Epidemiology of Coronary and Peripheral Atherosclerosis in Diabetes

Prevalence of Impaired Glucose Metabolism in Patients with Manifest Atherosclerosis

In patients with manifest coronary heart disease (CHD), based on the Ludwigshafen Risk and Cardiovascular Health (LURIC) study, approximately 20% of men and 25% of women had previously diagnosed diabetes. Newly detected diabetes was present in another 15% of men and approximately 10% of women. More detailed data on the prevalence of abnormal glucose regulation in patients with manifest CHD across Europe were collected in the Euro Heart Survey on Diabetes and the Heart. This survey involved 110 centers in 25 countries, recruiting more than 4000 patients referred to a cardiologist because of suspected coronary artery disease (CAD), including acute coronary syndrome (ACS) patients and those with stable symptoms. Of these patients, 31% had manifest diabetes. In 1920 patients without known diabetes, an oral glucose tolerance test (OGTT) was performed. The prevalence of impaired glucose tolerance (IGT) in those with ACS was 36%, with an additional 22% having newly detected diabetes. The corresponding figures in those with stable CAD were 37% and 14%, respectively. In another study from Spain, in 662 consecutive patients admitted to the hospital without a previous diagnosis of diabetes who were referred for coronary intervention in 2005 and 2006, the prevalence of diabetes was 45%. Analyses of more than 120,000 patients from randomized clinical trials in the late 1990s, at least 30% of them coming from the United States, revealed a prevalence of diabetes of 23% in women and 15% in men. Yet this certainly represents an underestimation as a result of selection bias and, most commonly, absence of systematic screening for incident diabetes.

Impaired glucose homeostasis (IGH) is also prevalent in this population. IGH comprises impaired fasting glucose (IFG) and IGT. IFG has been defined as a fasting glucose range of 110 to 126 mg/dL, and fasting glucose tolerance is pathologic if it exceeds 140 mg/dL according to 2-hour glucose levels after oral intake of 75 g of glucose (OGTT). Prevalence of these two entities may vary in different populations and may also not be consistently present in the same individual.

In the previously mentioned study from Spain in 662 consecutive patients without a previous diagnosis of diabetes who were referred for coronary intervention in 2005 and 2006, IGT was present in 24.5%, yet IFG was seen in only 1%. Thus again, more than two thirds of this population with manifest atherosclerotic disease had an abnormal glucose metabolism. In the LURIC study in patients with manifest CHD, approximately 20% of men and 15% of women showed an abnormal OGTT result. Putting these figures into perspective, approximately two thirds of patients with manifest cardiovascular disease (CVD) had impaired glucose metabolism, as results from the LURIC study have shown.

Thus, impaired glucose metabolism including manifest type 2 diabetes, pathologic oral glucose tolerance, and IFG are prevalent in patients with manifest atherosclerosis. In particular, manifest type 2 diabetes is at least twice as frequent in CHD patients compared with those free of CHD.

Diabetes Mellitus and Subclinical Atherosclerosis

Subclinical atherosclerosis represents the early manifestation of vascular disease without clinical symptoms. It can be assessed in all three major vascular beds by noninvasive but also by invasive imaging techniques. For carotid atherosclerosis, measurement of the intima-media thickness (IMT) or the presence of plaque on high-resolution transcutaneous ultrasound represents the method of choice. In the coronary vascular bed, computed tomography (CT) seems promising, but more precise data can be gathered by several invasive techniques such as intravascular ultrasound (IVUS) or optical coherence tomography (OCT). For the peripheral vascular bed, transcutaneous ultrasound and the ankle-brachial index (ABI) can be used ( Fig. 7-1 ). It has been well documented that the presence of subclinical atherosclerosis is associated with a higher incidence of clinical manifestations, yet in a number of studies the incremental value for risk prediction of these clinical manifestations (e.g., for IMT) above and beyond various risk scores such as the Framingham risk model did not yield clinically relevant results. At present, evidence is accumulating that coronary calcium scoring by CT may be superior to various blood biomarkers and also may produce significant incremental value over and above risk scoring. Epidemiologic data on the prevalence of subclinical coronary atherosclerosis and its association with incident CVD, as well as with all-cause mortality, have been published from several studies. One of the largest is the Cardiovascular Health Study (CHS), in which 1343 patients with diabetes, 1432 patients with IGT, and 2421 normoglycemic patients were identified by World Health Organization (WHO) criteria at the baseline examination in 1989-1990 and were followed on average for 6.4 years. Diabetic patients showed a higher prevalence of clinical and subclinical CVD at baseline, and the presence of subclinical disease was strongly related to CHD, stroke, and heart failure. This was particularly pronounced in diabetic patients, in whom the incidence of CHD and stroke was increased almost twofold in those presenting with subclinical disease versus those without ( Fig. 7-2 ). A similar relative increase in all three endpoints, but on a lower level, was seen in patients with IGT. Brohall and colleagues published a systematic review of the relevance of carotid IMT in patients with type 2 diabetes mellitus and IGT and focused on the differences between IMT in diabetic patients with IGT versus controls. They included 23 studies with 20,111 patients. Among those were 4019 with diabetes and 1110 with IGT. In 20 of the 23 studies, diabetic patients showed an increased IMT compared with individuals in the reference group, whereas in patients with IGT, the increase in IMT was approximately one third of that observed in patients with diabetes. These findings suggest an older vascular age in diabetic patients of approximately 10 years, and further conclusions drawn from this review estimated an increased relative risk for myocardial infarction (MI) and stroke in the presence of diabetes of almost 40%.

Several studies have looked into carotid atherosclerosis in the setting of the metabolic syndrome. Prospective data from the Bruneck study examined 888 patients aged 40 to 79 years, among whom 303 fulfilled the WHO criteria and 152 fulfilled the National Cholesterol Education Program—Adult Treatment Panel III (NCEP-ATP III) criteria for metabolic syndrome. Five-year changes in carotid status and its relation to incident fatal and nonfatal CHD were assessed. Patients with metabolic syndrome showed increased 5-year incidence and progression of carotid atherosclerosis, and also the incidence of clinical events was increased twofold compared with controls. This has recently been confirmed in the Multi-Ethnic Study of Atherosclerosis (MESA), in which individuals with metabolic syndrome or diabetes had a higher incidence and absolute progression of coronary artery calcification (CAC) compared with patients without diabetes. In addition, progression predicted future CHD events.

Several studies also assessed the prevalence of CAC in various populations. Data from the Dallas Heart Study, a large population-based study in which the individual and joint associations among metabolic syndrome, diabetes, and atherosclerosis as defined by CAC were assessed, suggested that both metabolic syndrome and diabetes mellitus are independently associated with an increased prevalence of atherosclerosis, with the highest prevalence seen in those fulfilling both criteria. In another study, by Wong and colleagues, 1823 patients aged 23 to 79 years were screened for CAC. Of these patients, 279 had metabolic syndrome and 150 had diabetes mellitus. Prevalence of CAC clearly increased from the reference group to those with metabolic syndrome or diabetes, with 53.5%, 58.8%, and 75.3%, respectively, among men and 37.6%, 50.8%, and 52.6%, respectively, among women. CAC also increased with the number of components of the metabolic syndrome (0 to 5) from 34% to 58%. Risk assessment by the Framingham score estimated 41% of patients with metabolic syndrome as having a more than 20% per 10 years increased risk for CHD or a CAC greater than 75% for age and gender. Diabetic patients had a higher incidence and progression of CAC, and progression of CAC predicted clinical CHD events ( Fig. 7-3 ). More recently, Elkeles and colleagues prospectively evaluated CAC score as a predictor for cardiovascular events in type 2 diabetes based on 589 patients from the PREDICT study. Follow-up was 4 years, and first CHD and stroke events were identified as primary endpoints. The CAC score was found to be a highly significant independent predictor, with a doubling of calcium being associated with a 32% increase in risk of events—which only slightly decreased in multivariable analysis considering traditional risk factors, and even after adjustment for homocysteine, c-reactive protein (CRP), and the homeostasis model assessment (HOMA) index. The only variable that predicted primary endpoints independently of CAC scoring was the HOMA index. CAC scoring contributed significantly to improved discrimination over and above the Framingham CHD risk score as well as the United Kingdom Prospective Diabetes Study (UKPDS) CHD and CVD risk scores, with an increase in the area under the curve on average of 0.1 (0.63 to 0.73).

Thus, among various measures of subclinical disease, CAC scoring may represent the most promising tool to improve risk prediction in asymptomatic patients with type 2 diabetes mellitus.

Clinical Manifestations of Atherosclerosis in Prediabetes, Metabolic Syndrome, and Type 2 Diabetes