Diabetes mellitus affects 18.2 million people in the United States1; 13 million were known to have diabetes and 5.2 million went undiagnosed in 2002. The majority of elderly patients with diabetes have type 2, which is associated with increasing age, obesity, family history, impaired glucose metabolism, a history of gestational diabetes, race, ethnicity, and a sedentary lifestyle. When analyzed by age it is clear that diabetes affects a higher percentage of older Americans (Figure).
With life expectancy continuing to increase, we can also expect an increase in the prevalence of diabetes. How one defines diabetes—by fasting plasma glucose (FPG) alone or 2-hour postchallenge plasma glucose (2-hr PG) during a 75-g oral glucose tolerance test—will affect the number of cases diagnosed. This is a very important point, because treatment of patients with diabetes includes management of not only hyperglycemia, but also more aggressive treatment of other commonly associated conditions such as hypertension and dyslipidemia.
The diagnosis of diabetes itself is considered a coronary heart disease risk equivalent that requires intensified cardiovascular risk factor modification. American Diabetes Association (ADA) Standards of Medical Care for individuals older than age 40 with diabetes and total cholesterol of 135 mg/dL or above but no overt cardiovascular disease include HMG-CoA reductase inhibitor (statin) therapy to achieve a low-density lipoprotein (LDL) cholesterol reduction of 30% to 40% regardless of baseline LDL cholesterol levels. The primary goal is an LDL cholesterol less than 100 mg/dL.2 For patients with diabetes and overt cardiovascular disease who are at very high risk for further events, treatment should include a high-dose statin with an LDL cholesterol goal of less than 70 mg/dL. The diagnosis of diabetes in the elderly also makes aspirin therapy for primary and secondary prevention of coronary heart disease an imperative, along with smoking cessation. Also recommended in diabetic patients older than 55 years with or without hypertension, but with another cardiovascular risk factor (history of cardiovascular
disease, dyslipidemia, microalbuminuria, or smoking), is treatment with an angiotensin-converting enzyme inhibitor to reduce the risk of cardiovascular events.3
The recently revised ADA diagnostic criteria for diabetes and impaired glucose regulation are listed in the Table. The fasting glucose cutpoint used to define normal glucose tolerance was lowered to less than 100 mg/dL, with impaired fasting glucose (IFG) now defined as 100 to 125 mg/dL. The 2-hr PG cutpoint used to define normal glucose tolerance remains at less than 140 mg/dL with impaired glucose tolerance (IGT) defined as 140 to 199 mg/dL. Both IFG and IGT are considered prediabetes. They have a similar risk of progression to diabetes, but they may have different risks of cardiovascular disease. The Expert Committee on the Diagnosis and Classification of Diabetes concluded that the FPG and the 2-hr PG (but not the glycosylated hemoglobin [A1C] test) remain the tests of choice for the diagnosis of diabetes.4 Using the 2-hr PG for the diagnosis of diabetes is more sensitive in most populations, particularly the elderly, while the FPG is more reproducible, less costly, and more convenient.
Data from oral glucose tolerance testing carried out in 2,844 subjects aged 40 to 74 years without known diabetes identified more people with diabetes based on elevated 2-hr PG levels than using the FPG alone.5 A 2-hr PG of 200 mg/dL or greater in the setting of an FPG less than 126 mg/dL defines isolated postchallenge hyperglycemia (IPH). In subjects aged 60 to 64 years, the weighted age-specific prevalence of diabetes diagnosed by the FPG was 8%, which was almost identical to the prevalence of diabetes diagnosed by IPH. In subjects aged 70 to 74 years, the values were 7% prevalence of diabetes based on FPG and 9.1% based on IPH. This confirmed that with advancing age, IPH was more common and reaffirmed the use of the oral glucose tolerance test to diagnose diabetes in elderly subjects.
IPH has been associated with increased cardiovascular disease mortality in a variety of studies. The Rancho Bernardo study, which performed 2-hour 75-g oral glucose tolerance tests in older subjects aged 50 to 89 years without known diabetes or myocardial infarction, found that a diagnosis of diabetes would have been missed in 88 of 125 women with diabetes (70%) and 60 of 133 men with diabetes (45%) using FPG alone.6 Seven-year age-adjusted cardiovascular mortality rates in women were doubled when isolated 2-hr PG levels of 200 mg/dL or above were used.
There are also important data from the Diabetes Epidemiology: Collaborative Analysis of Diagnostic Criteria in Europe study group, a collaborative study of 22 cohorts in Europe with baseline measurements of 29,714 subjects aged 30 to 89 years with 11-year follow-up.7 They found that high FPGs, as well as very low FPGs below 80 mg/dL, were associated with an increased risk of death, resulting in a J-shaped curve. Individuals with IFG of 110 to 125 mg/dL showed hazard ratios for all-cause mortality, cardiovascular mortality, and noncardiovascular mortality of 1.1, while patients with previously undiagnosed diabetes and an FPG of 126 mg/dL or greater had hazard ratios of 1.6 for these pa-rameters. The relationship between cardiovascular mortality and 2-hr PG was linear, with IGT showing hazard ratios of 1.5 for all-cause mortality, 1.4 for cardiovascular mortality, and 1.6 for noncardiovascular mortality, while patients with previously undiagnosed diabetes based on a 2-hr PG value greater than or equal to 200 mg/dL (IPH) showed hazard ratios of 2.0 for all-cause mortality, 1.9 for cardiovascular disease mortality, and 2.1 for noncardiovascular disease mortality. They concluded that all-cause, cardiovascular, and noncardiovascular disease mortality hazard ratios for subjects with 2-hr PG greater than 180 mg/dL were not significantly different from subjects with known diabetes. This was significantly higher than that for undetected diabetes based on FPG. They also concluded that all-cause and cardiovascular disease mortality was significantly higher for subjects with IGT than for IFG.
Further evidence to support the association of IGT with cardiovascular risk factors was obtained from the Baltimore Longitudinal Study on Aging.8 This was a prospective study of community-dwelling volunteers aged 21 to 96 years. They determined cardiovascular risk factors in 937 subjects without diabetes or cardiovascular disease at baseline and follow-up over 9.5 years. They excluded from analysis the 152 subjects who developed diabetes. When IFG was classified using the new criteria of 100 to 125 mg/dL, creating a similar baseline prevalence of IFG
and IGT, they found that 24.8% of the subjects remained with normal glucose tolerance throughout, while isolated IFG was seen in 21.2%, isolated IGT was seen in 16.3%, and both IFG and IGT were seen in 37.7%. The subjects with normal glucose tolerance or IFG were younger and had lower triglycerides and higher high-density lipoprotein (HDL) cholesterol levels than those with IGT and both IFG and IGT. There were 28% of subjects with IGT and 47% with both IFG and IGT meeting the criteria for metabolic syndrome, which was much greater than in subjects with normal glucose tolerance (4%) or IFG (6%).
The International Diabetes Federation IGT/IFG Consensus Statement workshop reported that the determinants of FPG and 2-hr PG differ.9 Elevated hepatic glucose production and a defect in early insulin secretion are characteristic of an elevated FPG. Peripheral insulin resistance, particularly at the skeletal muscle, is most characteristic of an elevated 2-hr PG. They found the prevalence of IFG tended to plateau in middle age, whereas the prevalence of IGT continued to rise in the elderly coincident with declining muscle mass (sarcopenia). While both IFG and IGT are similarly associated with a greater risk of developing diabetes (1%—10% per year), IGT is more strongly associated with cardiovascular disease. These reports have placed specific emphasis on the control of postprandial plasma glucose to prevent cardiovascular disease. Postprandial hyperglycemia may have a direct toxic effect on the vascular endothelium, mediated by oxidative stress that is independent of other classic cardiovascular risk factors.10
A retrospective claims-based Medicare dataset found the mortality rate was 100.2/1,000 person-years among seniors with diabetes compared with 60.6/1,000 person-years without diabetes, resulting in an age-adjusted relative risk of 1.83.11 The mortality risk associated with diabetes decreased with increasing age, but remained significantly elevated even among those 85 years and older. In contrast, the absolute excess mortality (defined as the difference in mortality rates) increased with increasing age. The excess mortality was 28.7/1,000 for the 65- to 69-year age group and 66.3/1,000 among those 85 years or older. These data support the goal of diabetes prevention to decrease mortality.
Lifestyle interventions are highly effective in delaying or preventing the onset of diabetes in people with IGT. The Diabetes Prevention Program found a 58% decrease in the onset of type 2 diabetes with a lifestyle modification program and a 31% decrease with metformin therapy in subjects with IGT who were followed up over 2.8 years.12 When subjects 60 years or older were analyzed, there was a 71% reduction in diabetes incidence with lifestyle intervention, while metformin was ineffective. The patients in the lifestyle modification group lost an average of 7% body weight and performed 150 minutes per week of moderate exercise. A low-fat (< 25%) diet was recommended, with caloric restriction added if needed. These are reasonable interventions for the prevention of diabetes in the elderly.
In conclusion, diabetes in the elderly is more than just hyperglycemia; it means a much greater burden for cardiovascular disease and therefore prevention of diabetes should be a priority. Good glycemic control is important
in elderly patients already diagnosed with diabetes, though larger reductions in morbidity and mortality may re-
sult from control of cardiovascular disease risk factors. Treatment of glucose, blood pressure, lipids, obesity, and platelet hyperactivity is needed. Treatment targets include an A1C less than 7%, blood pressure less than 130/80 mm Hg, LDL cholesterol less than 100 mg/dL, triglycerides less than 150 mg/dL, HDL cholesterol greater than 40 mg/dL in men and greater than 50 mg/dL in women, a body mass index less than 25 kg/m2 with aspirin therapy, and smoking cessation.13
Further research is needed to ascertain whether or not prevention of diabetes in the elderly will result in a decrease in cardiovascular events and mortality. Clinical trials are needed to see if there is a difference in cardiovascular morbidity and mortality in patients with normal glucose tolerance compared with IFG versus those with IGT and diabetes. Pharmacologic interventions are needed that will prevent the onset of diabetes in high-risk elderly patients with reversion to normal glucose tolerance a goal. The priority should be to improve cardiovascular outcomes and quality of life for the elderly.