INTRODUCTION
Diabetes mellitus has reached epidemic proportions, affecting more than 170 million individuals worldwide. Global estimates for the year 2025 predict a further increase of almost 50%, with the greatest increases in the developing countries of Africa, Asia, and South America. This trend is particularly ominous, as those in developing nations tend to develop diabetes earlier in life (ages 40-64 years) than their counterparts in the developed world (≥65 years), implying a longer duration of exposure during the most productive years and therefore a potentially greater risk of diabetes-associated morbidity and mortality.
Of the estimated 20 million persons in the United States who have diabetes mellitus, 90% to 95% have type II diabetes. Among those, 6 million are unaware of their condition. An additional 40 million show signs of insulin resistance and are at high risk of developing type II diabetes. Even more alarmingly, among obese white adolescents, 4% have type II diabetes and 25% have abnormal glucose tolerance, markedly increasing their likelihood of developing future premature cardiovascular complications.
Cardiovascular disease is the leading cause of death among patients with diabetes, accounting for over 60% of mortalities. While diabetes is uniformly recognized as an important risk factor for the development of atherosclerosis and its complications, it is perhaps less well acknowledged that diabetes is a powerful and independent risk factor for the development of heart failure.
The Framingham Heart Study was the first to demonstrate an increased risk for heart failure in patients with diabetes. The incidence of heart failure in these men and women was increased two- and fivefold, respectively, compared with men and women without diabetes. The association was even stronger in patients younger than 65 years of age, being fourfold higher in diabetic men and eightfold higher in diabetic women than in nondiabetic subjects. Since then, additional studies, including the Studies of Left Ventricular Dysfunction (SOLVD), the Heart Outcomes Prevention Evaluation (HOPE) study, and the Cardiovascular Health Study (CHS), have identified diabetes as a major risk factor for the development of heart failure. In a large registry of almost 50,000 diabetic individuals, poor glycemic control was associated with an increased risk of developing heart failure: each 1% increase in the level of glycosylated hemoglobin (HbA 1c ) was associated with an 8% increase in the risk of heart failure.
Conversely, the presence of heart failure was identified as a possible risk factor for diabetes. During a three-year follow-up of nondiabetic heart failure Italian patients, diabetes developed in 29% compared with 18% of matched controls; multivariate analysis showed heart failure to be an independent risk factor for the development of diabetes. Moreover, diabetic patients make up to 25% to 30% of all patients enrolled in large-scale heart failure clinical trials.
The dire prognosis of diabetic heart failure patients is well established. In SOLVD, diabetes was a major risk factor for cardiovascular and all-cause mortality, and in the Diabetes Insulin Glucose in Acute Myocardial Infarction (DIGAMI) study, heart failure was the most frequent cause of mortality in diabetics, accounting for 66% of deaths in the year following the first myocardial infarction.
PATHOPHYSIOLOGY
The term diabetic cardiomyopathy was first coined in 1972 by Rubler et al., who described four diabetic patients presenting with heart failure without evidence of coronary artery disease, hypertension, or valvular or congenital heart disease. Although initially controversial, the existence of diabetic cardiomyopathy has been confirmed in the past three decades by epidemiological, clinical, and laboratory studies, which have shed light on the biochemical and pathological mechanisms involved ( Fig. 26-1 ). The development of diabetic cardiomyopathy is likely multifactorial, with putative mechanisms including metabolic disturbances, myocardial fibrosis, small vessel disease, autonomic dysfunction, and insulin resistance ( Fig. 26-2 ).
Metabolic Disturbances
Isolated diabetic cardiomyocytes in diabetic patients exhibit a significant decrease in myocardial glucose supply and utilization, with the net effect of reduced adenosine triphosphatase (ATP) availability. The slow rate of glucose transport across the sarcolemmal membrane is probably due to the cellular depletion of glucose transporters 1 and 4, which can be corrected by insulin therapy. The reduced glucose oxidation in turn is caused by the inhibitory effect of fatty acid oxidation on pyruvate dehydrogenase complex, due to high levels of circulating free fatty acids. Experimental models have demonstrated that these metabolic abnormalities are associated with contractile dysfunction manifested by increased left ventricular (LV) end diastolic pressure and reduced cardiac output. Importantly, with improved metabolic control, these pathological mechanisms are potentially reversible in the early phases, with normalization of the cardiac function.
Free Fatty Acid Metabolism
Abnormalities in free fatty acid metabolism caused by insulin resistance may be important contributors to the abnormal myocardial function in diabetes. High levels of free fatty acids lead to an inhibition of glucose oxidation, resulting in reduced myocardial ATP availability. In addition, abnormally high oxygen requirements associated with increased fatty acid metabolism cause an intracellular accumulation of potentially toxic intermediates, leading to impaired myocardial performance and severe morphological changes. These metabolic changes are coupled with a relative carnitine deficiency that is common in diabetes and are potentially reversible once glycemic control is improved.
Abnormalities in Regulation of Calcium Homeostasis
The abnormal myocardial metabolism in diabetes leads to an accumulation of toxic molecules (e.g., long-chain acylcarnitines, free radicals), which in turn results in alterations in the function of regulatory and contractile proteins and decreased calcium sensitivity. The diminished calcium sensitivity along with shifts in cardiac myosin heavy chains (V1 to V3), reduction of sarcoplasmic reticulum calcium-ATPase (SERCA2a), and decreased SERCA2a pump gene expression may all contribute to impaired ventricular function. Finally, alterations in the expression of myosin isoenzymes and regulatory proteins and myosin phosphorylation have been demonstrated to contribute to the development of myofibrillar remodeling in the diabetic heart and are closely associated with abnormalities in diastolic function.
A magnetic resonance imaging (MRI) study of asymptomatic, normotensive, nonobese, well-controlled diabetic patients (mean HbA 1c , 6.1 g/dL) showed ventricular diastolic dysfunction compared with control subjects who were matched for age, gender, body mass index (BMI), and blood pressure. These findings were associated with a significantly lower ratio of myocardial phosphocreatine to ATP in diabetic patients compared with controls. Previous studies suggested that the lower phosphocreatine content and the switch in substrate preference from glucose to fatty acids may lead to lower levels of ATP in the sarcomeres, for which increased mitochondrial ATP production does not compensate. Lower cytosolic ATP concentration is associated with impaired calcium sequestration by the sarcoplasmic reticulum and impaired relaxation of cardiomyocytes.
Myocardial Fibrosis
Myocardial fibrosis and myocyte hypertrophy are among the most frequently proposed mechanisms to explain cardiac changes in diabetic cardiomyopathy. Collagen accumulation in the diabetic myocardium may be due in part to impaired collagen degradation resulting from glycosylation of the lysine residues on collagen. Hyperglycemia also results in the production of reactive oxygen and nitrogen species, which increases oxidative stress and causes abnormal gene expression, altered signal transduction, and activation of the pathways leading to programmed myocardial cell death or apoptosis. This process is associated with the glycosylation of p53, resulting in an increment in angiotensin II synthesis that has dose-dependent effects on collagen secretion and production in cardiac fibroblasts. In addition, chronic metabolic abnormalities present in diabetes (postprandial hyperglycemia, hyperinsulinemia, insulin resistance) lead to alterations in endothelin-1 and its receptors, decreased levels of insulin-like growth factor-I, and increased production of transforming growth factor-β1 that result in promotion of angiotensin II activity with an increase in myocardial collagen content.
The functional abnormalities in diabetic myocardium are associated with myocardial structural changes. Several studies using endomyocardial biopsies have shown a correlation between histological and clinical features in diabetes, with myocardial changes more pronounced in symptomatic patients and in those with cardiomegaly. The role of fibrosis in myocardial dysfunction is also supported by studies showing reversal of cardiac fibrosis by short-term pirfenidone and spironolactone treatment, with improvement in diastolic stiffness in diabetic rats. Alterations in myocardial structure are usually minimal in the early stages of diabetes and may be reversible or partially reversible. As diabetes progresses, accumulation of collagen becomes obvious and may play a major role in the development of diastolic dysfunction.
Small Vessel Disease
Structural Abnormalities of Vessels
Although the contribution of small vessel disease to diabetic cardiomyopathy is controversial, there are several structural abnormalities of small vessels that are evident in the diabetic myocardium. The capillary basement membrane is thicker in patients with diabetes, and its thickness seems to be greater compared with patients with glucose intolerance and those without diabetics. In addition, diabetes patients exhibit capillary microaneurysms with reduced capillary density, focal subendothelial proliferation, and interstitial fibrosis with myocyte atrophy.
Functional Abnormalities of Vessels
It has been proposed that diabetic cardiomyopathy is a consequence of repeated episodes of myocardial ischemia resulting from both structural and functional abnormalities in small vessels or from microvascular spasm during periods of increased myocardial demand. Such processes would lead to focal cell loss due to microvascular spasm and reperfusion injury, with the subsequent development of focal fibrosis and reactive hypertrophy in response to myocardial necrosis. The myocardial blood flow is not only reduced in diabetic patients but also correlates significantly with fasting glucose concentration and average levels of HbA 1c .
Endothelium-dependent responses of both small and large vessels are impaired in diabetic patients, including those with an otherwise low likelihood of atherosclerosis. The half-life of nitric oxide is reduced due to increased oxidative stress, and its activity is attenuated by accumulated glycosylation end products. In addition, diabetic endothelium manifests increased production of vasoconstrictor prostanoids and increased expression of protein kinase C activity. Protein kinase C activation is associated with abnormal retinal and renal hemodynamics in diabetic animals, and overexpression of the myocardial β-isoform is associated with cardiac hypertrophy and failure, implying that this may play a role in the development of diabetic cardiomyopathy by affecting the small vasculature.
Cardiac Autonomic Dysfunction
Studies in which sympathetic innervation was assessed quan-titatively using 123 I-metaiodobenzylguanidine or 11 C-hydroxyephedrine have shown a decreased myocardial uptake in 40% to 50% of diabetes patients, indicating the presence of cardiac autonomic dysfunction. It appears that this is a regional process, with the posterior myocardium being predominantly affected and with areas of proximal hyperinnervation complicating distal denervation. Myocardial autonomic dysfunction is associated with altered myocardial blood flow, with the regions of persistent sympathetic innervation exhibiting the greatest deficits of vasodilator reserve. Decreased myocardial perfusion reserve may be in part responsible for the abnormal response to exercise in the early phases of diabetic cardiomyopathy and may explain its association with impairments of diastolic function.
Insulin Resistance
Cellular insulin resistance may precede frank diabetes by a decade or more and is associated with requisite compensatory increases in plasma insulin levels to maintain glucose homeostasis in the face of impaired cellular insulin action, principally in skeletal muscle and liver. The nature and extent of the insulin resistance may be selective to certain organ systems and may vary in terms of their metabolic, mitogenic, pro-survival, and vascular actions. Insulin may act as a growth factor in the myocardium, a concept that is supported by the experimental observation that sustained hyperinsulinemia leads to increased myocardial mass and decreased cardiac output in rats. Hyperinsulinemia leads to sodium retention, which may contribute to decompensation in persons with otherwise subclinical myocardial dysfunction due to volume expansion. Hyperinsulinemia also leads to sympathetic nervous system activation, which is related to an increased response to angiotensin II and increases the stimulating effects of angiotensin II on cellular hypertrophy and collagen production, leading to myocardial hypertrophy and fibrosis and likely subsequent heart failure.
A very elegant recent study has shown that in a community-based sample of men free of heart failure and valvular disease at baseline, insulin resistance predicted heart failure incidence independently of diabetes and other established risk factors for heart failure. Furthermore, this study indicated that the previously described association between obesity and subsequent heart failure may be mediated, at least in part, by insulin resistance.
Interaction with Other Major Comorbidities
With the addition of untreated hypertension, myocardial ischemia, or both, the mild subclinical cardiomyopathy of diabetes may rapidly advance to clinical diastolic and systolic dysfunction. In clinical practice, it is difficult to distinguish the concurrent roles of hypertension and ischemia in the development of diabetic cardiomyopathy. Furthermore, the presence of silent ischemia in diabetic patients makes the diagnosis of diabetic cardiomyopathy more complicated.
Interaction with Hypertension
The prevalence of hypertension is approximately doubled in diabetic patients compared with nondiabetic controls, and the clinical and morphological features of heart disease in hypertensive diabetic patients are more severe than those of hypertensive patients or diabetic patients alone. Myocardial fibrosis and interstitial collagen deposition are greater when hypertension is associated with diabetes than when either entity exists in isolation, and these synergistic effects on neurohormonal activation and oxidative stress may promote apoptotic myocyte loss, initiating a transition from a subclinical, compensated/hypertrophied state to overt cardiomyopathy. At least one study has documented an association between diabetes, hypertension, and the development of dilated cardiomyopathy. In addition, patients with diabetes and hypertension in combination have more severe abnormalities of ventricular relaxation than those with either condition alone.
Interaction with Coronary Artery Disease
Although lipid metabolism abnormalities associated with diabetes do not have direct influence on the development of diabetic cardiomyopathy, they are at least partly responsible for enhanced coronary atherosclerosis in these patients. Enhanced coronary atherosclerosis is directly related to myocardial ischemia, increased oxidative stress, and vascular endothelial dysfunction. Compared with nondiabetic patients, those patients with diabetes demonstrate impaired recruitment of contractile reserve in noninfarct segments and greater reduction in global systolic function immediately following myocardial infarction, changes that may be related to diminished coronary flow reserve and microvascular dysfunction. Over the long term, these acute changes do not appear to be associated with a greater propensity for ventricular cavity dilation or progressive systolic dysfunction, and the increased incidence of heart failure in diabetic patients appears to be related to primary abnormalities of diastolic function.
The progression of diabetic cardiomyopathy is a dynamic process and takes several years to develop ( Table 26-1 ). In the initial phase, there is a short-term physiological adaptation to metabolic alterations that is potentially reversible once glycemic control has been restored. Thus, therapies during the early stages of diabetes can potentially prevent or delay the progression to more permanent sequelae. The late stage represents degenerative changes for which the myocardium has only limited capacity for repair. However, many factors, such as treatments, metabolic characteristics, lipid profile, blood pressure, and other individual differences, may affect the process of development of diabetic cardiomyopathy, and not all diabetic patients are affected by the same factors or to the same degree, which may result in marked variability in the clinical manifestations of diabetic cardiomyopathy.
| STAGES | CHARACTERISTICS | FUNCTIONAL FEATURES | STRUCTURAL FEATURES | DIAGNOSIS |
|---|---|---|---|---|
| Early |
| No overt functional abnormalities or possible overt diastolic dysfunction but normal ejection fraction | Normal ventricular size, wall thickness, and mass | Sensitive methods such as strain, strain rate, and myocardial tissue velocity |
| Intermediate |
| Abnormal diastolic function and normal or slightly decreased ejection fraction | Slightly increased ventricular mass, wall thickness, or size | Conventional echocardiography or sensitive methods such as strain, strain rate, and myocardial tissue velocity |
| Late |
| Abnormal diastolic function and ejection fraction | Significantly increased ventricular size, wall thickness, and mass | Conventional echocardiography |
CLINICAL RELEVANCE
LV diastolic dysfunction may be the first stage of diabetic cardiomyopathy. In the Olmsted County study, close to 50% of participants with diabetes had echocardiographic evidence of diastolic dysfunction, compared with 27% of nondiabetic subjects. Almost none of these participants had a prior diagnosis of heart failure. Remarkably, this observational study showed that even mild impairment in diastolic function is associated with an eightfold risk of all-cause mortality compared with normal diastolic function. In the same study, 14% of diabetic patients also had an LV ejection fraction (LVEF) below 0.50 compared with only 5% of nondiabetic patients, providing evidence that diabetes can affect both systolic and diastolic function.
In a study of 86 patients with diabetes (43% of whom were women), more than 40% had diastolic dysfunction: 26% had impaired relaxation and 17% had pseudonormalization on Doppler echocardiogram. These findings are noteworthy, as these subjects were young (mean age, 43 years), normotensive (mean blood pressure, 125/80 mmHg), and under excellent diabetic control (mean HbA 1c , 6.5 g/dL).
In the Strong Heart Study, enrolling 2411 Native Americans, diabetic participants had evidence of impaired LV relaxation on Doppler echocardiography. The association between diabetes and abnormal LV relaxation was independent of age, blood pressure, LV mass, and systolic function. These abnormalities were more severe in the group with both diabetes and hypertension, showing the additive deleterious effects on active LV relaxation when both these conditions are present.
In the Multi-Ethnic Study of Atherosclerosis, 6800 men and women from four ethnic groups (Americans of African, Chinese, European, and Hispanic descent) underwent cardiac MRI at enrollment. Diabetic patients manifested increased LV mass and lower end diastolic and stroke volumes compared with participants with impaired fasting glucose and with nondiabetic patients. These findings were independent of traditional risk factors and other measures of subclinical atherosclerosis (such as coronary artery calcium or carotid artery intima-media thickness) and were especially significant in African Americans and Hispanics. Although LVEF was not different among diabetic and nondiabetic women, diabetic men had a significantly lower LVEF compared with nondiabetic men, albeit within the normal range. These results strengthen the evidence in favor of a diabetic cardiomyopathy and suggest that this condition may start even before clinical diabetes is diagnosed and may have distinct characteristics among different gender and ethnic groups.
Several studies have shown that the first clinical manifestation of diastolic dysfunction is limited exercise tolerance. Poirier et al. reported that patients with well controlled diabetes and without overt coronary artery disease, hypertension, or heart failure had lower exercise performance on maximal treadmill testing than age-matched controls. The exercise limitation correlated with the severity of diastolic dysfunction as assessed by Doppler echocardiography.
Microalbuminuria appears to be an independent risk factor for the development of diastolic dysfunction, perhaps being a marker for intramyocardial microangiopathy. In the Strong Heart Study, after adjusting for age, gender, BMI, systolic blood pressure, duration of diabetes, coronary artery disease, and LV mass, the pre-valence of LV diastolic dysfunction increased as a function of increasing urinary albumin excretion. Further studies should address whether routine testing for microalbuminuria is indicated to identify diabetic patients with impaired LV relaxation.
The influence of diabetic complications on diastolic function has been investigated in several other studies. Most of these studies showed that patients with diabetic retinopathy, nephropathy, or neuropathy had significantly more abnormalities in diastolic function compared with diabetic patients without microvascular disease or with nondiabetic patients. In addition, the severity of diastolic dysfunction was related to the number of microvascular complications, glycemic control, or duration of diabetes.
Despite growing awareness of the burden of diastolic heart failure (DHF), there have been few randomized clinical trials of drug therapies for these patients (see Chapters 32 and 34 ), and no trial that specifically assessed the unique role of diabetes.
The Digitalis Investigation Group ancillary study randomized 988 patients with chronic heart failure who were in sinus rhythm and had an LVEF greater than 45% to digoxin or placebo, in addition to standard therapy. Although there were no differences in all-cause mortality (23.4% in both groups after a mean follow-up of 37 months) or in the endpoint of death or hospitalization for worsening heart failure (24% and 21% in the placebo and digoxin groups, respectively; p = 0.136), there was a trend toward benefit in the digoxin group for heart failure hospitalizations (22% and 18% in the placebo and digoxin groups, respectively; p = 0.094). This study is the basis for the level C recommendation for the use of digoxin to reduce symptoms of heart failure in patients with DHF.
The Candesartan in Heart Failure Assessment of Reduction in Mortality and Morbidity (CHARM)-Preserved arm is the first of several trials designed to study patients with DHF initiated in the last 5 years to have published its full results. The design of the study, in the context of the entire CHARM program, dictated the selection of patients with LVEF greater than 40%, allowing the inclusion of some patients with at least mild systolic dysfunction. The primary endpoint of cardiovascular death or heart failure hospitalization occurred in 22% of patients in the candesartan arm and 24% of those in the placebo arm (hazard ratio [HR], 0.89; 95% CI, 0.77-1.03). Cardiovascular deaths were identical in number in the two groups, but fewer patients in the candesartan group than the placebo group experienced heart failure hospitalizations (15.2% vs. 18.5%, p = 0.017). Although candesartan was associated with a reduction in the incidence of new-onset diabetes when compared with placebo, the authors did not present a separate analysis of its effect on mortality or heart failure hospitalizations in the diabetic patients in the study.
The Perindopril for Elderly Persons with Chronic Heart Failure (PEP-CHF) trial was a study of heart failure in elderly patients rather than a study of DHF. All 852 patients were older than 70 years and had evidence of chronic heart failure confirmed by clinical and echocardiographic criteria (LVEF greater than 40%). At the end of the 26 months of follow-up, there was no difference in the primary outcome of all-cause mortality or heart failure hospitalizations between the placebo and the perindopril groups (25.1% vs. 23.6%, p = 0.545). Perindopril decreased heart failure hospitalizations during the first year (HR, 0.63; 95% CI, 0.41-0.97) but not at the end of the follow-up (HR, 0.86; 95% CI, 0.6-1.20). In the authors’ opinion, the main reasons for the negative results were the enrollment of fewer patients ( N = 1000) than anticipated, a lower event rate than predicted, and the use of open-label angiotensin-converting-enzyme (ACE) inhibitors in these patients after the first year of follow-up.
The Study of Effects of Nebivolol Intervention on Outcomes and Rehospitalizations in Seniors with Heart Failure (SENIORS) evaluated the effect of a selective beta blocker with additional vasodilating properties in 2128 patients 70 years of age or older. Twenty percent of the study population had LVEFs above 45%. As with CHARM-Preserved, the characteristics of the overall population were quite different from those of populations with DHF in epidemiology studies, in that 63% were men and 76% had ischemic heart disease as the etiology of heart failure. Nebivolol produced a significant reduction in the primary endpoint of death or cardiovascular hospitalization, with 31.1% of patients in the nebivolol and 35.3% in the placebo group experiencing an event (HR, 0.86; 95% CI, 0.73-0.99). There was a modest trend toward an improvement in mortality (HR, 0.88; 95% CI, 0.71-1.08). The results in the group with LVEFs of 45% or greater were not presented separately. Patients without diabetes showed a significant benefit with nebivolol for the primary composite endpoint, while those with diabetes did not.
With the exception of candesartan and perhaps digoxin, which have been shown to reduce the incidence of hospitalizations, the management of DHF is based largely on clinical experience, with the goal of controlling the deleterious processes that are known to exert important effects on ventricular relaxation (i.e., hypertension, diabetes, ischemia, tachycardia, and atrial fibrillation) (see Chapter 32 ).
The first treatment goal is to provide symptomatic relief by decreasing pulmonary congestion at rest and during exercise. This can be achieved by a reduction in LV diastolic volume (with the goal of reducing LV diastolic pressure), maintaining synchronous atrial contraction (by maintaining sinus rhythm), and increasing the duration of diastole (by reducing heart rate).
Reduction in LV diastolic volume can be done by reducing total blood volume (through sodium and water restrictions and use of diuretics), decreasing central blood volume (through preload reduction with nitrates), and inhibiting the activation of the renin-angiotensin-aldosterone system (through ACE inhibitors, angiotensin-receptor blockers, and aldosterone antagonists, or a combination thereof) ( Table 26-2 ).
