Older adults with diabetes are at higher risk for atherosclerotic cardiovascular disease (ASCVD) than younger adults with diabetes and older adults without diabetes. The rationale to implement ASCVD risk–lowering therapies in older adults with diabetes is compelling.
The variability in health status among older adults is pertinent. Those with robust health are more likely to tolerate and derive benefit from many therapies when compared with those who have more complex health including frailty.
Therapeutic Lifestyle Interventions:
Therapeutic lifestyle interventions (TLI) with diet and exercise are important components of ASCVD risk reduction in older as well as younger patient populations. The efficacy of intensive TLI in older patient populations was demonstrated in the Diabetes Prevention Program (DPP) and Look AHEAD (Action for Health in Diabetes) studies. In the DPP, participants aged >60 years (range 60–85) achieved greater reductions in risk for progression to type 2 diabetes (T2D) with TLI compared with younger age-groups or those receiving metformin. In Look AHEAD, older participants receiving TLI experienced greater improvements in physical functioning than younger subjects.
TLI components in the DPP and Look AHEAD were similar. Dietary interventions included meal replacements (shakes, food bars) for 1–2 meals a day to achieve caloric restrictions. Exercise prescriptions targeted 150–175 min of weekly physical activity. The primary outcome of reductions in ASCVD events in Look AHEAD was not achieved, but there were improvements in ASCVD risk with reduced body weight, waist circumference, and HbA1c and increased physical functioning and fitness.
Nutrition counseling is important for older adults who are at risk for undernutrition as well as obesity. Many older adults focus on food as a source of satisfaction and comfort, potentially viewing meal replacements as interfering with HRQL. Although there is no one optimal diet, the Mediterranean diet with an emphasis on intake of fresh fruit and vegetables, legumes, and nuts may have advantages for some older adults. This diet has been demonstrated to be superior to traditional diets in several studies that included seniors aged ≥65 to 90 years for weight loss, lipid lowering, blood glucose control, and CVD outcomes.
The American College of Cardiology (ACC)/American Heart Association (AHA) lipid-lowering guidelines recommend moderate-intensity statins for patients with diabetes aged 65–75 years without ASCVD and high-intensity statins for those with ASCVD. For adults aged ≥75 years, moderate-intensity therapy is recommended for those with and without ASCVD as it is presumed that this older group would be unable to tolerate higher dosages. These recommendations are consistent with those of the ADA and AGS with the added modification that decisions for intensity of therapy be based on health status rather than age alone. These guidelines counterbalance risk-based therapeutic intensity with recognition of the complexities of care and vulnerabilities associated with aging.
Nonstatin lipid-lowering agents are not generally recommended. Fibrates alone or in combination with statins are not recommended based on the absence of efficacy in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) and the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) trials. In one recent study of adults with ASCVD aged 56–72 years, ezetimibe reduced risk of ASCVD events when used with a statin. Subgroup analyses of participants with diabetes (27%) and those >75 years of age (15%) revealed significantly greater reductions in the primary outcome when compared with those without diabetes or <75 years of age, respectively. There was no subgroup analysis of subjects with diabetes >75 years of age.
Strong evidence supports the ACC/AHA recommendations for adults with diabetes aged <75 years, but data for older adults are more limited. A meta-analysis of studies using statins as primary prevention in 18,686 individuals with diabetes showed a 21% reduction in major vascular events for each mmol/L reduction in LDL cholesterol, with no differences between subjects younger or older than age 65 years. The Prospective Study of Pravastatin in the Elderly at Risk (PROSPER) randomly assigned men and women aged 70–82 years with preexisting ASCVD or one or more CVD risk factors to pravastatin 40 mg or placebo. Significant reductions in the primary outcome (fatal or nonfatal myocardial infarction [MI], stroke) were observed with pravastatin (hazard ratio 0.85) primarily in the secondary prevention group (22% reduction in primary outcome). Subgroup analysis of subjects with diabetes (11%) did not demonstrate a benefit with pravastatin; however, the small numbers limited the ability to interpret these findings.
For patients with diabetes aged ≥80 years, evidence for statin use is limited. In one population-based study of participants aged 66–96 years, all-cause mortality was reduced among those with diabetes receiving statin therapy (HR 0.47, 95% CI 0.32–0.71) compared with nonusers. Another retrospective observational study investigated the efficacy of statin therapy in 1,712 community-dwelling adults with diabetes aged ≥65 years (mean 81 ± 7) grouped according to health status. Subjects with mild, moderate, and severe impairments in health experienced a similar mortality benefit with statins irrespective of age, underlying clinical complexity, or frailty. The authors acknowledged that statin therapy in older individuals with severe health impairments remains controversial. No information was presented regarding adverse side effects in this vulnerable group of patients.
Statins have been associated with a spectrum of muscle concerns in older adults; however, concerns regarding statin-related adverse events in older patients are not substantiated in pooled analyses from clinical trials. Rhabdomyolysis is rare, but myalgias without creatinine phosphokinase elevations are common. This can be especially debilitating in an older population prone to frailty and diminished HRQL. Reports of statin-induced cognitive changes are offset by inconsistent data, with some reports suggesting benefit attributable to reduced inflammation and improved vascular function. Again, few trials include older adults with intermediate and poor health, indicating a need for further study.
In summary, given the powerful ASCVD risk associated with diabetes, there is strong rationale to use statins, especially in patients with known ASCVD. Although clinical data have clearly demonstrated aggressively lowering LDL as an optimal strategy in younger adults, in older patients these benefits are counterbalanced by risks associated with age, polypharmacy, frailty, other comorbidities, and limited life expectancy. A suggested strategy would be to start with low-intensity statin doses and gradually increase dosing as tolerated while monitoring for adverse side effects that prompt dose reductions or discontinuation. There are currently no studies using the new class of humanized monoclonal antibodies that inactivate proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors in older populations with diabetes.
Hypertension is common among older adults (prevalence rate ∼70%), but the associated ASCVD risks are compounded by diabetes. The ADA, AGS, and Eighth Joint National Committee (JNC 8) provide recommendations regarding benefits of BP-lowering therapy in older adults, emphasizing that goals should be tailored to individual patient characteristics in order to minimize risk for harm. The majority of these recommendations are based on studies focusing on systolic blood pressure (SBP); however, there is evidence that intensive control of diastolic blood pressure (DBP) can be also be associated with harm in older adults with diabetes. In one study of >34,000 subjects (aged 64.2 ± 12.1 years) with T2D without known CVD at baseline followed for 11 years, an increase in ASCVD events was observed with low as well as high DBP in the older subgroup, with the lowest risk observed with DBP of 80–90 mmHg.
Several trials have investigated treatment of hypertension as a way of reducing risk of ASCVD events. The observed benefits of BP-lowering therapies have been inferred primarily from larger trials that included only subsets of older subjects with diabetes. These trials consistently show that reducing BP from high to moderate levels reduces ASCVD risks. Some trials have demonstrated that more aggressive BP reductions achieve proportionally greater benefits but at the expense of greater iatrogenic risks, particularly in those with poor health or frailty. Falls and syncope are particularly concerning, as are issues of incontinence, fatigue, and diminished HRQL.
The Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE) trial randomized 11,140 adults with T2D (6,601 were aged ≥65 years) to combination perindopril with indapamide or placebo. SBP and DBP were reduced by 5.6 and 2.2 mmHg, respectively, with perindopril–indapamide compared with placebo. There were greater reductions in microvascular and CVD events in the older subgroup (relative risk reduction [RRR] 11% vs. 6% in those aged <65 years).
The UK Prospective Diabetes Study (UKPDS) compared tight (<150/85 mmHg) versus less tight (<180/105 mmHg) therapy in 1,148 patients aged 56.4 ± 8.1 years with newly diagnosed T2D. The mean BP achieved was 144/82 and 154/87 mmHg, respectively. Significant RRR demonstrated with tight control for diabetes-related death, microvascular disease, and stroke at 10 years were not sustained at the posttrial follow-up. No age-based subgroup analysis was performed; however, the average age of subjects at the posttrial evaluation was 63 ± 8 years, suggesting relevance to older adult populations.
The ACCORD–Blood Pressure (ACCORD–BP) trial compared intensive (target SBP <120 mmHg) with standardized (target <140 mmHg) treatment among adults aged 40–79 years with T2D and high risk of ASCVD. Intensive therapy did not reduce MI or all-cause mortality, but it was associated with a reduction in strokes (secondary outcome) with the number needed to treat of 89 over 5 years. Results were similar when subjects were grouped according to age (<65 and ≥65 years). Post hoc analysis of the International Verapamil-Trandolapril Study (INVEST) had findings similar to those of ACCORD–BP with no ASCVD benefit with BP targets of <130/85 vs. <140/95 mmHg among patients ≥55 years of age (mean age 66 ± 6 years).
The Systolic Blood Pressure Intervention Trial (SPRINT) is an important trial that renewed interest in intensified BP-lowering therapy that included older adults but excluded those with diabetes. Aggressive BP treatment to achieve SBP <120 vs. <140 mmHg resulted in greater reductions in CVD and non-CVD mortality that was more pronounced in subjects aged ≥75 years. These favorable findings are balanced against an almost doubling of serious adverse events with intensive treatment (hypotension, syncope, falls, electrolyte abnormalities, and acute kidney injury), which argues against imposing these more aggressive targets to older adults with diabetes.
In summary, pursuing moderate BP control according to ADA recommendations in older adults with diabetes according to health status is an appropriate strategy for reducing ASCVD risk without significantly increasing iatrogenic risks.
The ADA suggests that aspirin therapy may benefit older adults whose life expectancy exceeds the time frame for primary or secondary prevention trials. The AGS recommends aspirin 75–325 mg a day only for older adults with known CVD unless contraindicated. These recommendations are derived primarily from trials of younger populations without diabetes; however, age-specific trials now in progress may help elucidate the extent of potential benefits and risks for older individuals with diabetes in relation to cardiovascular events and mortality as well as subclinical disease associated with microemboli.
A study of Japanese individuals with diabetes but no ASCVD demonstrated a benefit of aspirin use on ASCVD events among subjects aged ≥65 years (6.3% vs. 9.2%, P = 0.047) but not <65 years. The observed benefits were counterbalanced by increases in serious gastrointestinal (GI) bleeding with four participants receiving aspirin and none receiving placebo. Another study showed no benefit from low-dose aspirin in older Japanese adults at high risk for ASCVD, approximately one-third of whom had diabetes. Low-dose aspirin again did not significantly reduce the composite outcome of cardiovascular death, nonfatal stroke, and MI but was associated with a reduced incidence of nonfatal MI (HR 0.53, P = 0.02) and transient ischemic attack (HR 0.57, P = 0.04), again at the expense of an increased risk of extracranial hemorrhage requiring transfusion or hospitalization (HR 1.85, P = 0.004).
Overall, data regarding efficacy of aspirin in older adults with diabetes is limited, with some evidence of therapeutic benefits (particularly for those with known ASCVD) that are counterbalanced by notable bleeding risks. A recent publication suggests a new algorithm to help decision making for aspirin use pertinent to older adults (still unvalidated), recommending a highly individualized approach to care that centers on patient preferences and tacitly counters the precept that management can and should be stratified in a standardized manner relative to specific levels of risk.
The ADA and AGS have similar recommendations for modification of glycemic goals in older adults according to health status. Although the AGS recommends an upper limit for HbA1c up to 75 mmol/mol (9%), there is evidence that many elderly adults are being treated more aggressively. The modified targets for HbA1c in older adults are based in part on the lack of evidence for ASCVD risk reduction as well as concern for harm with tight glycemic targets in older adults. Several large randomized controlled trials with relevance to an elderly population (average age at baseline 60–66 years) were unable to identify any ASCVD benefit with glycemic management strategies targeting HbA1c values of ≤42 to 48 mmol/mol (≤6–6.5%). The ACCORD trial was stopped prematurely because of an observed 22% increase in cardiovascular death with intensive therapy. The ADVANCE trial demonstrated neither benefit nor harm with intensive versus conventional therapy. Similar to findings in the ACCORD trial, more than 75% of intensively treated participants in the Veterans Affairs Diabetes Trial (VADT) were using thiazolidinediones or insulin, raising questions as to whether the negative results could be ascribed to the glucose-lowering strategies used.
The UKPDS legacy study and two recent studies with the sodium–glucose cotransporter 2 inhibitor (SGLT2i) empagliflozin and the glucagon-like peptide 1 receptor agonist (GLP-1RA) liraglutide each demonstrated benefits that have relevance to ASCVD risk management in older adults with T2D. In the UKPDS legacy study, participants with newly diagnosed T2D at enrollment demonstrated significant reductions for MI and all-cause mortality with a strategy of intensive versus conventional glycemic control. Many participants in this legacy analysis were eligible to be considered older adults based on age >65 years; however, subgroup analysis was not performed.
Selected cardiovascular outcome studies in older adults with diabetes
The Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG OUTCOME) is the first study demonstrating reductions in ASCVD events with glucose-lowering medication in obese or overweight subjects with T2D and CVD. RRR of 14% for MI and stroke (absolute risk reduction [ARR] 10.5% vs. 12.1%), 38% for cardiovascular death (ARR 3.7% vs. 5.9%), with a 35% decrease in hospitalizations for congestive heart failure (CHF) (2.7% vs. 4.1%) were observed with empagliflozin. Subgroup analyses demonstrated more favorable outcomes in subjects aged ≥65 years when compared with younger counterparts. Reductions in ASCVD events were observed early in the study (at 3 months), raising questions regarding the mechanisms for these favorable outcomes. Given the modest differences in HbA1c between the empagliflozin and placebo groups, it is likely that the observed reductions in BP, plasma volume, and weight contributed more to the favorable results than changes in glycemic control. The liraglutide and cardiovascular outcomes study also demonstrated significant reductions in the primary outcome of ASCVD events; however, subgroup analyses based on age demonstrated significance only in subjects <60 but not ≥60 years of age. Several large studies with dipeptidyl peptidase 4 inhibitors (DPP-4is) did not show any cardiovascular benefit. An increase in hospitalizations for CHF was observed with saxagliptin; however, a meta-analysis suggested no increase in CHF with any of the oral or injectable incretins.
Treatment Strategies for Achieving Glycemic Goals in Older Adults:
In addition to insulin, there are several different classes of oral or noninsulin injectable glucose-lowering therapies available for achieving glycemic control. There are no absolute contraindications to use of any specific agent; however, there are special considerations for their prescribing and monitoring in older adults. Insulin-providing (IP) therapies with insulin secretogogues or insulin are associated with risk for hypoglycemia, particularly in those with compromised renal function, which is more frequent in older adults with diabetes, emphasizing the need for caution when prescribing these agents. Patient education regarding hypoglycemia prevention, recognition, and treatment is important. Metformin is no longer contraindicated for individuals aged >80 years or for those who have mild chronic kidney disease (eGFR 45–60 mL/min).
The use of thiazolidinediones in older adults is limited by concerns for weight gain, edema, CHF, reductions in bone mineral density, and increased fracture risk. In one study, subjects with insulin resistance without diabetes randomly assigned to receive pioglitazone following a stroke experienced significant improvements in insulin sensitivity with a reduction in the combined primary outcome of stroke and MI compared with placebo. However, this was at the expense of more weight gain, CHF, and fractures. If a decision is made to use these agents, close monitoring for adverse effects is recommended.
DPP-4i's have a low incidence of side effects, do not cause hypoglycemia when used alone, and appear to be neutral in regard to ASCVD risk. Injectable GLP-1RAs can be associated with gastrointestinal side effects but do not have specific contraindications in the elderly. Despite the favorable results in EMPA-REG OUTCOME, preventive monitoring of older adults for dehydration, genital infections, and urinary tract infections (UTIs) with use of the SGLT2is is important.
In summary, although there are no absolute contraindications to use of any of the oral or injectable agents available for treating diabetes, there are special considerations for prescribing and monitoring of these agents in older adults that can impact cardiovascular safety, including risk for hypoglycemia and patient preference.
Hypoglycemia as a Cardiovascular Risk Factor
Despite recommendations to modify glycemic goals for older adults, particularly those with intermediate or poor health status, there is evidence that older adults may be treated too aggressively, increasing risk for hypoglycemia and associated adverse events including falls and cardiac and neurological events. In one study, >50% of participants aged ≥65 years with intermediate or poor health had HbA1c <53 mmol/mol (<7%). IP therapy was the most frequently used glucose-lowering therapy and has been observed to account for the majority of emergency department visits for hypoglycemia among adults aged ≥65 years.
Older adults with diabetes are particularly vulnerable to hypoglycemia because of impaired counterregulatory responses that interfere with the ability to detect and recover from hypoglycemia. Questions regarding the contribution of hypoglycemia to adverse outcomes were raised in the ACCORD, ADVANCE, and VADT studies, where severe hypoglycemia occurred more frequently in intensively treated subjects; however, the relationship between these events and mortality has been the subject of debate. In ACCORD, severe but not mild hypoglycemia was associated with an increase in mortality that was more pronounced with standard therapy. In ADVANCE, severe hypoglycemia was associated with an increased incidence of microvascular and macrovascular events as well as CVD and non-CVD mortality. In VADT, severe hypoglycemia was associated with progression of coronary artery calcifications in those receiving standard but not intensive glycemic management.
There are several suggested mechanisms for the contribution of hypoglycemia to risk for microvascular and macrovascular events. In studies using continuous glucose and cardiac monitoring devices, hypoglycemia was associated with an increase in arrhythmias and prolonged QT intervals. Enhanced sympathetic nervous system activation, catecholamine excess, and abnormal cardiac repolarization observed with hypoglycemia can potentially contribute to ASCVD risk and events.
Together, this information emphasizes the importance of minimizing risk for hypoglycemia among older adults with diabetes through patient education and use of therapies that are associated with low risk for hypoglycemia when possible.
In summary, older adults with diabetes are at high risk for ASCVD, warranting consideration of strategies that reduce this risk. Approximately 50% of adults aged 64–75 and >75 years fall in the category of those with good to intermediate health for whom targeted risk-factor management is reasonable. The remaining population requires careful consideration of any potential benefit or risk with therapies targeting specific numbers for lipid, BP, or glycemic parameters. There is a need for additional research that addresses the risk-to-benefit ratio of strategies among older adults according to aggregate health status. There is growing appreciation of the contribution of hypoglycemia to ASCVD risk, warranting caution with use of IP therapy in this vulnerable population, particularly in those with compromised renal function. The approaches suggested by the ADA and AGS provide a framework for approaching risk-reduction strategies in this age-group; however, there is a need for further investigation into this growing population of older adults to guide clinical practice.