Lifestyle Intervention

The basis for risk reduction in patients with diabetes, as in nondiabetic subjects, is lifestyle intervention. Lifestyle intervention has been shown to prevent the development of CVD in primary prevention, but the benefit of lifestyle intervention including diet, physical activity, and weight loss is less well established in patients with existing chronic CAD. However, general aspects are covered by various guidelines such as the American Heart Association (AHA), American Diabetes Association, and European Society of Cardiology (ESC)/European Association for the Study of Diabetes (EASD) : these include recommendations for smoking cessation, ideally guided by structured advice or a specially developed program, as well as a Mediterranean diet with fruit, vegetables, and olive oil. To what extent weight loss reduces CV risk in patients with existing CAD has not been established. Due to the lack of evidence, current guidelines do not recommend supplementation with vitamins or micronutrients to reduce CV risk in this population. With respect to physical activity, moderate to vigorous physical activity, at least 150 min/week, is recommended to prevent vascular disease in patients that can exercise.

Glucose Control

In patients with type 2 diabetes mellitus intensive glucose control can reduce microvascular complications such as retinopathy or nephropathy. The effect on macrovascular events in patients with diabetes and chronic CAD is less well established. The United Kingdom Prospective Diabetes Study (UKPDS) was the first large study examining the effect of an intensive glucose control regimen on macrovascular events; the study compared conventional versus intensive therapy in 3867 patients with newly diagnosed type 2 diabetes and no history of CVD. Intensive therapy significantly reduced microvascular events such as nephropathy and retinopathy, but after a follow-up of 10 years only a nonsignificant reduction in macrovascular events such as MI was found. Only after an additional 10 years of follow-up did the initial intensive therapy translate into a significant decrease in macrovascular events. These results, albeit difficult to interpret because of the nature of this nonprespecified follow-up analysis, suggested that early intervention with a stringent glucose control strategy may eventually reduce macrovascular events in patients with diabetes without a history of CV disease.

Over the last decade various CV outcome trials in high-risk patients with diabetes have assessed the effect of a tight glucose control strategy compared with standard therapy on the incidence of CV events. The Action to Control Cardiovascular Risk in Diabetes (ACCORD) Trial examined whether an intensive glucose control with the HbA _1c target of less than 6.5% (46 mmol/mol) compared with standard therapy with an HbA _1c target of less than 7.5% (58 mmol/mol) reduced CV events in 10,251 patients with type 2 diabetes. A very aggressive glucose-lowering approach with various combination therapies including insulin and up to 5 oral antidiabetic drugs was chosen to bring the HbA _1c value to target. After 3.5 years the study was stopped prematurely due to a higher mortality in the intensive treatment arm. The primary endpoint of MI, stroke, and CV death was not significantly different between groups, despite a significant difference in HbA _1c of 7.5% in the standard therapy and 6.4% in the intensive glucose-lowering group. Increased mortality associated with intensive therapy was mainly observed in subjects with multiple CV risk factors, as well as in those subjects in whom HbA _1c lowering was very difficult.

The Action in Diabetes and Vascular Disease: PreterAx and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial included 11,140 patients and tested whether intensive glucose-lowering therapy with an HbA _1c target below 6.5% compared with standard therapy with an HbA _1c target according to local guidelines might reduce the primary combined endpoint of macrovascular (MI, stroke, or CV death) and microvascular (nephropathy or retinopathy) events. The therapeutic algorithm in this trial to reduce HbA _1c levels was less aggressive than in ACCORD, and after a follow-up of 4.3 years the two groups significantly differed with an HbA _1c of 7.3% versus 6.5%. This HbA _1c difference translated into significant 10% RR reduction for microvascular events ( p =0.013) but did not have a significant effect on the combined macrovascular endpoint. In contrast to ACCORD, there was no increase in mortality in this study.

The third trial, the Veterans Affairs Diabetes Trial (VADT), was a smaller trial randomizing 1791 patients with type 2 diabetes to intensive or standard glucose therapy with an HbA _1c target of 6.0% in the intensive group and 9.0% in the standard group. Despite a highly significant difference in glucose control with an HbA _1c of 8.5% in the standard group and 7.0% in the intensive group, no reduction of the combined primary endpoint of MI, stroke, CV death, CHD intervention, or amputation was achieved. Patients in these three studies had a long duration of diabetes and a large proportion had preexisting CVD and a high number of associated risk factors such as hypertension or dyslipidemia. A meta-analysis of ACCORD, ADVANCE, and VADT suggested that an HbA _1c reduction of 1% may lead to a 15% RR reduction in nonfatal MI but no benefit on stroke and all-cause mortality. Further analyses suggested that patients with a short duration of diabetes, no history of CVD, and low HbA _1c at baseline may still benefit from an intensive glucose-lowering therapy. However, lower HbA _1c targets should only be achieved without increasing the risk for hypoglycemia; in addition, weight gain and uncontrolled combination therapy of oral antidiabetic medication and/or insulin should be avoided.

Hemoglobin A _1c Targets

Current guidelines from various diabetes and heart professional associations favor an individualized strategy for HbA _1c target based on age, history, duration of diabetes, and presence of CVD as well as other comorbidities and the risk of hypoglycemia. In general, a near-normal HbA _1c level below 7% (53 mmol/L) should be achieved to decrease microvascular complications. A tighter blood glucose control with an HbA _1c target less than 6.5 might be appropriate in selected subjects with a short duration of diabetes and a low risk of hypoglycemia. For older patients with diabetes as well as those with preexisting CV disease, less stringent HbA _1c lowering to a target less than or equal to 8% is recommended.

Glucose-Lowering Agents

Most guidelines recommend metformin as the first-line therapy for glucose-lowering because of its weight-loss effect and low risk of hypoglycemia. In addition, data from the UKDPS suggested a beneficial effect on CV outcome: in the subgroup of 753 overweight patients metformin significantly reduced the risk of MI versus conventional therapy by 39%. Such data were confirmed in two meta-analyses suggesting reduced CVD in patients treated with metformin. The majority of patients with type 2 diabetes require combination therapy to achieve glycemic targets. Metformin can be combined with any other antidiabetic drug including sulfonylureas (SUs), α-glucosidase inhibitors, pioglitazone, glucagon-like peptide 1 (GLP1)-receptor agonists, dipeptidyl peptidase 4 (DPP4)-inhibitors, sodium-glucose cotransporter-2 (SGLT2)-inhibitors, and insulin. Of note, any of these agents can be used as monotherapy in subjects in whom metformin is contraindicated or not tolerated.

The PROspective pioglitAzone Clinical Trial In macroVascular Events (PROACTIVE) analyzed whether addition of pioglitazone or placebo to baseline antihyperglycemic therapy has an effect on CV events. It showed no benefit on the combined primary endpoint of all-cause mortality, nonfatal MI, acute coronary syndrome (ACS), coronary artery bypass graft (CABG), percutaneous coronary intervention (PCI), stroke, major leg amputation, or major leg revascularization. However, because this endpoint included non-CV composites such as reduction of leg amputation or revascularization—events that are unlikely to be reduced by medical therapy alone—a principal secondary endpoint was predefined. Pioglitazone significantly reduced this secondary outcome of MI, stroke, and CV mortality (HR 0.84; 95% CI, 0.72–0.98; p = 0.027) versus placebo. For another thiazolidinedione, rosiglitazone, no such effects have been observed. However, PROACTIVE showed an increase in heart failure, a class effect of these insulin-sensitizing agents. The Study to Prevent Non-Insulin Dependent Diabetes Mellitus (STOP-NIDDM) showed a 49% RR reduction of CV events by acarbose versus placebo in patients with impaired glucose tolerance. Still, this was not the primary endpoint of this study and the study population did not have manifest diabetes. Therefore, the effect of acarbose on CV events is currently being tested in a large CV outcome trial in China, the Acarbose Cardiovascular Evaluation (ACE), enrolling patients with established type 2 diabetes.

Conflicting data exist with respect to the effects of SUs on CV events. The University Group Diabetes Program (UGDP) was the first study conducted in the 1960s that raised concerns about the safety of the first-generation SU tolbutamide. It showed a significant increase of overall and CV mortality in subjects receiving tolbutamide versus placebo. Still, this study was not designed or powered to test CV safety, and it has been criticized because the results were not corrected for higher preexisting CV risk in the tolbutamide group versus the placebo group. In addition, it is unclear to what extent the findings of this study can be applied to current clinical practice, given the fact that modern diabetes management including a multifactorial approach was not applied. It is also unclear whether these findings apply to modern SUs. In contrast to the findings in UGDP, UKPDS demonstrated that tolbutamide, glyburide, and glimepiride were not associated with adverse CV events. Other trials of longer-term duration also indicated that SUs are not associated with an increased CV risk when compared head-to-head with other agents, such as thiazolidinediones, DPP4-inhibitors, metformin, or GLP1-analogs. In addition, a large meta-analysis of 40 randomized controlled trials of glucose-lowering drugs found no increased risk of macrovascular events and all-cause mortality in second-generation SUs versus other oral agents or placebo. However, most of the trials included in this meta-analysis were not designed or powered to examine CV events. Moreover, the inconsistent reporting of adverse events and the short-term duration of these studies make it difficult to make final conclusions on the effect of SUs on CV events. Interestingly several observational studies have shown higher rates of all-cause and CV mortality associated with SU monotherapy or in combination with metformin compared with metformin monotherapy, but this was not confirmed in other studies. Overall, there is an absence of conclusive outcome data on the impact of SUs on CV events. The ongoing Cardiovascular Outcome Trial of Linagliptin Versus Glimepiride in Type 2 Diabetes (CAROLINA) trial may shed more light on this issue.

Newer Treatment Options

Over the last years, multiple novel antidiabetic therapies have come to the market and the Food and Drug Administration (FDA) and Europe, the Middle East, and Africa (EMEA) requirements have made it mandatory for the industry to perform CV outcome trials to show safety. The FDA required a demonstration of noninferiority of these agents versus placebo with regard to CV events, utilizing a noninferiority margin of 1.3. This has led to carrying out and publishing results of large CV outcome trials in patients with type 2 diabetes and high CV risk. So far, three large CV outcome trials with DPP4-inhibitors, three large trials with GLP1-receptor agonist, and the first outcome trial for an SGLT-2-inhibitor have been published. The three DPP4-inhibitor trials, SAVOUR (saxagliptin), EXAMINE (alogliptin), and TECOS (sitagliptin), examined in a high-risk population of patients with a long duration of diabetes, prior CVD, and/or various risk factors whether the addition of the given drug increases CV risk versus placebo ( Table 24.1 ).

These trials were designed as noninferiority trials and did not show an increased CV risk of any of these DPP4-inhibitors. Of note, they were designed to achieve glycemic equipoise between groups, not to examine whether a difference in HbA _1c levels in the two treatment arms translates into a reduction of CV events. Interestingly, SAVOUR-TIMI showed a significant increase in hospitalization for heart failure in patients treated with saxagliptin versus placebo, whereas such a significant signal was not found in the two other trials of DPP4-inhibitors. Three similar trials were performed with GLP-1 receptor agonists. The ELIXA trial confirmed CV safety of lixisenatide versus placebo without showing a portenial benefit with respect to CV events.

In contrast, the LEADER cardiovascular outcome trial testing the effect of the long acting GLP1 receptor agonist Liraglutide showed a significant reduction of the primary endpoint of cardiovascular death, myocardial infarction and stroke and the results were mainly driven by a significant reduction of cardiovascular death. In addition Liraglutide reduced overall mortality in a population of 9340 patients with diabetes and high cardiocardiovascular risk.

Most recently SUSTAIN 6 was reported. This study examined once weekly Semaglutide in 3297 patients with type 2 diabetes and high cardiovascular risk. Compared to placebo Semaglutide significantly reduced the combined cardiovascular endpoint of cardiovascular death, non-fatal myocardial infarction and non-fatal stroke. Interestingly this result was mainly driven by a significant 39% reduction of non-fatal stroke. The trend for myocardial infarction was statistically not significant NEJM 2016 inline). A similar trial with the GLP-1 receptor agonist lixisenatide versus placebo confirmed this drug’s safety, without showing a potential benefit with respect to CV events.

Cardiovascular Risks Associated with Hypoglycemia

Both insulin and SUs can lead to hypoglycemia in patients with diabetes. Severe hypoglycemia is defined as an event requiring external assistance for recovery, whereas milder episodes may be treated by the patient alone. Clinical trials in patients with type 2 diabetes raised concerns about an increased risk for CV events after hypoglycemic events. In the previously mentioned trials, ACCORD, ADVANCE, and VADT, the rates of severe hypoglycemia were substantially higher in patients with intensive versus standard therapy. In ACCORD severe hypoglycemic events occurred in 16.2% versus 5.1%, in ADVANCE 2.7% versus 1.5%, and in VADT 21.2% versus 9.7% in the intensive glucose target groups versus control groups. After publication of these data, intensive discussions have taken place to what extent severe hypoglycemic events contribute to CV events and excess mortality. To date it is recognized that hypoglycemia is a serious and common complication of diabetes treatment and is associated with CV events and mortality. Several mechanisms such as cardiac arrhythmias due to abnormal cardiac repolarization in high-risk patients, for example, those with CAD or cardiac autonomic neuropathy, increased thrombotic tendency/decreased thrombolysis, CV changes induced by catecholamines, and silent myocardial ischemia, have been discussed to link hypoglycemia with CV events. Although clear causality is as yet unproven, the avoidance of hypoglycemia is one of the key goals in diabetes therapy. Because a direct causal link with death or CV events has not been shown so far, hypoglycemia may serve as a marker of a patient’s overall vulnerability to adverse clinical outcomes. Therefore, patients treated with SUs or insulin should be carefully monitored with respect to hypoglycemic events and, whenever possible, other agents that do not cause hypoglycemia should be used. Table 24.3 summarizes therapeutic pharmacologic options to treat type 2 diabetes.

Lipid Lowering

Low-Density Lipoprotein Cholesterol Lowering

Studies over the last three decades have revealed that LDL-C lowering is among the most potent strategies to reduce CV events in patients with diabetes. However, to date there is controversy about the strategy: the American guidelines are in favor of a “fire-and-forget” approach, whereas the European guidelines propose a “treat-to-target” concept. In patients with diabetes, the American guidelines distinguish two groups: those between 40 and 75 years of age and a high CV risk (10-year risk assessed by new pooled cohort equations > 7.5%) and a moderate-risk group with a 10-year risk less than 7.5%. In patients with diabetes in the high-risk group a high-intensity LDL-C reduction of at least 50% with a potent statin such as high-dose atorvastatin (80 mg ) or rosuvastatin (20 to 40 mg) is recommended, whereas subjects in the moderate-risk group should receive a less-intensive statin therapy to achieve an LDL-C reduction between 30% and 50%. This less-intensive daily therapy includes atorvastatin 10 to 20 mg, rosuvastatin 5 to 10 mg, simvastatin 20 to 40 mg, pravastatin 40 to 80 mg, pitavastatin 2 to 4 mg, or lovastatin 40 mg. In the US guidelines no target LDL-C levels are recommended. The American guidelines are based on statin RCTs only, whereas the European guidelines included RCTs, population epidemiology, and genetic epidemiology as a basis for their recommendations. Therefore the European guidelines recommend that patients are categorized based on their CV risk, and, depending on their individual risk, LDL-C target values are recommended. All patients with diabetes belong to the very-high-risk group, and ESC guidelines recommend an LDL-C target below 70 mg/dL (< 1.8 mmol/L) in these subjects.

Since publication of these guidelines, another large lipid-lowering CV outcome trial, Improved Reduction of Outcomes: Vytorin Efficacy International Trial (IMPROVE-IT), has presented results. IMPROVE-IT examined whether ezetimibe, an inhibitor of the cholesterol transporter NPC1L1, which reduces intestinal cholesterol absorption, added to simvastatin versus simvastatin alone may affect CV event incidence in a population of 18,144 post-ACS patients with LDL-C levels above target. After a mean follow-up of 5.7 years, the addition of ezetimibe to simvastatin led to a reduction of LDL-C to 53.7 mg/dL (1.4 mmol/L) versus 69.5 mg/dL (1.8 mmol/L) in the simvastatin-alone group. This LDL-C reduction translated into a significant 6.7% RR reduction for the primary combined endpoint of CV death, MI, stroke, hospitalization for unstable angina, or revascularization with a number-needed-to-treat of 52 to prevent 1 event. Further subgroup analyses of IMPROVE-IT showed that the significant benefit in the overall population is mainly driven by a highly significant effect in patients with diabetes (European Society of Cardiology 2015 Congress. Presented August 30, 2015). Abstract 1947). These data challenge current lipid-lowering guidelines: first of all, IMPROVE-IT shows that a nonstatin lipid-lowering strategy can reduce CV events in high-risk patients. In addition, it demonstrates that further lowering of LDL-C to levels below the currently recommended targets translates into a further reduction of CV events, raising the hypothesis that “the lower, the better” strategy may apply for LDL-C reduction. In summary, lowering LDL-C is a very potent strategy to reduce CV risk in patients with diabetes and CAD independent of a fire-and-forget or treat-to-target strategy. The ESC/EASD guidelines recommend LDL-C target values less than 70 mg/dL in patients with diabetes mellitus and CAD.

Novel strategies, such as inhibition of PCSK9 with antibodies such as alirocumab or evolocumab (both approved in Europe and the United States), have shown promising results in various patient populations including those with diabetes. In subjects with familial hypercholesterolemia, as well as in high-CV-risk patients with LDL-C levels not at target with currently available lipid-lowering drugs, these antibodies have been shown to exhibit a very potent LDL-C-lowering effect. In patients with diabetes, subgroup analyses of phase III trials showed that PCSK9 inhibition is as effective as in nondiabetic subjects. Two large outcome trials, ODYSSEY Outcome and FOURIER, are examining whether the reduction of LDL-C by PCSK9 inhibition with alirocumab or evolocumab translates into a reduction of CV events. Both trials enrolled a large proportion of patients with diabetes, and the results will provide further insights into the effect of intensive LDL-C lowering in the high-risk population of patients with diabetes and CVD.

Various clinical trials have shown that statin use may increase the risk of incident diabetes in patients without diabetes. A large meta-analysis suggested that statin therapy is associated with a slightly increased RR (9%) of development of diabetes ; still the risk is low in absolute terms and does not outweigh the benefit with respect to the reduction in coronary events.

Therapies to Increase High-Density Lipoprotein Cholesterol in Patients with Type 2 Diabetes

Current guidelines do not recommend HDL-C-raising strategies in the lipid management of patients with diabetes. Over the last few years various approaches have been developed to increase HDL-C but no trial showed a significant reduction in CV events. The Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides: Impact on Global Health Outcomes (AIM-HIGH) trial compared niacin versus placebo in statin-treated patients with known CV disease. It had a large subgroup of patients with diabetes (34%). After 2 years niacin raised HDL-C from 35 mg/dL to 42 mg/dL, lowered triglycerides from 164 mg/dL to 122 mg/dL, and decreased LDL-C from 74 mg/dL to 62 mg/dL. However, AIM-HIGH was terminated after a mean follow-up of 3 years because of futility. The primary endpoint of CV events or hospitalization for unstable angina did not differ among groups. In the subgroup of patients with diabetes the data were similar with no significant effect on CV outcome in niacin-treated patients. In addition, the Treatment of HDL to Reduce the Incidence of Vascular Events (HPS2-THRIVE) trial confirmed these data by showing that the addition of extended-release niacin/laropiprant to simvastatin (or ezetimibe/simvastatin) did not reduce the risk of CV events.

Inhibition of cholesteryl ester transfer protein (CETP) was another strategy to increase HDL levels. Two large outcome trials (with the CETP inhibitors torcetrapib and dalcetrapib) did not show a reduction of CV events despite a 30% to 40% increase in HDL-C. Clinical and experimental data suggest that this may be due to abnormal functional characteristics of HDL particles in patients with diabetes and/or CVD, suggesting that HDL function may be more important than the overall number of measured HDL particles. Thus, current recommendations suggest that only lifestyle modification is indicated to address low HDL in patients with diabetes.

Blood Pressure Lowering

Hypertension is one of the CV risk factors associated with diabetes, and in patients with type 2 diabetes more than 60% have arterial hypertension. Various pathophysiologic mechanisms such as increased renal sodium reabsorption due to hyperinsulinemia, increased sympathetic activity, and RAAS activation are thought to contribute to arterial hypertension in these patients. Both hypertension and diabetes are additive risk factors for CVD, and various data suggest that the presence of hypertension in subjects with diabetes leads to a 4-fold increase in CV risk. Over the last decades blood pressure targets in patients with diabetes have changed and are still a matter of debate. The first trial to suggest stringent blood pressure lowering in patients with diabetes was the High-Potential Optimal Treatment (HOT) Trial. It showed that diastolic targets below 80 mm Hg significantly decreased CV risk versus a less stringent strategy with diastolic targets less than 100 or less than 90 mm Hg. Still, the mean diastolic blood pressure in the first group was still above 80 mm Hg and the mean systolic blood pressure was 144 mm Hg.

Data from UKPDS published in 1998 showed that reduction of mean blood pressure from 154/87 mm Hg to 144/82 mm Hg led to a 24% reduction in CV events. In addition, post-hoc UKPDS data suggest that a blood pressure drop of 10 mm Hg decreases diabetes-related mortality by 15%. The lowest systolic blood pressure achieved in this context was approximately 120 mm Hg. The results from these studies suggested that there may be no threshold for the beneficial effect of BP lowering. These data were in contrast to those from the ACCORD trial in which 2700 patients with type 2 diabetes were randomized to intensive (mean systolic blood pressure at study end: 119 mm Hg) versus standard therapy (mean systolic blood pressure at study end: 134 mm Hg) over a mean follow-up of 4.7 years. There was no significant effect of the intensive therapy on the primary composite endpoint of nonfatal MI, nonfatal stroke, or CV death. Still, the incidence of fatal or nonfatal strokes was significantly reduced in the intensive therapy group, translating into a number-needed-to-treat for over 5 years of 98 to prevent 1 stroke event. However, the intensive blood pressure-lowering approach significantly increased serious adverse events from 8.3% versus 3.3% with a significant increase in hypotension, syncope, arrhythmia, and hyperkalemia, as well as renal failure and a decrease of glomerular filtration rate below 30 mL/min per 1.73 m ² . These data from ACCORD showing an increase in serious adverse events do not support a reduction of systolic blood pressure below 130 mm Hg.

A 2011 meta-analysis including 13 RCTs and 37,736 patients with diabetes, impaired fasting glucose, or impaired glucose tolerance suggests that intensive blood pressure control (with a systolic blood pressure ≤135 mm Hg in this group) compared with a standard group (systolic blood pressure ≤140 mm Hg) leads to a 10% relative reduction in overall mortality and a 17% reduction of stroke incidence. However, this meta-analysis confirmed the ACCORD observation that more intensive blood pressure control leads to an up to 20% increase in serious adverse events. These trials and analyses set the basis for the ESC guideline recommendation that patients with diabetes should achieve a blood pressure target of less than 140/85 mm Hg. In certain subgroups, including those with nephropathy and overt proteinuria, further reduction of systolic blood pressure to targets below 130 mm Hg may be considered, but the evidence to support this recommendation is scarce. In addition, the risk/benefit balance of intensive blood pressure management needs to be carefully considered individually with special attention in elderly patients and those with a long duration of diabetes.

The management of blood pressure in patients with diabetes, as in nondiabetic subjects, includes lifestyle intervention with salt restriction and weight loss, as well as pharmacologic treatment. Lifestyle intervention is recommended for all patients with hypertension but it is often insufficient to adequately control blood pressure, making pharmacologic intervention necessary.

Pharmacologic Intervention to Lower Blood Pressure in Patients with Diabetes

In principle, all blood pressure–lowering agents can be used to treat patients with diabetes to a blood pressure target of less than 140/85 mm Hg. However, several RCTs enrolled large subgroups of patients with diabetes and demonstrated that blockade of the RAAS by angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers provides the largest benefit in the reduction of CV events in these high-risk patients—in particular in patients with overt proteinuria. Good evidence exists that for initial therapy, ACE inhibitors rather than calcium-channel blockers should be used to prevent or retard the occurrence of microalbuminuria in these patients. A combination of ACE inhibitors and angiotensin receptor blockers did not show any CV benefit versus ACE inhibition alone in the ONTARGET trials and was even associated with more adverse events, suggesting that this combination therapy should not be used for blood pressure lowering. The Aliskiren Trial in Type-2 Diabetes Using Cardio-renal Endpoints (ALTITUDE), examining the addition of the renin inhibitor aliskiren to RAAS blockade in patients with diabetes at high CV and renal events risk, did not show a reduction in CV events but an increase in adverse events, suggesting that this combination therapy should also be avoided.

Other important points taken into consideration in antihypertensive therapy in patients with diabetes are the metabolic effects of various blood pressure–lowering agents. As such, thiazides and β-blockers are associated with an increased risk for the development of type 2 diabetes compared with RAAS inhibitors or calcium-channel blockers. However, it is not known whether β-blockers and/or thiazides have similar effects in patients with prevalent type 2 diabetes, and the clinical importance of these adverse metabolic effects remains undetermined. Based on the unfavorable metabolic effects of diuretics and β-blockers, these agents should be avoided as first-line therapy in subjects with metabolic syndromes or high risk for diabetes. Despite this, in patients with established diabetes the beneficial effect of blood pressure lowering seems to outweigh the potential negative metabolic effects by far and therefore diuretics and β-blockers should be used as combination therapy once RAAS inhibition is established. The Avoiding Cardiovascular Events through Complication Therapy in Patients Living with Systolic Hypertension (ACCOMPLISH) trial compared addition of the calcium-channel blocker amlodipine versus hydrochlorothiazide on top of an ACE inhibitor. The study had 11,506 patients, including 6946 patients with diabetes. In the diabetic subgroup, there was a significant reduction of the primary endpoint of CV death and CV events in the amlodipine arm. These data suggest that once RAAS inhibition has been established, the second-line drug should be a calcium-channel blocker given the neutral metabolic effects and favorable results seen in ACCOMPLISH. Overall, current blood pressure targets are only achieved in a subset of patients with diabetes, underscoring the necessity to improve blood pressure lowering therapies in them.

Antiplatelet Therapy in Patients with Diabetes

In patients with diabetes platelet function is disturbed leading to a more frequent response to subthreshold stimuli, increased platelet turnover, and accelerated thrombopoiesis of hyperreactive platelets. Various factors such as hyperglycemia itself with glycation of platelet membrane proteins, oxidative stress with increased reactive oxygen species production, formation of advance glycation end products, and endothelial dysfunction with the release of mediators that affect platelet activity may be crucial in this context. Conflicting data exist for the benefit of aspirin therapy in primary prevention in patients with diabetes: various CV outcome trials as well as large meta-analyses suggest a limited net benefit of aspirin in primary prevention when assessing reduced CV events compared with increased bleeding risk. Therefore, current guidelines do not uniformly recommend low-dose aspirin in primary prevention. The most recent ACC/AHA guideline in 2015 states that low-dose aspirin (75 to 162 mg/day) is reasonable among those with a 10-year CV risk of at least 10% without an increase of bleeding (class II b, level of evidence B) and that low-dose aspirin is reasonable in adults with diabetes mellitus at intermediate risk (10-year CV risk 5% to 10% [ACC/AHA class II b level of evidence C]). In contrast, the ESC and EASD guidelines in 2013 state that antiplatelet therapy with aspirin in patients with diabetes and low risk is not recommended (class III) and antiplatelet therapy for primary prevention may be considered in high-risk patients with diabetes on an individual basis with a class II b level C recommendation.

In secondary prevention it is recommended that patients with diabetes receive low-dose aspirin similar to what is recommended in patients without diabetes. The evidence for this statement stems from the Anti-platelet Trialists’ Collaboration (ACT) showing that aspirin leads to a clear CV benefit in patients with preexisting systemic CVD, both in the presence or absence of diabetes. This analysis included more than 4000 patients with diabetes in RCTs and showed that aspirin clearly reduced CV events (nonfatal MI, nonfatal stroke, and CV death) in them. In case of aspirin intolerance clopidogrel is recommended as an alternative antiplatelet therapy. In the setting of stable CAD current data do not support the use of novel antiplatelet agents such as prasugrel or ticagrelor.

In the ACT trial of 1000 patients with diabetes, aspirin reduced 42 vascular events in secondary prevention. In addition, interesting analyses of the ACT trial suggest that low-dose aspirin (75 to 150 mg/day) seems to be as effective as higher doses (150 mg/day). Moreover, low-dose aspirin was associated with a lower risk of bleeding complications than the higher dose. These data were supported by an observational analysis from the Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilisation, Management Avoidance (CHARISMA) trial demonstrating that aspirin doses above 100 mg/day are not associated with increased efficacy compared with a lower dose. In addition, CHARISMA suggests an increased, albeit not significant, risk of CV death, MI, and stroke (adjusted HR 1.16, 95% CI 0.93–1.14) as well as an increased risk of severe or life-threatening bleeding (adjusted HR 1.3, 95% CI 0.83–2.04) when aspirin doses above 100 mg/day were combined with clopidogrel. These data were supported by a 2010 trial examining the optimal aspirin dose also showing that a higher aspirin dose is not effective in reducing CV events but may increase the risk of bleeding. Table 24.4 summarizes the current recommendation of the ESC/EASD on risk factor management in diabetes.

Coronary Revascularization in Patients with Diabetes

The concurrent presence of diabetes in patients with multivessel CAD bears an enhanced risk of in-stent restenosis after PCI and subsequently may result in a worse prognosis following CABG in contrast to patients without diabetes. This refers largely to the aggressive nature of the disease with smaller coronary arteries affected in a diffuse manner, and consequently CABG has been the preferred treatment strategy in patients with diabetes requiring coronary revascularization. Numerous trials have demonstrated an efficacy benefit (less repeat revascularization) and moreover a safety benefit (lower mortality) with CABG than with PCI in patients with diabetes. The advantages of CABG in patients with diabetes reflect reductions in CV events caused by both nonculprit and culprit lesions. Treatment with PCI usually focuses mainly on the coronary culprit lesion, whereas angiographically and/or functionally nonsignificant nonculprit lesions are generally not treated. However, data from the PROGRESS trial have indicated that particularly those lesions that appear rather mild angiographically have a tendency to rupture in the future due to enhanced plaque vulnerability and plaque burden. In contrast, CABG offers complete treatment of culprit and nonculprit lesions throughout the bypassed segments. Thus, CABG provides an effective protection against secondary CV events due to rupture of initially nonflow-limiting but unstable plaques and thereby avoids the occurrence of concomitant CV events such as MI and sudden cardiac death.

Percutaneous Coronary Intervention with Balloon Angioplasty Versus Coronary Artery Bypass Grafting in Patients with Diabetes

In a subgroup analysis of the Early Bypass Angioplasty Revascularization Investigation (BARI) including 353 patients with diabetes mellitus, the revascularization strategies of CABG versus PCI with plain old balloon angioplasty (POBA) were compared. There was a survival benefit for CABG over PCI (80.6% vs 65.5%, p =0.003). A more recent meta-analysis comprising 68 RCTs and a total of 24,015 patients with diabetes compared CABG with different modes of PCI (POBA, bare metal stent [BMS], first- and second-generation drug-eluting stents [DESs]). In the overall study cohort CABG was associated with numerically lower rates of both death (RR 0.8, 95% CI 0.55–1.23) and MI (RR 0.86, 95% CI 0.28–2.86); however, the difference was not statistically significant.

Percutaneous Coronary Intervention Using Bare Metal Stent or First-Generation Drug-Eluting Stent Versus Coronary Artery Bypass Graft in Patients with Diabetes

In 2005 the Arterial Revascularizaton Therapies Study (ARTS) compared PCI using BMS with CABG. However, it did not primarily focus on patients with diabetes, and only a small diabetic subcohort of 208 patients was available for evaluation. In this subcohort PCI using BMS compared with CABG was found to have numerically, but not statistically significant, higher rates of overall mortality (13.4% vs 8.3%, p =0.27) and MI (10.7% vs 7.3%, p =0.47) at 5-year follow-up. However, rates for repeat revascularization (42.9% vs 10.4%, p <0.001) and MACCE (54.5% vs 25%, p < 0.001) were significantly higher in the PCI group than in the CABG group.

As previously mentioned the BARI-2D trial addressed the question of revascularization in a diabetic patient cohort comparing OMT alone versus OMT plus revascularization therapy (either CABG or PCI). Within the PCI group the stent types were 56% BMS and 35% DES. Despite indicating a treatment benefit in favor of CABG over PCI, this study had a major limitation in that it did not directly compare CABG and PCI. This has subsequently led to several further RCTs comparing CABG with PCI with the use of first-generation DES. In the Coronary Artery Revascularization in Diabetes (CARDia) trial, 510 patients with diabetes were enrolled and randomized to either PCI or CABG. The results did not show any differences regarding the primary composite endpoint of death, MI between the two strategies (PCI 13.0% vs CABG 10.5%, p =0.39). However, when adding repeat revascularization to the composite endpoint, there was a relevant benefit favoring CABG (11.3% vs 19.3% with PCI, p =0.016) at 1-year follow-up. A relevant limitation of the CARDia trial was the mixed use of BMS (31%) and first-generation sirolimus-eluting stent (SES) within the PCI arm.

In the Veterans Affairs Coronary Artery Revascularization in Diabetes Study (VA CARDS), a small study enrolling 198 patients, most patients received a first-generation DES (60% SES or paclitaxel-eluting stent [PES]) while approximately 20% received a second-generation cobalt-chromium everolimus-eluting stent (CoCr-EES). Thus, VA CARDS was considered as a trial mainly comparing CABG with first-generation DES. It showed a significant reduction of overall mortality from 21% with PCI to 5% with CABG at 2 years of follow-up.

A subcohort of the Synergy Between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery (SYNTAX) trial, consisting of 452 patients with diabetes with left main or 3-vessel disease, similarly demonstrated higher rates of MACCEs with PCI using PES compared with CABG at 1 year (26% vs 14.2%, p =0.003) and after 5 years of follow-up (46.5% vs 29.6%, p <0.001). These findings were predominantly related to a higher rate of repeat revascularization within the PCI group at 1 year (PCI 20.3% vs CABG 6.4%, p < 0.001) and 5 years (PCI 35.3% vs CABG 14.6%, p <0.001). Regarding the anatomic severity according to the SYNTAX score, a treatment benefit with CABG was only seen in those patients with diabetes and complex disease (SYNTAX score ≥33); whereas in intermediate and less complex CAD no differences in terms of the composite endpoint were demonstrated. However, as SYNTAX and ARTS were not performed specifically in a diabetic population, the limited number of patients in these subgroups limits conclusions that can be drawn. The ARTS trial only used a historic control group, the CARDia trial was stopped early due to low enrollment and was conclusively underpowered for true evaluation, and the diabetic subgroup analysis of the SYNTAX trial was initially not designed to test differences in mortality. Hence, these studies did not provide sufficient evidence to clearly endorse one of the two revascularization strategies. Moreover, apart from current clinical practice CARDia, VA CARDS, and the diabetic subcohort of SYNTAX did not use newer-generation DES. Instead, predominantly BMS or first-generation DES were implanted and compared with CABG.

As a consequence of these limitations, the FREEDOM study was designed. This prospective multicenter study has emerged as a true landmark trial and the only RCT that was adequately powered to compare PCI and CABG in an exclusive cohort limited to patients with diabetes and multivessel disease. However, as in previous trials, first-generation DES (SES: 51% and PES: 43%) and not second-generation stents were implanted, which is different than the current standard of care in interventional cardiology. In FREEDOM, which enrolled 1900 patients with diabetes, the primary composite endpoint (all-cause mortality, nonfatal MI, or nonfatal stroke) was lower in patients treated by CABG compared with PCI at 1-year (CABG 18.7% vs PCI 26.6%, p =0.005) and 5-year follow-up (CABG 11.8% vs PCI 16.8%, p =0.004). Of note, this was related to significant differences in overall mortality (CABG 10.9% vs PCI 16.3%, p =0.049) and rate of MI (CABG 6.0% vs PCI 13.9%, p <0.001) at 5 years. Moreover, the incidences of repeat revascularization at 1 year after initial revascularization were significantly higher in the PCI than the CABG group (12.6% vs 4.8%, p <0.01). However, the rate of strokes was conversely higher in the CABG group (5.2 vs 2.4%, p =0.03) and no difference concerning cardiac-specific mortality was found (DES 10.9% vs CABG 6.8%, p =0.12). FREEDOM was still limited by the relatively low inclusion rate of women (28.6%), patients with an ejection fraction below 40% (2.5%), and patients with less advanced CAD (35.5%), defined as a SYNTAX score less than 22.

Still, prior to SYNTAX and FREEDOM, lesion anatomy was hardly characterized and scores for lesion severity were not used at all. Thus, stratification of patient cohorts according to the predescribed SYNTAX score tertiles of coronary lesion severity strengthens the impact of both the SYNTAX diabetic substudy analysis as well as the FREEDOM trial. Unlike previous trials, the baseline characteristics in the FREEDOM cohort targeted a high-risk diabetic population with rather advanced and complex CAD, as characterized by an 83.3% rate of multivessel disease, an average number of lesions of 5.7 ± 2.2, an average lesion length of 77.6 mm ± 33.8 mm, and an average SYNTAX score of 26.2 ± 8.6. Unlike SYNTAX, FREEDOM documented good adherence to concomitant medication in both groups: specifically, in the PCI group approximately 90% of patients received dual antiplatelet therapy for at least 12 months. Furthermore, FREEDOM showed superiority of CABG over PCI in all categories of the SYNTAX score with no significant subgroup interaction ( p =0.58). Table 24.5 lists a survey of randomized trials on revascularization in patients with diabetes.

TABLE 24.5

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Epidemiology of Chronic Coronary Artery Disease
2. Goals of Therapy
3. Global Risk Assessment
4. CT and MRI

Stay updated, free articles. Join our Telegram channel

Join