myocardial infarction was 43% lower in patients with diabetes but no personal history of coronary disease compared to patients without diabetes but with prior myocardial infarction [23]. A contemporary study from a membership-based health service in the United States, conducted using electronic medical record data from over 1.5 million adults, also found existing coronary disease to be associated with a higher incidence of myocardial infarction than diabetes alone (22.5 vs. 12.2 events per 1,000 person-years) [24]. Knowing whether or not diabetes is a risk equivalent to known coronary disease is relevant in guiding the approach to management of cardiovascular risk factors in patients with diabetes. If diabetes is a risk equivalent, perhaps intensive therapy, as used in secondary prevention of ischaemic heart
