Manual of Cardiovascular Medicine

I.INTRODUCTION. The presence of type 1 diabetes mellitus (T1DM) or type 2 diabetes mellitus (T2DM) confers an increased risk of coronary artery disease (CAD). A diagnosis of T2DM in the fifth decade is associated with a 6 to 7 years decrease in longevity largely because of complications of atherosclerosis. In the United States, the prevalence of T2DM correlates closely with obesity and has shown a sharp rise over recent decades. A similar trend has been noted across the world, with urbanization, adoption of sedentary lifestyle, and changing dietary patterns playing a major contributory role. Consequently, the management of diabetes, along with hypertension, dyslipidemia, and obesity, has become a prime focus in both the primary and secondary prevention of cardiovascular events. In response to the guidance from the Food and Drug Administration (FDA), large cardiovascular outcome trials (CVOT) have evaluated the impact of newer therapies on patients with T2DM. The findings from these completed and ongoing trials have the potential to significantly alter the care and outcomes of patients with T2DM. The purpose of this chapter is to discuss the current epidemiology and pharmacotherapy of T1DM and T2DM with an emphasis on the latest evidence-based practice for the treatment and prevention of cardiovascular events in individuals with this disease.

II.EPIDEMIOLOGY

A.Epidemiology of diabetes

1.Diabetes is one of the most common chronic diseases in both developed and developing nations. This rise in prevalence is driven partly by rising levels of obesity, physical inactivity, and urbanization, coupled with the aging population and greater longevity. The International Diabetes Federation most recently reported that 415 million people are currently living with diabetes; many are yet undiagnosed and untreated. By the year 2040, the global population with diabetes is expected to rise to 642 million (Table 44.1).

TABLE 44.1 International Diabetes Federation Estimates of Prevalence of Diabetes
Location	Population with Diabetes in million (2017)	Projected Population with Diabetes in million (2045)
North America and Caribbean	46	62
Middle East and North Africa	39	67
Europe	58	67
South and Central America	26	42
Southeast Asia	82	151
Africa	16	41
Western Pacific	159	183

2.The greatest burden of diabetes lies in developing countries. India and China are notable for the exceptionally high prevalence in 2000 (31.7 and 20.8 million, respectively, compared with 17.7 million in the United States). The projected 2030 figures for individuals with diabetes are 79.4 million in India, 42.3 million in China, and 30.3 million in the United States.

3.T2DM has grown increasingly more prevalent in childhood and in ethnic minorities such as African Americans, Hispanic Americans, and Native Americans. Currently up to 45% of cases in adolescents are designated T2DM. Diabetes prevalence is higher in men than in women, but there are more women in total with diabetes.

4.In the United States, 95% of cases with diabetes are T2DM, but T1DM has also shown increased incidence over recent years. The increase occurs largely in the youngest individuals (<5 years) and those with moderate genetic susceptibility.

B.Spectrum of insulin resistance to T2DM

1.There is a continuum between the early pathologic features of disordered glucose metabolism and T2DM.

2.A diagnosis of diabetes can be made on the basis of a fasting blood glucose ≥126 mg/dL (7.0 mmol/L) or a 2-hour plasma glucose of ≥200 mg/dL (11.1 mmol/L) during a 75-g oral glucose tolerance test. HbA1c measurement ≥6.5% may also be utilized to establish the diagnosis of diabetes. In the absence of unequivocal hyperglycemia, a positive result should be confirmed on repeat testing. In the setting of classic hyperglycemic symptoms, a single random glucose ≥200 mg/dL is considered diagnostic. These diagnostic criteria are summarized in Table 44.2.

TABLE 44.2 Criteria for the Diagnosis of Diabetes

Diabetes

•HbA1c ≥ 6.5% (48 mmol/mol)^a

•Fasting plasma glucose ≥ 126 mg/dL (7.0 mmol/L)^a

•2-H plasma glucose ≥ 200 mg/dL (11.1 mmol/L) in 75-g oral glucose tolerance test^a

•In a patient with classic symptoms of hyperglycemia or hyperglycemic crisis, a random plasma glucose ≥ 200 mg/dL

Prediabetes

•HbA1c 5.7%–6.4% (39–46 mmol/mol)

•Fasting plasma glucose 100–125 mg/dL (5.6–6.9 mmol/L) = IFG

•2-H plasma glucose 140–199 mg/dL (7.8–11.0 mmol/L) in 75-g oral glucose tolerance test = IGT

HbA1c, glycosylated hemoglobin; IGT, impaired glucose tolerance; IFG, impaired fasting glucose.

^aIn the absence of unequivocal hyperglycemia, results should be confirmed by repeat testing.

a.It is now recognized that the microvascular and macrovascular complications are not limited only to individuals meeting the diagnostic criteria for established diabetes. The progression from normal glucose tolerance to T2DM involves intermediate stages of impaired fasting glucose (IFG) and impaired glucose tolerance (IGT), also collectively termed “prediabetes.” This pathologic continuum arises from dysregulation of the balance between insulin sensitivity and insulin secretion.

b.There is usually a long latent asymptomatic phase prior to the development of overt T2DM. Longitudinal studies suggest that insulin resistance onset occurs 10 to 20 years prior to the diagnosis of diabetes and is the best predictor of whether an individual will develop T2DM in the future. Insulin resistance places pressure on the pancreatic β-cells to augment secretion of insulin and, therefore, promotes β-cell dysfunction. Once the β-cell is unable to compensate sufficiently for the peripheral insulin resistance state, progression to T2DM will ensue. The underlying etiology of T2DM is complex, with both environmental and genetic predisposing factors promoting insulin resistance and β-cell dysfunction.

c.Epidemiologic evidence regarding the impact of prediabetes states comes from the Diabetes Epidemiology: COllaborative analysis of Diagnostic criteria in Europe (DECODE) group. The relationship between 2-hour postload glucose and cardiovascular mortality was linear, with a continuum of risk extending into and below the prediabetes glucose range. In subjects without a diagnosis of diabetes, the investigators observed no threshold level of fasting or 2-hour postload glucose concentration above which the risk of all-cause or cardiovascular mortality death increased sharply.

3.There is evidence that the pathogenic effect of hyperglycemia on the endothelial cells already exists in the prediabetes stage. The full transition from the early metabolic abnormalities of prediabetes to established diabetes probably occurs in about two-thirds of individuals.

a.There is interest in determining interventions that may decrease the likelihood of progression to diabetes. The Diabetes Prevention Program Research Group randomly assigned 3,234 participants without diabetes, but with elevated fasting and postload glucose concentrations, to placebo versus metformin (850 mg twice daily) versus a lifestyle modification program promoting exercise and weight loss. The incidence of diabetes over an average follow-up of 2.8 years was 11.0, 7.8, and 4.8 cases per 100 person-years in the placebo, metformin, and lifestyle groups, respectively. The lifestyle intervention was significantly more successful in preventing diabetes than the metformin strategy. Also of note is the Study to Prevent NIDDM (STOP-NIDDM)trial, which randomized individuals with IGT to acarbose (which primarily affects postprandial glycemia) or placebo. Patients in the acarbose group were observed to have a lower rate of diabetes diagnosis than those receiving placebo. Furthermore, a lower rate of cardiovascular disease (CVD) and hypertension over a 3.3-year mean follow-up period was reported in the acarbose group.

III.DIABETES AND ATHEROSCLEROSIS

A.Diabetes accelerates the atherosclerotic process. Autopsy series have revealed more diffuse coronary involvement, greater severity of vessel stenosis, and more severe left main disease in persons with diabetes, compared with those without. Similarly, adverse coronary anatomy findings have been observed in autopsy studies of patients with a history of childhood onset of T1DM who died before the age of 40 years. The pathophysiology of atherosclerosis in diabetes remains incompletely understood but is thought to involve hyperglycemia, lipid abnormalities, and dysfunctional endothelial and vascular smooth muscle function, coupled with a propensity for inflammation, thrombosis, and platelet activation.

B.It is recognized that the synthesis of nitric oxide (NO) by the vascular endothelium is reduced in subjects with diabetes. NO is among several key substances that maintain healthy endothelial function, which includes freedom from adhesion molecule activation, leukocyte diapedesis, platelet aggregation, and activation of thrombosis. However, the bioavailability of NO is adversely affected by hyperglycemia, high levels of free fatty acids, and insulin resistance. Current research reveals the impact of excessive oxidative stress (which can be induced by hyperglycemia, fatty acids, or insulin resistance) on NO production and also on the generation of advanced glycation end products, which are suspected to mediate various negative cellular effects in diabetes. Reactive oxygen species also activate protein kinase C with wide-ranging effects on cellular metabolism, including activation of the phosphoinositide 3-kinase (PI3K) pathway, which has a role in mediating cell adhesion and migration. The ability of NO to act on vascular smooth muscle and cause vasodilation is reduced; endothelial cell dysfunction in diabetes is coupled with increased production of vasoconstrictors such as prostanoids and endothelin.

C.Another key pathogenic feature of atherosclerosis in diabetes is the PI3K- and nuclear factor κB–stimulated alterations in adhesion molecule expression on the endothelial surface. This causes the endothelium to become more adherent to passing cells, and selectins on the surface of leukocytes attach to receptors such as intercellular adhesion molecule 1 and vascular cell adhesion molecule 1 and migrate into the intima. Here monocytes take on oxidized low-density lipoprotein (LDL) and become macrophage foam cells, which are instrumental in the development of an atherosclerotic plaque. The tendency toward small, dense LDL particles in diabetes is another factor contributing toward atherosclerosis. Smooth muscle cell proliferation ensues, leading to deposition of collagen and other extracellular matrix proteins into the plaque.

D.Platelet function is also abnormal in diabetes, with overexpression of the glycoprotein IIb/IIIa receptor promoting inappropriate platelet adhesion and activation. Upregulation of multiple coagulation factors such as factor VII, thrombin, tissue factor, and plasminogen activator inhibitor-1 increases the tendency toward clotting, therefore contributing to the risk of thrombosis at the site of a plaque rupture or stent placement.

IV.IMPACT OF DIABETES ON HEART DISEASE

A.Both T1DM and T2DM confer significantly elevated risks of CAD, acute coronary syndrome, post–myocardial infarction (MI) complications, heart failure, and probably also sudden cardiac death. In addition, the incidence of peripheral arterial disease, stroke, and end-stage renal failure are elevated in individuals with diabetes. The cardiovascular complications of T2DM account for the majority of the socioeconomic burden of this chronic disease on both individuals and society.

1.As many as 80% of individuals with diabetes will die from cardiovascular causes. Per the Framingham Heart Study, the risk of CAD is doubled in men and tripled in women with diabetes, compared with age-matched subjects without diabetes.

2.The Emerging Risk Factors Collaboration undertook a large meta-analysis that demonstrated a twofold excess risk of outcomes such as coronary heart disease, coronary death, and nonfatal MI among individuals with established diabetes. They also found diabetes to be even more strongly related to fatal than to nonfatal MI, possibly suggesting more severe manifestations of coronary disease in those with diabetes. Hazard ratios (HRs) were particularly elevated for CVD among individuals with diabetes who were female, younger, and nonsmokers or had lower-than-average blood pressure (BP).

3.Patients with diabetes are much more likely to present with atypical symptoms in the setting of an acute coronary syndrome and experience more silent MI.

4.Various angiographic trials have demonstrated that patients with diabetes undergoing percutaneous coronary intervention (PCI) or coronary artery bypass surgery (CABG) tend to have significantly more severe CAD, with a preponderance of multivessel disease and greater severity of lesions than those without diabetes. Despite the more severe plaque burden, diabetes correlates with lesser collateral vessel formation.

5.The residual risk in individuals with T2DM following their first cardiovascular event is significantly greater than that noted in the nondiabetic group even after revascularization.

6.Post-MI complications and mortality in patients with diabetes correlate with post-MI ejection fraction and the presence of multivessel coronary disease. Cardiogenic shock is more common and more severe in post-MI patients with diabetes. The higher in-hospital mortality among the postacute coronary syndrome population with diabetes is largely related to the greater incidence of acute decompensated heart failure and to a lesser extent the increased risk of reinfarction and infarct extension.

7.The CAD mortality arising from T1DM has also been studied. Laing et al. reported a prospective study of 23,751 individuals with T1DM where subjects were followed for up to 29 years. As is typical of a T1DM cohort, the relatively young age correlated with a low event rate. However, their CAD risk was several times higher than that of a matched population without diabetes.

V.RISK FACTOR MODIFICATION IN PATIENTS WITH DIABETES

A.The risk factors that predispose individuals with diabetes to develop CVD are the same as those that raise cardiovascular risks in those without diabetes. However, the prevalence and consequences of known major risk factors for CAD is generally amplified among persons with diabetes. Given the overall higher cardiovascular risk conferred, the benefits of tighter risk factor control is greater in those with diabetes than those without.

1.Dyslipidemia

a.This is a major risk factor among individuals with diabetes. Diabetes is associated with small, dense LDL particle composition, increased levels of apolipoprotein B and E, low levels of high-density lipoprotein (HDL) cholesterol, and high triglyceride (TG) levels. These lipid composition abnormalities cluster with insulin resistance and abdominal adiposity and appear to induce endothelial dysfunction and an increased susceptibility to thrombosis.

b.LDL reduction with statin therapy for both primary and secondary prevention is a cornerstone of T2DM management. The 2013 American College of Cardiology (ACC)/American Heart Association (AHA) guidelines recommend primary prevention treatment with moderate- (LDL 70 to 189 mg/dL) or high-dose statin (LDL ≥ 190 mg/dL) in all patients with diabetes aged >40 years with a 10-year risk >5%. Treatment may also be initiated in select patients with diabetes who do not meet these criteria. Some of the factors that may influence decision making include family history of premature atherosclerosis, lifetime atherosclerosis risk, abnormal coronary artery calcium score or ankle-brachial index, or high-sensitive C-reactive protein ≥2 mg/L. For secondary prevention, high-dose statin for those aged <75 years is recommended, with moderate dosages suggested in those aged >75 years and intolerant of higher dosages. Although these guidelines did not support uptitration to specific LDL targets, trials of proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibition as well as ezetimibe have now confirmed clinical benefits achieved with LDL down to 30 mg/dL with no harm. As a result, it appears prudent to achieve LDL goals <70 mg/dL in the secondary prevention setting with addition of alternate agents (ezetimibe/PCSK9 inhibitor) when these targets are not achieved with statin alone or the patient is statin intolerant.

c.Hypertriglyceridemia. Elevated fasting TG levels are characteristic of the lipid panel in diabetes and constitute an independent cardiovascular risk factor. Hypertriglyceridemia correlates with abdominal adiposity and fibrates have traditionally been considered an appropriate therapy to target hypertriglyceridemia and have often been added to statin therapy for this indication. The Fenofibrate Intervention and Event Lowering in Diabetes trial, however, showed no benefit with fenofibrate therapy and the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial did not demonstrate any benefit for subjects with T2DM with the addition of fenofibrate to simvastatin. A meta-analysis, however, suggested a benefit for fibrate therapy in this population in the subset with low HDL and low TG levels. High levels of ω-3 fatty acids are commonly utilized to reduce TG levels. Large outcome trials are currently underway testing the utility of this intervention.

d.HDL. Low HDL levels predict poor cardiovascular outcomes in a number of epidemiologic studies and are commonly present in patients with diabetes. Pharmacologic agents called cholesterylester transfer protein inhibitors can dramatically increase and maintain HDL levels with favorable effects on LDL (evacetrapib and anacetrapib). Large randomized clinical trials with this group of agents have failed to show a benefit with raising HDL in this manner. Similarly trials with nicotinic acid in the background of statin utilization have also failed to impact favorably on clinical outcomes. This agent can worsen glycemic control and should not be utilized in patients with diabetes.

2.Hypertension

a.The presence of an elevated BP in patients with diabetes serves as a strong risk factor for development of atherosclerotic events. There are multiple large trials supporting the benefit of BP lowering in subjects with diabetes. In the UK Prospective Diabetes Study (UKPDS), lowering the BP to a modest mean of 144/82 mm Hg (compared with 154/87 mm Hg) significantly reduced strokes, diabetes-related deaths, and heart failure, as well as microvascular complications. In the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial, a fixed dosage combination of perindopril–indapamide was tested against matching placebo. The intervention resulted in a 5.6 and 2.2 mm Hg reduction in systolic and diastolic BP respectively, which in turn translated to a 14% reduction in mortality, 14% reduction in coronary events, and a 21% reduction in renal events over 4.3 years of follow-up. Several trials since have demonstrated the renal protective effects of angiotensin-converting enzyme (ACE) inhibitors (or angiotensin receptor antagonists) over alternate agents.

b.The Hypertension Optimal Treatment trial demonstrated the benefit of a lower diastolic target. Among the 3,000 subjects with diabetes, but not in those without diabetes, the relative cardiovascular risk was significantly reduced in the ≤80 mm Hg group, compared with the ≤90 mm Hg group. The ACCORD trial randomized subjects with diabetes to a target systolic BP of <120 mm Hg or <140 mm Hg. The lower target group showed no difference in the primary cardiovascular outcomes end point but did have a significantly lower stroke rate; however, this was at the expense of significantly more adverse drug events and an increased risk of a creatinine rise of >1.5 mg/dL. The 2017 AHA/ACC guidelines recommend initiation of treatment in hypertensive patients with diabetes if their BP is 130/80 mm Hg or higher. This assumes that the individual has either known CVD or has a >10% risk of CVD over 10 years. The 2018 American Diabetes Association (ADA) guidelines treat BP ≥ 140/90, but note a lower goal of <130/80 mm Hg in the high-risk group and those with established CVD.

3.Tobacco

a.There is evidence that smokers with diabetes have a markedly increased risk of MI and peripheral arterial disease. Lipid abnormalities and the development of atherosclerotic plaques appear to be promoted by smoking in the setting of diabetes.

4.Obesity

a.The prevalence of obesity has more than doubled in the United States since 1980. Obesity and overweight are typically defined in terms of body mass index (BMI), with overweight being 25 to 30 kg/m², class I obesity 30 to 35 kg/m², class II obesity 35 to 40 kg/m², and class III obesity > 40 kg/m². Waist circumference and waist-to-hip ratio better reflect abdominal adiposity and are more reliable predictors of CAD outcomes than BMI. Obesity is an important determinant of cardiovascular health and is associated with widespread alterations in cardiac and vascular structure and function. It has been shown to be an independent risk factor for the development of CVD. T2DM correlates closely with obesity, especially central obesity.

b.Caloric restriction, behavior modification, and increased physical activity form the basis of weight management programs. Unfortunately, sustained weight loss is difficult to achieve with these conservative measures. Several medications have been marketed for temporary assistance with weight loss. The cardiovascular safety of these agents with the exception of liraglutide at lower dosages has not been evaluated.

(1)Orlistat is currently the only drug that is available for modulation of fat digestion. It inhibits gastric and pancreatic lipases, thus increasing the proportion of fat that is not completely hydrolyzed and is fecally excreted. The recommended prescription dose is 120 mg three times daily. A 60-mg over-the-counter version is available in some countries, including the United States. Major side effects include abdominal cramps, flatus, fecal incontinence, diarrhea, and oily stools; there is a rare association with severe liver injury.

Multiple trials have demonstrated a greater initial weight loss (approximately 3%) with orlistat, compared with placebo, and also slower weight regain in the longer term. In obese individuals with diabetes, orlistat not only promotes weight loss but also decreases HbA1c at 1 year in comparison with placebo. Vitamin supplementation is warranted because of the lack of absorption of fat-soluble agents.

(2)Lorcaserin is a 5HT2c receptor agonist that is approved at 10 mg tablet twice daily and is thought to decrease food intake and promote satiety. Cases of serotonin syndrome or neuroleptic malignant syndrome have been reported with its use and about 2% may develop valvular regurgitant lesions. In a study of 604 patients with diabetes and BMI >27, the percentage of patients losing >5% of their body weight at 1 year was 37.5% compared with 16.1%. Overall, the weight loss with this agent is about 3% to 3.6% compared with placebo.

(3)Phentermine/topiramate acts through norepinephrine release and γ-aminobutyric acid receptor modulation. It is initiated as a 3.75/23-mg extended release tablet that is then uptitrated. It was approved in 2012 and in two randomized controlled trials (RCTs), it was associated with 8.6% to 9.3% weight loss compared with placebo. Experience with utilizing this agent is critical as the combination has known adverse effects on mood as well as on cognition.

(4)Naltrexone/bupropion is an opioid receptor antagonist that was approved in 2014. It is formulated as naltrexone (32 mg) and bupropion (360 mg) twice daily. The main limiting side effect is nausea and, in RCT, the use of this agent resulted in a 3.3% to 4.8% weight loss compared with placebo.

(5)Liraglutide is a glucagon-like peptide-1 (GLP-1) agonist and, for weight loss, it is administered at a daily subcutaneous dosage of 0.6 mg that is uptitrated to 3 mg. It is contraindicated in the setting of medullary thyroid cancer and multiple endocrine neoplasia and its main side effects include nausea, vomiting, and risk for pancreatitis. A 4% to 5% weight loss compared with placebo can be expected with this agent.

5.There is growing evidence regarding the beneficial effects of significant weight loss achieved by bariatric surgery on glucose metabolism. The 2018 ADA guidelines recommend bariatric surgery as a treatment option in appropriate surgical patients with BMI ≥40 kg/m² (37.5 kg/m² Asians) regardless of the level of glycemic control. When glycemic control is a challenge, the threshold to consider surgery is 35 to 39 kg/m² (32.5 to 37.4 kg/m² in Asians). Bariatric surgery may also be considered in the group with BMI 30 to 34.9 kg/m²^{if hyperglycemia is an issue despite optimal lifestyle and treatment interventions.}

a.Bariatric surgery options include malabsorptive procedures such as the Roux-en-Y gastric bypass and restrictive procedures such as laparoscopic adjustable gastric bands and sleeve gastrectomy.

b.Weight loss post bariatric surgery is typically expressed in terms of “excess weight,” which refers to the difference between the actual and the ideal weights for an individual. Weight loss after malabsorptive bariatric surgery tends to reach a nadir at 12 to 18 months with an average of 70% excess body weight loss and a 35% decrease in BMI, with an approximate 10% weight regain in the following 10 years.

c.A meta-analysis of 22,000 patients demonstrated that an average excess body weight loss of 61% was accompanied by significant improvements in T2DM, hypertension, dyslipidemia, and obstructive sleep apnea. Indeed, bariatric surgery has demonstrated an ability to completely reverse established diabetes in a large number of subjects. In the Swedish Obese Subjects Study, a prospective, nonrandomized, intervention trial of 4,047 obese subjects, 72% of individuals with diabetes who chose the bariatric surgery option showed reversal of their diabetes at 2 years, compared with 21% of those who followed a conservative weight loss regimen of diet and exercise. At 10 years follow-up, diabetes was reversed in 36% of the bariatric surgery group and 13% of the control group. In a smaller study of 165 obese patients with diabetes by Pories, 83% showed diabetes remission at a mean of 9.4 years. In the recent Surgical Treatment and Medications Potentially Eradicate Diabetes Efficiently (STAMPEDE) trial, 150 patients with uncontrolled T2DM and a BMI between 27 and 43 kg/m² were randomized to intensive medical therapy alone, versus medical therapy plus Roux-en-Y gastric bypass or sleeve gastrectomy. The proportion of patients with HbA1c ≤6.0% at 12 months was 12% in the medical therapy group, versus 42% in the gastric bypass group (p = 0.002) and 37% in the sleeve gastrectomy group (p = 0.008).

6.Multifactorial risk factor interventions should be targeted in all patients with diabetes, regardless of whether this is a primary or a secondary prevention strategy.

a.The Steno-2 trial randomized 160 patients with T2DM and microalbuminuria to intensive versus conventional therapy for a mean period of 7.8 years. The trial was designed to evaluate the effect on cardiovascular events of an intensified, targeted, multifactorial intervention comprising behavior modification and polypharmacologic therapy aimed at several modifiable risk factors in patients with T2DM. The intensive management regimen included dietary fat restriction, 30 minutes of exercise three to five times per week, smoking cessation, ACE inhibitors or angiotensin receptor blocker administration irrespective of BP, additional agents to target BP <130/80 mm Hg, 150 mg aspirin, stepwise glycemia management to target A1c <6.5%, statins for hyperlipidemia, and fibrates for isolated hypertriglyceridemia. The intensive group achieved significantly lower BP, HbA1c, LDL, and TGs. There was an absolute risk reduction of 20% in cardiovascular events. The subjects were then followed up for a further mean of 5.5 years, at which point the primary end point of all-cause mortality was assessed. Significant differences in BP, HbA1c, LDL, and TGs were absent by this later follow-up point. The primary end point at 13.3 years was time to all-cause death, and an absolute risk reduction of 20% was found. Even beyond the period of tight risk factor control, the Kaplan–Meier curves for the first cardiovascular event continued to diverge. During the mean 13.3-year follow-up period, the mortality rate among the conventional therapy subjects was 50%, a finding that highlights the poor overall outcomes in patients with diabetes who are not intensively managed. This study established that there were long-term benefits to aggressive multifaceted risk factor management, and that tight glycemic control and treatment with aspirin, antihypertensives, and lipid-lowering drugs appeared to be additive. Therefore, current society and national guidelines stress the importance of a broad approach to targeting multiple cardiovascular risk parameters.

VI.MICROVASCULAR TRIALS

A.The role of tight control of glycemia was firmly established in the 1990s with the publication of two large trials demonstrating decreases in microvascular complications—primarily nephropathy and retinopathy—with lower glucose goals.

1.The Diabetes Control and Complications Trial (DCCT) recruited 1,441 patients with T1DM, of whom 726 had no retinopathy at baseline and 715 had mild retinopathy. Subjects were randomly assigned to an external insulin pump or three or more daily insulin injections to target a fasting glucose <6 mmol/L. Conventional therapy had no glucose goals beyond those needed to prevent symptoms and comprised one or two daily injections of insulin. The two treatment groups were followed for a mean of 6.5 years between 1983 and 1993, with the mean HbA1c attained being 7.4% and 9.1%, respectively. In the primary prevention cohort (those without baseline retinopathy), intensive therapy reduced the adjusted mean risk of retinopathy development by 76%. With the two cohorts combined, intensive glucose control reduced the occurrence of microalbuminuria by 39%.

2.The UKPDS recruited 5,102 newly diagnosed T2DM patients from 1977 to 1991 with a median baseline HbA1c of 9.1%. The 4,209 patients who could not be controlled on diet alone were managed with differing therapies to determine if there were any specific advantages or disadvantages between glucose-lowering agents. A total of 342 obese subjects were allocated to the metformin group; of the remaining patients, 30% were randomized to conventional therapy and 70% to insulin or a sulfonylurea. The intensive group aimed for a fasting plasma glucose <6 mmol/L. The median HbA1c achieved was 7.0% in the tight control group versus 7.9% in the conventional group. There was a significant 25% risk reduction (95% confidence interval [CI] 7% to 40%) in the end point of renal failure or death from renal failure, vitreous hemorrhage, or photocoagulation in the intensive control group. In summary, the two large trials published in the 1990s, DCCT and UKPDS, confirmed significant improvements in microvascular outcomes with tighter glycemic control in both T1DM and T2DM.

VII.MACROVASCULAR TRIALS

A.Despite the convincing evidence for a reduction in microvascular complications, the relationship between glycemia and cardiovascular events was not readily apparent in earlier trials using insulin and older antihyperglycemic agents only.

B.The UKPDS revealed a nonsignificant 16% risk reduction (95% confidence interval [CI] 0% to 29%, p = 0.052) for fatal or nonfatal MI, or sudden death, with intensive management. In those subjects with >120% ideal body weight who were allocated to the metformin strategy, there were significant benefits in terms of diabetes-related end points and all-cause mortality and a 39% risk reduction in MI. However, these results should be interpreted with caution because of the small numbers in the metformin subgroup.

C.Two follow-up studies, and three trials of intensive glucose control, next sought to clarify the relationship between glucose control and cardiovascular events. Both the DCCT and the UKPDS cohorts were followed up for more than a decade from recruitment and their macrovascular outcomes assessed.

1.The DCCT follow-up, known as Epidemiology of Diabetes Interventions and Complications (EDIC), monitored 93% of the original cohort until 2005, for a mean of 17 years of follow-up. Despite the loss of metabolic separation, there was emergence of a macrovascular risk reduction in the group that had received the period of tight control earlier in the course of their T1DM; a 42% risk reduction for all cardiovascular events (95% CI 9 to 63, p = 0.02) was observed in this group.

2.Similarly, the UKPDS subjects were invited for posttrial monitoring, with 3,277 patients attending annual clinics for a further 5 years. No attempts were made to maintain their previously assigned therapies, and indeed there was no persisting difference in HbA1c between groups at 1 year after initial trial conclusion. Data were then analyzed by original trial groupings. The sulfonylurea/insulin group showed a sustained decrease at 10 years for MI (15%, p = 0.01) and death from any cause (13%, p = 0.007). The metformin group showed a significant risk reduction in MI (33%, p = 0.005) and all-cause mortality (27%, p = 0.002).

3.The DCCT/EDIC and UKPDS follow-up findings can be interpreted as showing “legacy” effects, whereby tighter management of glycemia early in the disease course confers cardiovascular and survival benefits more than a decade later.

D.Three further large trials added additional information regarding the potential relationship between glycemic control and cardiovascular outcomes (Table 44.3).

TABLE 44.3 Characteristics of the Major Randomized Controlled Trials of Intensive Glucose Control in Type 2 Diabetes
	UKPDS 33	ACCORD	ADVANCE	VADT
Participants (n)	3,867	10,251	11,140	1,791
Diabetes duration (y)	0	10	7.9	11.5
Mean age (y)	53	62	66	60
History of CV disease (%)	Not reported	35	32	40
Duration of intervention (y)	10.0	3.5	5.0	5.6
Intensive group treatment	Sulfonylurea or insulin	Any oral drugs, insulin (91% rosiglitazone)	Gliclazide, other drugs including insulin	Glimepiride or metformin, plus rosiglitazone, or insulin
Standard group treatment	Diet	Any oral drugs, insulin (57% rosiglitazone)	Any oral drugs, insulin	Glimepiride or metformin, plus rosiglitazone, or insulin
Intensive group goal	FPG < 6.0 mmol/L	A1c < 6.0%	A1c ≤ 6.5%	1.5% absolute A1c reduction, cf. standard group
Standard group goal	FPG 6.1–15.0 mmol/L	A1c 7.0%–7.9%	Local standards	Local standards
Baseline median A1c (%)	7.1	8.1	7.5	9.4
Achieved median A1c (%)	7.0 vs. 7.9	6.4 vs. 7.5	6.5 vs. 7.3 (mean)	6.9 vs. 8.4
Primary CV outcome	MI and sudden cardiac death	Nonfatal MI or stroke, or CV mortality	Nonfatal MI or stroke, or CV mortality	Nonfatal MI or stroke, CV mortality, heart failure, vascular surgery, inoperable CAD, amputation
HR for primary outcome	16% risk reduction (p = 0.052)	0.90 (95% CI 0.78–1.04)	Macrovascular 0.94 (95% CI 0.84–1.06)	0.88 (95% CI 0.74–1.05)
HR for death	0.87 (p = 0.006)	1.22 (95% CI 1.01–1.46)	0.93 (95% CI 0.83–1.06)	1.07 (95% CI 0.80–1.42)

ACCORD, Action to Control Cardiovascular Risk in Diabetes; ADVANCE, Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation; CAD, coronary artery disease; CI, confidence interval; CV, cardiovascular; FPG, fasting plasma glucose; HR, hazards ratio; MI, myocardial infarction; UKPDS, UK Prospective Diabetes Study; VADT, Veterans Affairs Diabetes Trial.

Data from Kelly TN, Bazzano LA, Fonseca VA, et al. Systematic review: glucose control and cardiovascular disease in type 2 diabetes. Ann Intern Med. 2009;151(6):394–403. Copyright © 2009 American College of Physicians. All Rights Reserved. Reprinted with the permission of American College of Physicians, Inc.

1.The ADVANCE trial randomized 11,140 patients with T2DM to standard versus intensive glucose control. Intensive control was achieved with the use of gliclazide (a sulfonylurea) plus other agents as necessary to achieve an HbA1c ≤ 6.5%. After 5 years’ median follow-up, the mean A1c was 7.3% in the standard group and 6.5% in the intensive group. Insulin was prescribed for 40.5% and 24.1% of patients in the intensive group and the standard control group, respectively. There was a decrease in the incidence of microvascular events (primarily microalbuminuria) but no significant effect of the type of glucose control on major macrovascular events (HR with intensive control 0.94), death from cardiovascular causes (hazard ratio [HR] 0.88), or death from any cause (HR 0.93). Of relevance, the nonglycemic risk factors were not fully optimized in many participants, with a mean systolic BP of 135.5 ± 17.6 in the intensive glucose group and 137.9 ± 18.4 in the standard group and an LDL mean of around 120 mg/dL ± 20 in both groups.

2.The ACCORD trial also tested the effects of tight glucose control, recruiting a total of 10,251 T2DM patients who were randomized to a target HbA1c of <6% versus 7.0% to 7.9%. These participants were slightly younger than those studied in ADVANCE; they also tended to be heavier, had a higher median baseline HbA1c, and were much more likely to be using insulin prior to entry. ACCORD was discontinued at a mean follow-up of 3.5 years because of an increased risk of death in the intensive treatment arm. The primary outcome of nonfatal MI, nonfatal stroke, and cardiovascular mortality occurred in 352 patients in the intensive group versus 371 in the standard group (HR 0.90, 95% CI 0.78 to 1.04, p = 0.16). Hypoglycemia requiring medical attention and weight gain >10 kg were both more common in the intensive therapy group. The finding of harm with a very aggressive A1c target was surprising and could be related to adverse effects of hypoglycemia, especially in long-standing diabetes—19 of the 41 cardiovascular deaths were attributed to “unexpected or presumed CVD,” and so could have been hypoglycemia related. In addition, the average weight gain of 3.5 kg in the intensive control group (presumably drug mediated) may have impacted on outcomes. It has also been noted that 91.2% of intensive therapy patients were receiving rosiglitazone, which was subsequently associated in other literature with an excess cardiovascular mortality, versus 57.2% of the standard therapy group.

3.The Veterans Affairs Diabetes Trial randomly assigned 1,791 predominantly male military veterans who had a suboptimal response to therapy for T2DM to receive either intensive or standard glucose control. The mean age was 60 years, 40% had already experienced a cardiovascular event, 52% were receiving insulin, and the mean baseline HbA1c was 9.4%. This high-risk cohort was followed for a mean of 5.6 years, during which time the mean A1c was 6.9% in the intensive group and 8.4% with standard care. However, there was no significant difference observed between the two groups in any component of the primary outcome or in all-cause mortality (HR 1.07, 95% CI 0.81 to 1.42; p = 0.62).

E.The lack of clear benefits in these strategy trials was likely due to a number of factors. Duration of therapy, lack of significant separation in glycemic control between the treatment and control groups, and the nature of medications utilized may all have contributed to these findings. An uncertain era began when clinicians while being able to control hyperglycemia were unable to favorably impact the cardiovascular complications that ensued. At the turn of the century, a large number of new targets to control blood sugar in diabetes were being evaluated. At this point, drug approval for newer agents simply required evidence of glycemic control. This resulted in small, short-term studies that targeted low-risk patients with DM that on occasion led to the approval/near approval of agents that could even induce harm. Recognizing this deficiency, the FDA issued a guidance that mandated the conduct of large CVOT in patients with established/high-risk CV profile prior to the approval of agents for T2DM. These trials were large in number, had significant follow-up, utilized standard definition, and had central adjudication of all clinical events. The results of completed trials are summarized in Table 44.4.

TABLE 44.4 Completed Randomized, Double-Blind, Placebo-Controlled CV Outcome Trials in Type 2 Diabetes Mellitus
Trial	Drug	N	Primary CV Outcome	HR	p-Value
DPP-4 Inhibitors
EXAMINE	Alogliptin	5,380	Composite: CV death, nonfatal MI, or nonfatal stroke	0.96	0.32
TECOS	Sitagliptin	14,671	Composite: CV death, nonfatal MI, nonfatal stroke, hospitalization for UA	0.98	0.65
SAVOR	Saxagliptin	16,492	Composite: CV death, MI, stroke	1.00	0.99
GLP-1 Agonists
ELIXA	Lixisenatide	6,068	Composite: CV death, MI, stroke, hospitalization for UA	1.02	0.81
LEADER	Liraglutide	9,340	Composite: CV death, nonfatal MI, or nonfatal stroke	0.87	0.01 (for drug superiority)
SUSTAIN-6	Semaglutide	3,297	Composite: CV death, nonfatal MI, nonfatal stroke	0.74	<0.001 for noninferiority
SGLT-2 Inhibitors
EMPA-REG Outcome	Empagliflozin	7,020	Composite: CV death, nonfatal MI, nonfatal stroke	0.86	0.04 (for drug superiority)
CANVAS	Canagliflozin	10,142	Composite: CV death, nonfatal MI, nonfatal stroke	0.86	0.02 (for drug superiority)

CV, cardiovascular; DPP-4, dipeptidyl peptidase-4; GLP-1, glucagon-like peptide-1; HR, hazards ratio; MI, myocardial infarction; SGLT-2, sodium glucose transport protein subtype 2; UA, unstable angina.

Trials: EXAMINE, Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care; TECOS, Trial Evaluating Cardiovascular Outcomes with Sitagliptin; SAVOR, Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus; ELIXA, Evaluation of Lixisenatide in Acute Coronary Syndrome; LEADER, Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results; SUSTAIN-6, The preapproval Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes; EMPA-REG, Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients–Removing Excess Glucose; CANVAS, The Canagliflozin Cardiovascular Assessment Study.

F.Trials with dipeptidyl peptidase-4 (DPP-4) inhibitors. These agents held much promise to improve macrovascular outcomes as a number of favorable mechanisms were identified in early animal and human studies. Three large CVOT with sitagliptin (Trial Evaluating Cardiovascular Outcomes with Sitagliptin, n = 14,671), alogliptin (Examination of Cardiovascular Outcomes with AlogliptIn Versus Standard of Care, n = 5,380) and saxagliptin (Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus, n = 16,492) were conducted. In all of these trials, composite cardiovascular outcomes were similar in the treatment and control groups confirming noninferiority. An increased risk of heart failure was noted with this class of agent. The signal for heart failure was most apparent with saxagliptin, increased with alogliptin, but absent with sitagliptin.

G.Trials with sodium glucose transport protein subtype 2 (SGLT-2) inhibitors. The Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients–Removing Excess Glucose (EMPA-REG) Outcomes trial evaluated 7,028 patients with T2DM and high risk of cardiovascular events to empagliflozin + standard of care versus placebo + standard of care. Over a median follow-up of 3.1 years, treatment with empagliflozin (pooled for patients receiving 10 and 25 mg once-daily empagliflozin) reduced the primary composite outcome (death from cardiovascular causes, nonfatal MI, or nonfatal stroke) from 12.1% to 10.5%; HR 0.86, p = 0.04. A significant decrease in all-cause (5.7% vs. 8.3%) and cardiovascular mortality (3.7% vs. 5.9%) was noted in the treatment group along with a remarkable 35% relative risk reduction in hospitalization for heart failure. This favorable action on heart failure has led to trials in heart failure in patients with both preserved and reduced ejection fraction EMPagliflozin outcomE tRial in Patients With chrOnic heaRt Failure With Preserved Ejection Fraction (EMPEROR) regardless of the presence of diabetes. The SGLT-2 inhibitor canagliflozin was evaluated in The Canagliflozin Cardiovascular Assessment Study program that enrolled 10,142 patients at high cardiovascular risk, randomized to canagliflozin + standard of care versus placebo + standard of care. Utilizing similar end points as with empagliflozin, canagliflozin resulted in a reduced event rate (26.9 vs. 31.5 participants per 1,000 patient-years; HR 0.86; p = 0.02 for superiority) over a mean 188 weeks of follow-up. An increased risk of amputation (6.3 vs. 3.4 participants per 1,000 patient-years; HR 1.97) was associated with canagliflozin. A large outcome trial with a third agent dapagliflozin is currently underway.

H.Trials with GLP-1 receptor agonists. Three outcomes trials evaluating this group of agents have been completed. The largest trial Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results evaluated liraglutide compared with placebo in 9,340 patients over 3.8 years of clinical follow-up. The event rate for the composite end point of cardiovascular death, nonfatal MI, and nonfatal stroke was 13% versus 14.9%, HR 0.87, p = 0.01 for superiority. After being superimposed early, the event curves diverge over time, suggesting an antiatherosclerotic mechanism. The preapproval Trial to Evaluate Cardiovascular and Other Long-term Outcomes with Semaglutide in Subjects with Type 2 Diabetes trial reported on cardiovascular outcomes in 3,297 subjects randomized to once-a-week semaglutide compared with placebo. The primary triple end point was significantly reduced with semaglutide (6.6% vs. 8.9%, HR 0.75, p < 0.001 for noninferiority). Lixisenatide was tested versus 6,068 subjects with diabetes and within 6 months of an acute coronary syndrome in the Evaluation of Lixisenatide in Acute Coronary Syndrome trial. The primary end point was a composite of cardiovascular death, MI, stroke, or hospitalization for unstable angina. Over 25 months of follow-up, event rates in both groups were similar (13.4% vs. 13.2%). Outcome trials with dulaglutide, albiglutide, and exenatide are currently ongoing.

VIII.DIABETES PHARMACOTHERAPY

A.The spectrum of medications available to manage diabetes has expanded greatly over recent years. There has been increasing recognition that there may be significant differences between the cardiovascular impact of the various medications, even if the glycemic effects are equivalent. Agents with proven cardiovascular benefits should be preferentially considered in treating patients with and at high risk for CVD.

1.Insulin is available in short-, medium-, and long-acting preparations, with synthetic human insulin and analogs of human insulin differing in their rates of absorption and durations of action. Mixtures of rapid short-and intermediate-acting insulin are also commonly used. Recombinant human insulin is synthesized using Escherichia coli bacteria, and subtle variations in the amino acid chain deter the insulin molecules from forming aggregates, thus creating a faster-onset, shorter-acting drug than regular insulin.

2.The different recombinant rapid-acting insulins—lispro, aspart, and glulisine—each have differing amino acid substitutions. All three can be injected subcutaneously to prevent postprandial glucose elevations, for rapid correction of elevated glucose, and in insulin pumps. The onset of action is 15 minutes, with peak effect around 1 hour and a duration of action around 3 to 4 hours.

3.Regular insulin is designated as a short-acting preparation. It has onset of action within 30 to 60 minutes, a peak time of 2 to 4 hours, and a duration of action around 6 to 8 hours. It can be used in intravenous infusions, as well as subcutaneously.

4.Neutral protamine Hagedorn (NPH) is an intermediate-acting insulin. It’s time to onset is 1 to 3 hours, and the total duration of action is 12 to 16 hours. It can be injected either once daily at bedtime to lower fasting glucose levels (typically in individuals on oral hypoglycemics who are above A1c target) or as a twice-daily basal medication in combination with a shorter-acting prandial insulin. This latter approach is known as a basal-bolus regimen and is widely used to achieve tight glucose control. In the basal-bolus regimen, a twice-daily intermediate-acting insulin, or a once-daily long-acting insulin, provides basal coverage and controls the fasting glucose levels. The three prandial insulin doses, administered just before mealtimes, cover the ingested carbohydrates and so limit postprandial glucose elevation. This is considered to be the most “physiologic” method of insulin administration, because it simulates the insulin-release patterns of the β-cells.

5.Glargine and detemir are the available long-acting agents, usually injected at night, and show minimal peaking. Glargine has a 24-hour profile of action, whereas detemir is more variable at 6 to 24 hours.

6.Insulin can cause weight gain and carries a risk of severe hypoglycemia. The hypoglycemic risk is increased in patients with renal and/or hepatic dysfunction, because the liver and kidneys are responsible for the majority of gluconeogenesis and glycogenolysis; in addition, insulin is renally excreted. The associated weight gain and hypoglycemia raised questions when the ACCORD trial revealed higher mortality in the intense therapy arm, where 77% of subjects were using insulin (vs. 55% in the standard care arm). The recent Euro Heart study also suggested harm related to insulin use. Insulin-treated patients with diabetes had an adjusted 1-year HR for mortality of 2.23 (95% CI 1.24 to 4.03, p = 0.006) and for cardiovascular events of 1.27 (95% CI 0.85 to 1.87, p = 0.230) compared with those taking oral hypoglycemic agents (predominantly sulfonylurea, metformin, or a combination of the two). However, this study was nonrandomized and the choice to prescribe insulin was made by the treating physicians. The Outcome Reduction With Initial Glargine Intervention trial was a 6.2-year dedicated cardiovascular outcomes trial of glargine which demonstrated no increased cardiovascular risk.

B.Insulin pumps are devices that can be programmed to release a continuous infusion of insulin into the subcutaneous tissue. An abdominal infusion site is the usual location, and the catheter should be changed every 2 to 3 days. These pumps are generally used in patients with T1DM and are most suited to knowledgeable and motivated individuals. They have been demonstrated to achieve greater success with target HbA1c levels and reduce the number of severe hypoglycemic episodes, compared with traditional subcutaneous insulin injections. A newer innovation is the subcutaneous continuous glucose-monitoring system that detects glucose levels in the interstitium of subcutaneous tissue. The real-time display of glucose levels can assist patients in anticipating insulin requirements and avoiding severe hypoglycemia. The currently marketed devices are each limited to a few days of continuous wear. Closed-loop systems that consist of both a subcutaneous continuous glucose sensor and an insulin pump are in development.

C.Metformin addresses glycemia by several mechanisms. It reduces hepatic gluconeogenesis by inhibiting glucose-6-phosphate dehydrogenase and works as a peripheral insulin sensitizer, with promotion of insulin-induced glucose movement into skeletal myocytes and adipocytes. Because it does not stimulate insulin release, there is minimal hypoglycemic risk. This drug is considered first line in overweight patients with T2DM and is not associated with the weight gain that is encountered with insulin and the sulfonylureas. Metformin’s insulin-sensitizing mechanisms have also found a role in the treatment of metabolic syndrome, polycystic ovarian syndrome, and nonalcoholic fatty liver disease, each of which is associated with insulin resistance. There is evidence for improvements in lipid profiles with metformin use. Metformin is contraindicated with impaired renal function (creatinine > 1.5 mg/dL in males and > 1.3 mg/dL in females is often quoted as the threshold) and also decompensated heart failure, where the risk of lactic acidosis may be elevated. Lactic acidosis, which can arise in the setting of reduced renal lactate clearance coupled with insufficient uptake of lactate into the liver because of gluconeogenesis inhibition, is considered to be rare. However, it is standard of care to discontinue metformin during periods of renal impairment or inpatient heart failure treatment and for 24 hours before and 48 hours after injection of iodinated contrast agents. The majority of the reported side effects are gastrointestinal and can include diarrhea, nausea, early satiety, and abdominal pain.

D.Sulfonylureas are insulin-stimulating oral agents, with action on the adenosine triphosphate (ATP)-sensitive potassium channel found on the pancreatic β-cell. Binding of a sulfonylurea causes a decrease in potassium efflux through the K⁺_ATP channel, inducing membrane depolarization that leads to calcium influx into the β-cell. Insulin release from secretory granules results. The main side effect is hypoglycemia; weight gain can also be a complication. Questions have also been raised regarding their potential to inhibit ischemic preconditioning via blockade of myocardial K⁺_ATP channels. The commonly used second-generation sulfonylureas, such as glipizide, glimepiride, and glyburide (also known as glibenclamide outside the United States), are largely metabolized by the liver and excreted renally. Glyburide is not recommended as first line for patients with heart failure given that it has active metabolites that can linger in patients with renal dysfunction.

E.Thiazolidinediones are a newer class of drugs that have attracted much controversy with regard to their effects on the heart. The two agents developed in this class, rosiglitazone and pioglitazone, act by increasing insulin sensitivity in target peripheral tissues. This is achieved via activation of the peroxisome proliferator–activated receptor (PPAR)-γ nuclear receptor in myocytes and adipocytes, encouraging insulin-stimulated glucose transport into the cell. Pioglitazone also acts as a partial agonist of the PPAR-α receptor, which is believed to be the reason for increased HDL and decreased LDL and TGs observed with this drug. Peripheral edema is a noted side effect; hence, this class of drugs should not be used in patients with New York Heart Association class III–IV heart failure because of concerns for sodium and water retention and possible precipitation of heart failure decompensation. Weight gain, by expansion of subcutaneous adipose tissues, is also an associated complication. There is some evidence, mostly using surrogate outcomes such as carotid intimal thickness and progression of coronary atherosclerosis by intravascular ultrasound, for a deterrent effect on atherosclerosis by pioglitazone in comparison with a sulfonylurea. Analysis of a database including 16,390 clinical trial participants demonstrated that pioglitazone is associated with a significantly lower risk of death, MI, or stroke among a diverse population of patients with diabetes. However, the risk of heart failure was increased with pioglitazone, although without an associated increase in mortality. Conversely, rosiglitazone has been associated with an excess risk of MI and cardiovascular mortality in a meta-analysis. This led to the subsequent publication of the Rosiglitazone Evaluated for Cardiovascular Outcomes in Oral agent combination therapy for type 2 Diabetes (RECORD) trial, which demonstrated overall noninferiority for a combination of metformin and sulfonylurea, but an increased risk of heart failure hospital admission or death with rosiglitazone. The PROspective pioglitAzone Clinical Trial In macroVascular Events (PROactive) study, however, demonstrated that pioglitazone decreased primary cardiovascular composite outcome when compared with placebo although the result did not quite reach statistical significance. In addition, the Insulin Resistance Intervention after Stroke (IRIS) study which compared pioglitazone to placebo in insulin-resistant patients with prior transient ischemic attack or stroke showed a significant reduction in further microvascular complications.

F.Meglitinides, such as repaglinide and nateglinide, are relatively short-acting oral insulin secretagogues. They are taken just prior to a meal and help to lower postprandial glucose levels. Because of their short duration of action, meglitinides are useful in the elderly and individuals with erratic eating habits.

G.α-Glucosidase inhibitors such as acarbose and miglitol delay the absorption of complex carbohydrates via their action on the glycosidase enzymes in the brush border of the small intestine. They are generally used as an adjunctive therapy and can be useful in controlling postprandial glycemia. As described above, the STOP-NIDDM trial suggested a role for acarbose in reducing the progression of IGT to diabetes. The major side effects are flatulence and abdominal pain.

H.Amylin is also an adjunctive agent and can be combined with prandial insulin therapy in T1DM or T2DM patients not meeting glucose targets. Pramlintide is a synthetic version of endogenous amylin, which is synthesized by β-cells and secreted with insulin in response to a carbohydrate load. The major effects appear to be inhibition of gastric emptying and suppression of glucagon release. Insulin doses should be reduced on initiation of pramlintide injections to avoid potential hypoglycemia. This agent may assist in reducing body weight.

I.Incretins are a new class of agents for T2DM and include the GLP-1 and glucose-dependent insulinotropic peptide (GIP). The incretin hormones stimulate β-cell proliferation in animal models and are found at lower-than-normal levels in patients with T2DM. The active form of GLP-1 is the amide GLP-1(7 to 36), which is secreted by entero-endocrine L-cells of the ileum and colon in response to a carbohydrate load. These agents reduce HbA1c and decrease body weight, BP, and LDL cholesterol. The main side effects with this class are gastrointestinal and pruritus at the injection site. Exenatide is a synthetic short-acting GLP-1 analog that is subcutaneously injected twice daily. It works by potentiating insulin secretion, decreasing postprandial glucagon, delaying gastric emptying, and promoting weight loss. Liraglutide works in a manner similar to exenatide but has a longer half-life and is dosed once daily. Liraglutide and exenatide have demonstrated the ability to reduce systolic BP. Liraglutide also has a positive effect on the lipid profile and cardiovascular risk biomarkers. Side effects include diarrhea and vomiting, and hypoglycemia is also a concern. DPP-4) inhibitors work by deterring the DPP-4-mediated degradation of GLP-1 and GIP into inactive metabolites. DPP-4 inhibitors are weight neutral and carry a low risk of hypoglycemia.

J.SGLT-2 inhibitors are the newest class of diabetes medications and work by inhibiting glucose reabsorption in the proximal tubule of the kidney by binding to the SGLT-2 receptor in the first part of the proximal convoluted tubule. This results in glycosuria and in weight loss along with a 1% lowering of HbA1c. It also increases insulin sensitivity in the periphery. These agents decrease BP, and decrease TG and uric acid levels while mildly increasing LDL levels. The increased amount of glucose in the urine can also worsen the infections already associated with diabetes, particularly urinary tract infections and thrush (candidiasis). There are also concerns it may increase risk of diabetic ketoacidosis. Examples of drugs in this category include dapagliflozin, canagliflozin, and empagliflozin.

IX.REVASCULARIZATION IN PATIENTS WITH DIABETES

A.The cardiovascular management of patients with diabetes incorporates not only careful consideration of medical regimens but also review of the available literature with regard to revascularization strategies. There have been many advances in evidence-based management of obstructive coronary lesions over recent decades, and many of the trials have identified strategies that confer particular benefit in individuals with diabetes. The choice of CABG versus PCI applies when the coronary anatomy and patient characteristics are amenable to both modalities of revascularization. In practice, there are significant proportions of patients in whom only one modality is realistic. A heart team approach involving a clinician, surgeon, and interventionalist should be utilized to make treatment decisions in this set of patients.

1.CABG versus PCI. A series of landmark trials in patients with obstructive CAD have defined current practice with regard to the decision between CABG and PCI. The finding in 1997 that the Bypass Angioplasty Revascularization Investigation (BARI ) subgroup with diabetes had a significant survival benefit in favor of CABG heralded the onset of specific coronary strategies for patients with diabetes. The subgroup was not prespecified, but the survival benefit was considerable: In subjects receiving insulin or oral agents for diabetes (347 of the total cohort of 1,829), those randomized to CABG had a 80.6% 5-year survival, whereas balloon angioplasty was associated with 65.5% survival. This was in the interventional era prior to coronary stenting, and the explanation for the poor PCI outcomes in BARI subjects with diabetes is a presumed high rate of vessel restenosis. In addition, patients with diabetes tend to have more diffuse CAD, and lesions distant from the angioplasty site are left unprotected by a PCI strategy.

2.CABG versus PCI with bare metal stents (BMSs) and glycoprotein IIb/IIIa inhibitors. The widespread adoption of coronary stenting—at this stage with BMS—was successful in reducing restenosis rates. Sustained angiographic coronary patency also correlated with superior survival outcomes, symptom scores, and improved regional left ventricular systolic function. The randomized Arterial Revascularization Therapy Study of CABG versus PCI in multivessel disease showed equivalence of major outcomes for CABG and PCI strategies, with the caveat that the incidence of repeat revascularization was higher in the PCI group. This trial was also notable for its diabetes subgroup results; there was a benefit in terms of 1-year repeat revascularization and major cardiovascular events in patients with diabetes who underwent CABG, as compared with stenting. Other studies of this era, including the Stent or Surgery trial published in 2002, demonstrated superiority of CABG over PCI in patients with multivessel disease in general.

3.CABG versus drug-eluting stents (DESs). The Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery (SYNTAX) trial investigated the strategies of PCI with DES versus CABG in patients with triple-vessel or left main disease; approximately 26% of SYNTAX participants had diabetes. An evaluation of the 452 SYNTAX subjects with diabetes, 40% of whom received insulin, showed that mortality was significantly higher among subjects with diabetes compared with those without, regardless of the revascularization strategy. At 1 year, the only outcome difference between PCI and CABG in patients with diabetes was the excess of repeat revascularization in the PCI arm. Patients with diabetes had significantly increased repeat revascularization rates compared with those without diabetes when treated with DES, but not after CABG. Stroke was nonsignificantly higher in the SYNTAX CABG arm. The Optimal Management of Multivessel Disease (FREEDOM) trial, completed in 2010, is the largest randomized trial of diabetic patients undergoing multivessel CABG. The T2DM had to have at least three-vessel coronary disease and one-third had a recent acute coronary syndrome. The study found that FREEDOM patients had significantly improved survival free of death, MI, or stroke and increased overall survival after CABG compared with percutaneous intervention. The rate of the composite outcome in the CABG group compared with PCI was 18.7% versus 26.6%, p = 0.005. However, the stroke rate was greater following CABG than PCI (5.2% for CABG vs. 2.4% for PCI; p = 0.03).

4.CABG outcomes. Bypass conduit attrition rates are known to be higher in patients with diabetes. As in patients without diabetes, the use of the internal mammary arteries (IMAs) offers higher long-term patency (in the region of 90% at 10 years) compared with reversed saphenous veins (long-term patency varying between 40% and 75%). Therefore, IMAs are particularly attractive in patients with diabetes, although the practice of using bilateral IMA grafts is controversial because of the potential for sternal wound infections. In addition to wound infections, patients with diabetes have an elevated risk of perioperative stroke with CABG surgery as shown in the FREEDOM trial.

5.Revascularization versus optimal medical therapy. The Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial raised questions regarding the relative merits of a purely medical versus medical and PCI strategy in patients with obstructive CAD and stable symptoms. This somewhat controversial study concluded that PCI added to optimal medical therapy did not reduce the primary composite end point of death and nonfatal MI or reduce major cardiovascular events over a 2.5- to 7.0-year period. A third of the participants had diabetes. The 1,605 patients with diabetes enrolled in the PCI versus optimal medical therapy stratum of BARI-2D did not show any difference in cardiovascular outcomes. However, both of these trials randomized patients after an initial angiogram that defined anatomy, and the decision to enroll in the PCI arm of BARI-2D (as opposed to the CABG arm) was made once anatomy was known and at the discretion of the investigators.

6.Screening asymptomatic patients with diabetes for CAD. Given the elevated risk of CAD in patients with diabetes, investigators have attempted to determine whether identification of asymptomatic CAD can improve future outcomes for patients with diabetes. The Detection of Ischemia in Asymptomatic Diabetes study used a nuclear cardiac stress imaging protocol to identify patients likely to have CAD. At 5 years, there was no difference in nonfatal MI or mortality between the nuclear screening and standard care groups. However, the screening arm was notable for a very low rate of significant ischemia detection, and there was a very successful protocol of optimal medication therapy among trial participants.

ACKNOWLEDGEMENTS: The authors wish to thank Amanda Vest, MD, and Leslie Cho, MD, for their contributions to an earlier edition of this chapter.