Diabetes and Cardiovascular Disease in Older Adults: Current Status and Future Directions

Jeffrey B. Halter; Nicolas Musi; Frances McFarland Horne; Jill P. Crandall; Andrew Goldberg; Lawrence Harkless; William R. Hazzard; Elbert S. Huang; M. Sue Kirkman; Jorge Plutzky; Kenneth E. Schmader; Susan Zieman; Kevin P. High


Diabetes. 2014;63(8):2578-2589. 

In This Article

Prevention and Treatment for Diabetes and CVD

As suggested by the effects of exercise and dietary changes on mitochondrial and vascular function, addressing lifestyle risk factors can have a substantial impact on diabetes and CVD in the older population. In the Diabetes Prevention Program (DPP), a clinical trial with individuals at risk for type 2 diabetes, lifestyle intervention proved especially effective in preventing diabetes among older participants (Fig. 3), with a persistent effect for up to 10 years.[5,36] These participants lost more weight, and they showed a lower tendency, compared with middle-aged participants, to regain the weight they had lost.[5]

Figure 3.

Diabetes incidence rates by age-group in the DPP (36), demonstrating the effectiveness of the lifestyle intervention vs. placebo, especially in the oldest age-group. Reprinted and modified by permission of Oxford University Press on behalf of the Gerontological Society of America.

Obese, nondiabetic older adults randomized to dietary interventions or exercise maintained weight loss up to 30 months following the study.[37] Also, dietary changes and exercise each improved parameters of physical function, but they exerted additive effects when combined. These findings are consistent with those of the Look AHEAD (Action for Health in Diabetes) study in middle-aged and older people with type 2 diabetes, where weight loss and improved fitness lowered the risk for loss of mobility.[38] Lifestyle interventions also improve intrahepatic fat content, insulin secretion, insulin sensitivity, and metabolic risk factors for CHD,[39] and they increase the likelihood of partial remission of type 2 diabetes.[40]

However, weight-loss interventions remain controversial for older adults due to concerns that weight loss will exacerbate sarcopenia and frailty and that attempts to change lifelong habits will cause anxiety and distress. Yet in DPP, otherwise healthy older adults were more active and enthusiastic in adopting lifestyle changes and more successful in achieving their goals.[36] Older adults with diabetes and multiple comorbidities can have both sarcopenia and visceral obesity. In such individuals, weight loss could exacerbate age-related declines in physical and metabolic function, leading to development of the frailty syndrome.[3]

Treatment strategies used to address CVD risk factors in younger populations with diabetes also apply to older adults with diabetes. Clinical trials of interventions for hypertension and lipid disorders in these populations have been summarized.[2,3] Consistent, positive outcomes have been found in older people and people with diabetes, but few studies have included enough older people with diabetes to have adequate statistical power. Older adults may in some cases be more sensitive to therapy than younger adults. For example, in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, older adults responded similarly to protocol-driven therapies as younger participants did, and with some outcomes, such as LDL cholesterol and triglyceride levels, older participants fared better than their younger counterparts. However, in both the intensive and standard glycemia therapy arms of the trial, these same older adults were more prone to severe hypoglycemia than their younger counterparts. Although ACCORD was stopped early because of excessive mortality in the intensive therapy arm, this excessive mortality was confined to participants younger than 65 years.[41]

Primary percutaneous intervention (PCI) with drug-eluting stent is the treatment of choice for acute coronary syndromes, regardless of whether diabetes is present. However, mortality and rates of MI and stroke are higher among diabetic patients receiving PCI than among those receiving coronary artery bypass graft (CABG).[42,43] The Bypass Angioplasty Revascularization Investigation in 2 Diabetes (BARI 2D) study of patients with type 2 diabetes and coronary artery disease (CAD) showed no significant difference in the rates of death and major cardiovascular events between patients undergoing prompt revascularization and those undergoing medical therapy.[44] Although the majority of participants in these studies were in their 60s and 70s, results were not stratified by age.


Impairment of protective mechanisms against hypoglycemia (defined as a plasma blood glucose concentration of 70 mg/dL or lower) is a serious issue in individuals with diabetes.[45] Such impairments may worsen with duration of diabetes and most likely with age, and data from the ACCORD trial suggest a relationship between poor baseline cognitive function and the risk for severe hypoglycemia, which requires the assistance of others for recognition and treatment (Fig. 4).[41] Indeed, age is a strong predictor of acute hypoglycemia. Concerns about hypoglycemia and its potential impact are important factors in diabetes self-management. For example, patients with diabetic kidney disease, and particularly those who are older and receiving an intensive glucose-lowering regimen, are at increased risk for severe hypoglycemia.[3] However, severe hypoglycemia is a difficult outcome to study, and mild hypoglycemia is almost never reported. Thus, the potential impact of hypoglycemia is unclear. Hypoglycemia increases QTc interval, production of proinflammatory markers, platelet activation, and markers of oxidative stress, and it decreases endothelial function and myocardial blood flow.[3,45] Although these changes may increase the risk for CVD and death, it is not clear that hypoglycemia is a direct cause of these outcomes. Results from several studies indicate a correlation between severe hypoglycemia and future mortality.[3,46] However, correlative findings do not establish causality.

Figure 4.

Relationship between baseline cognitive function and risk for severe hypoglycemia in the ACCORD trial. Kaplan-Meier curves are shown for the proportion of subjects with severe hypoglycemia events according to baseline tertiles of the Digit Symbol Substitution Test (DSST) score. Crude incidence rates and 95% CIs are shown for each group. Log-rank test P = 0.0001. Hazard ratios (HRs) for the middle and highest score groups are with reference to the lowest DSST score group. Patients who scored in the worst tertile on the DSST had the highest rate of severe hypoglycemia, at 2.90%/year or approximately 10% (proportion 0.10) cumulatively over 4 years. The numbers below the x-axis are the actual number of patients in each tertile at each time point (41).