Cardiologists in general are strong advocates of evidence-based medicine, due to the critical nature of ischemic heart disease in patients. The use of up-to-date scientific evidence from research is paramount as the basis for making many medical decisions. There are three major advantages to using this approach. First, it offers some of the most objective ways to determine and maintain consistent high-quality safety standards. Second, in many instances it can shorten the time from research to clinical practice and even change current national practice guidelines. Third, good practice of medicine does not mean more cost; in some cases it can reduce costs. There are several helpful evidence-based research documents available online; for example, the Cochrane systematic reviews, the UK National Institute for Health and Clinical Excellence guidelines, and many US guidelines from various organizations (such as the American College of Cardiology or the American Diabetes Association). Here, we discuss the most current evidence-based data on diabetes therapy and its cardiovascular effects.
Evidence Suggesting Elevated Blood Sugar Is Related to Increased Cardiovascular Events
Type 2 diabetes is a cardiovascular disease presenting as a metabolic disorder. Clearly, the morbidity and mortality related to type 2 diabetes are cardiovascular in nature. A significant threefold increase in cardiovascular events began to manifest approximately 15 years before the diagnosis of type 2 diabetes in the Nurses’ Health Study (see Figure 1) and is primarily responsible for the costs associated with treating the disease.1 Consequently, early, aggressive intervention to reduce global cardiovascular risk before the clinical diagnosis of diabetes is warranted.
Evidence-based data to support the association between fasting blood sugar/post-prandial hyperglycemia and risk of cardiovascular events primarily come from eight large epidemiological trials. Briefly, the Hoorn study was a population-based cohort study conducted from October 1989 to February 1992 among 1,342 non-diabetic white residents of Hoorn, the Netherlands, aged 50–75 years at baseline. Not only was it found that the incidence of diabetes was strongly related to both impaired fasting glucose and impaired glucose tolerance but, in a subsequent analysis, the two-hour post-load glucose was a better predictor of mortality than glycosylated hemoglobin (HbA1c).2 The Honolulu heart program reported that one-hour glucose was predictive of future coronary heart disease.3
Other studies that support a strong relationship between elevated glucose and increased cardiovascular disease are the Chicago heart study,4 the UK prospective diabetes study (UKPDS) 35,5 the Whitehall study,6 and the Helsinki study.6 The Diabetes epidemiology: collaborative analysis of diagnostic criteria in Europe (DECODE)7 study group assessed the risk of death by different glucose categories in 18,048 men and 7,316 women aged 30 years or older. They assessed glucose concentrations at fasting and two hours after a 75 g oral glucose tolerance test from 13 prospective European cohort studies. Using the WHO two-hour glucose criteria (≥11.1 mmol/l), they reported an increase in all-cause mortality in men and women with newly diagnosed diabetes and in those with previously diagnosed diabetes. They then showed that the hazard ratio for death in both men and women increased significantly with increasing two-hour glucose concentrations, regardless of fasting glucose concentrations, thus demonstrating that abnormalities in two-hour glucose concentrations are better predictors of mortality than fasting glucose when applied alone in screening. In summary, the link between elevated blood sugar and mortality is significant. However, treatments that merely reduce blood sugar levels without reducing CV events, or those that increase cardiovascular events, can have a significant unwanted off-target effect.
Large Type 2 Diabetes Prospective Randomized Trials with Cardiovascular Endpoints
Unfortunately, in the past few years, three large trials (Action in diabetes and vascular disease: preterax and diamicron MR controlled evaluation [ADVANCE],8 Action to control cardiovascular risk in diabetes [ACCORD],9 and Veterans Affairs diabetes trial [VADT])10 have reported negative results on macrovascular and microvascular events when using intensive glucose control targets. These three trials used intensive oral hypoglycemic and insulin therapies to achieve lower HbA1c values. All three of these trials failed to show a benefit on cardiovascular mortality with more intensive glucose control, and one trial (ACCORD) showed an increase in all-cause mortality. The lack of mortality benefit was due in part to severe recurrent hypoglycemia. In addition, insulin is well known to promote weight gain and worsen central obesity, and these effects add to the increased risk of cardiovascular disease.11 In the ADVANCE trial, the baseline HbA1c was 7.2 % and was reduced to 6.5 % in the intensive treatment arm (37 units insulin).
The weight gain was very little, at 0.2 lb, with only 2.7 % of patients having hypoglycemia events. In contrast, the baseline HbA1c in the ACCORD trial was 8.1 % and with insulin (35 units) it was reduced to 6.4 %. This was attended by a 7.7 lb weight gain and a 16.2 % increase in hypoglycemia events. Furthermore, in the VADT study, HbA1c decreased from 9.4 % to 6.5 % in the intensive treatment arm, with an average of 52 units of insulin given daily. The weight gain was tremendous at 17 lb and associated hypoglycemia was reported in 24 % of patients in the intensive treatment arm.
Cardiometabolic Drugs that Lower Glucose Levels
Cardiometabolic drugs that lower glucose (direct effect on insulin secretion or sensitivity) consist of insulin sensitizers, secretagogues, peptide analogs, and insulin. The most commonly used is metformin, which was found in the UKPDS 35 to reduce cardiovascular events in obese patients with moderate weight loss. This compound is a potent biguanide that primarily reduces hepatic glucose output and increases modest peripheral uptake of glucose in skeletal muscle. Metformin’s ability to produce weight loss instead of weight gain and its low risk of hypoglycemia place it as first-line treatment in most clinical practice guidelines. The usual expected reduction in HbA1c is generally between 1.5 and 2 %. Thiazolidinediones (TZDs), such as pioglitazone and rosiglitazone, are another class of insulin sensitizers which bind to the nuclear regulatory proteins (peroxisome proliferator-activated receptor-γ [PPAR-γ]) that regulate genes affecting glucose and fat metabolism. TZDs increase production of the messenger RNA of insulin-dependent enzymes (stimulation of insulin-sensitive genes) that increase the uptake of glucose primarily in skeletal muscle and fat cells. This class has achieved good HbA1c control in the long term compared with other oral hypoglycemics12 and has even been shown to prevent progression from impaired glucose tolerance to diabetes.13 In addition, some beneficial effects on the heart have commonly been reported. Pioglitazone has demonstrated an ability to halt the progression, or even reduce the burden, of atherosclerotic plaques as demonstrated by intravascular ultrasound.14,15 The Prospective pioglitazone clinical trial in macrovascular events (PROactive), though it failed to meet its primary endpoint, which included revascularization of leg arteries and lower extremity amputations, did find a significant reduction in cardiovascular events in the principal secondary endpoint of all-cause mortality, non-fatal myocardial infarction, and stroke with the use of pioglitazone.16
This study also demonstrated a decrease in overall incidence of cardiac events in patients with type 2 diabetes and prior myocardial infarctions. Despite these benefits, the use of this class of drugs remains controversial due to bone fractures in women, mild weight gain, fluid retention, heart failure, and now cancer concerns. Cardiovascular safety concerns relating to rosiglitazone arose from a retrospective meta-analysis published in the New England Journal of Medicine by Nissen et al., which showed a significant risk of myocardial infarction and a trend toward an increase in cardiovascular death.17 In closing, PPAR drugs are very complex, but have the potential to have beneficial effects beyond diabetes on the endovascular bed, which potentially are much more powerful than those of statins. The challenge will be finding the one that has a good safety profile with few off-target effects.
The second major class is secretagogues, which trigger endogenous insulin release by inhibiting the adenosine triphosphate-sensitive potassium (KATP) channels of the pancreatic β cells. The average HbA1c reduction varies between 1 and 2 %. This class of drugs can cause hypoglycemia, which is related to increased all-cause mortality. The original sulfonylureas adversely affected the heart by blocking myocardial preconditioning related to KATP channel blockade. At present, this class of compounds have no proven cardiovascular efficacy in randomized placebo-controlled double-blind trials.18
The third group of cardiometabolic drugs is peptide analogs that include glucagon-like peptide-1 (GLP-1) injectable agonists and glucose-dependent insulinotropic peptide (GIP). These compounds are insulin secretagogues that are both rapidly inactivated by the enzyme dipeptidyl peptidase-4 (DPP-4). The incretins are one of the most exciting groups of compounds from a cardiometabolic aspect because of recent human trials. Lonborg et al.19 studied 172 patients presenting with ST-segment elevation myocardial infarction (STEMI). Patients were randomly assigned to exenatide or placebo (saline) intravenously. The primary endpoint was salvage index calculated from myocardial area at risk (AAR), measured in the acute phase, and final infarct size measured 90 days after percutaneous coronary intervention by cardiac magnetic resonance (CMR). The authors found a 15 % larger salvage index in the exenatide group than in the placebo group (p<0.003) and the infarct size:AAR ratio was 23 % smaller in the exenatide group (p<0.003). One of the limitations was that 117 patients returned for follow-up CMR.
However, the implication for potential cardiovascular treatment from this pilot study is huge. Less than 10 % of these patients had diabetes. At a more basic level, GLP-1 compounds can improve endothelial function and have a preventive effect on endothelial dysfunction caused by ischemia-reperfusion injury via KATP channels.20 DPP-4 inhibitors increase circulating levels of GLP-1. In a study by DeFronzo et al.,21 injectable GLP-1 agonists produced three- or fourfold higher blood levels than DPP-4 inhibitors. DPP-4 inhibitors take advantage of the effects of GLP-1 by a mild increase in plasma levels of endogenous active (non-degraded) GLP-1. Both compounds (GLP-1 receptor agonists and DPP-4 inhibitors) have exceptional safety profiles with regard to hypoglycemia, with combined insulinotropic and glucagonostatic effects being exerted in a glucose-dependent fashion.
Non-glucose Lowering Medications and their Use in Diabetics
Recently, concern has arisen about high-dose angiotensin-converting enzyme inhibitors (ACEIs) in patients receiving DPP-4 inhibitors. Marney et al.22 reported the use of a DPP-4 inhibitor in metabolic syndrome patients testing the hemodynamic effects of a DPP-4 inhibitor in combination with an ACEI. In a well-designed cross-over study using sitagliptin and enalapril, 16 patients were evaluated. The design was a parallel-group cross-over study. Patients were randomized to receive placebo or sitagliptin (100 mg/day) for five days prior to each of two study days in a cross-over fashion. They were then randomized in parallel to receive an acute dose of placebo (group A, called enalapril 0 mg throughout the remainder of the article to avoid confusion with the placebo for sitagliptin), enalapril 5 mg (group B), or enalapril 10 mg (group C). The primary endpoint was blood pressure measured by automatic cuff. In addition, heart rate and norepinephrine levels were measured on each day. The results revealed important differences in patients on low- or high-dose ACEI. In patients receiving placebo (enalapril 0 mg) and low-dose ACE inhibition (enalapril 5 mg), sitagliptin was associated with lower blood pressure. However, higher-dose ACE inhibition (enalapril 10 mg) increased blood pressure and heart rate significantly in patients being treated with a DPP-4 inhibitor (see Figure 2). In addition, norepinephrine levels were significantly elevated in the high-dose ACEI group receiving a DPP-4 inhibitor.
This suggests that the antihypertensive effect of a high-dose ACEI is lost due in part to activation of the sympathetic nervous system in this study. Translating basic science to clinical practice is not perfect, and always needs to be looked at with larger double-blind randomized trials. This is one small look at an important group of high-risk patients who require very careful cardiovascular care.
The use of statins is key to reducing cardiovascular events; however, not all statins appear to have the same profile. One example is the recent publication by Koh et al., who studied the metabolic effects of rosuvastatin and pravastatin in patients with hypercholesterolemia.23 They completed a randomized single-blind placebo-controlled parallel study in 54 patients (each group) matched for age, gender, and body mass index. Over a two-month period patients received rosuvastatin 10 mg, pravastatin 40 mg, or placebo once daily. The authors evaluated adiponectin levels, insulin sensitivity by quantitative insulin sensitivity check index (QUICKI), insulin levels, HbA1c, and lipids. Rosuvastatin 10 mg/day significantly reduced lipids (total, low-density lipoprotein [LDL], and apolipoprotein B) and C-reactive protein (CRP) (p<0.05). However, there were increased fasting insulin levels (28 % increase, p<0.05) (see Figure 3) and a 1 % increase in HbA1c. Adiponectin levels were reduced with rosuvastatin by 9 % (p<0.01). Pravastatin reduced lipids and inflammation compared with placebo, but significantly reduced insulin levels and HbA1c and increased adiponectin. Both drugs improved endothelial function. The metabolic effects of statins differ significantly, and the concurrent treatment of diabetes in addition to hyperlipidemia may make a difference in statin choices. Clearly, more research is needed on this important topic.
Recent Meta-analysis with All-cause Mortality and Cardiovascular Endpoints
The core problem in type 2 diabetes is insulin resistance. Insulin resistance is associated with increased cardiovascular events, as demonstrated in many prior studies. At the heart of diabetes cardiovascular treatment is controlling global risks with proven drugs that reduce cardiovascular events. At the present time we have very limited data to show that current hypoglycemic drugs reduce cardiovascular events and, in many trials, they actually increased hypoglycemic events.
Additionally, Boussageon et al.24 completed a meta-analysis of randomized controlled trials involving 13 studies of 34,533 patients in total, with 18,315 receiving intensive glucose lowering treatment and 16,218 receiving standard treatment. During the five-year treatment period, 117–150 patients would need to be treated to avoid one myocardial infarction and 32–142 patients to avoid one episode of microalbuminuria. Off-target hypoglycemia occurred in every 15–52 patients treated. There was evidence of increased congestive heart failure with intensive treatment (p<0.001) without clinical risk reduction. In summary, this meta-analysis found a 19 % increase in all-cause mortality and a 43 % increase in death from cardiovascular causes. It did have inherent limitations, particularly publication bias as well as differing targeted glucose levels, outcome measurements, trial designs, and duration of follow-up between trials. Nonetheless, 13 randomized controlled trials found no benefit of intensive glucose control on all-cause mortality or death from cardiovascular causes in patients with type 2 diabetes. This underscores the continued importance of global risk reduction with proven agents rather than agents that simply lower blood glucose numbers.
While increased blood glucose is shown to lead to increased mortality, intensive glucose control with our current available treatments has not been shown to reduce cardiovascular events and may even worsen mortality. Even the use of non-glucose-lowering drugs such as ACEIs and statins in diabetics is not clear. Newer agents such as the GLP-1 agonist exenatide may have great potential in cardiac patients with type 2 diabetes, and potentially in non-diabetes patients, to reduce cardiovascular events. However, for now, more studies are needed on this matter. In the meantime, lifestyle and weight control remain the cornerstones of the treatment of type 2 diabetes. Ôûá