SYSTOLIC VERSUS DIASTOLIC FAILURE

Multiple studies across countries have now shown that the long-term prognosis of diastolic heart failure is not very different from that of systolic heart failure.1^–3 If that is the case then why bother differentiating between the two types of heart failure? The answer is because the treatment options differ greatly between the two conditions. Whereas treatment (drug and device) is well characterised for systolic dysfunction,4 there are precious few data and options available for treating diastolic dysfunction.4 While ejection fractions and left ventricular cavity size differ between the two forms of heart failure, signs and symptoms and neurohumoral abnormalities are similar between the two. These have been elegantly summarised recently by Chatterjee and Massie5 (table 1). It has even been proposed that these two forms of heart failure may be different phenotypes of the same pathophysiological basis.6 The stimuli responsible for these phenotypic differences are largely unknown.5^–7

EPIDEMIOLOGY

Despite advances in treatment for congestive heart failure (CHF), mortality remains 40–80% higher for diabetic subjects with CHF than for non-diabetic subjects.8 Diabetes prevalence is increasing world wide, with prevalence of diabetes among patients with CHF increasing at an even faster pace.9 A recent report based on a nationwide registry in the United States showed that 44% of patients with CHF have diabetes.10 Diabetes has a particularly pernicious effect among women for the development of CHF.11 While multiple mechanisms are responsible for development of CHF in diabetes, ischaemic heart disease and comorbidities such as obesity and hypertension play a major role.8 In the foreseeable future, doctors will have to deal with increasing numbers of subjects with diabetes, coronary disease and heart failure. Several recent developments in the field of heart failure have revolutionised the way in which patients are treated for CHF, with improvements in quality of life and mortality. Although prospective studies specifically dealing with heart failure in diabetes are lacking, extrapolation of data from recent large trials has shed light on the management of CHF in diabetes.8

BASIC MECHANISMS AND PATHOPHYSIOLOGY

Cardiac dysfunction in diabetes—both systolic and diastolic—results from the complex interplay of derangements involving activation of the sympathetic nervous system and renin–angiotensin–aldosterone system, increased levels of circulating cytokines, alterations in heart rate variability and increased oxidative stress, as reviewed recently.12 At the molecular level, hyperglycaemia-induced activation of protein kinase C has been implicated in some of the maladaptive changes seen at the cellular level in diabetes. Protein kinase C activation leads to changes in contractile protein function, stimulation of angiotensin converting enzymes (ACE) genes and inducible nitric oxide synthase activity. Increased ACE activity in diabetes leads to many of the maladaptive changes seen in the diabetic patient with heart and vascular complications. These include apoptosis and necrosis of cardiomyocytes and endothelial cells and increased interstitial fibrosis. Animal and human studies show that chronic hyperglycaemia leading to glycation of collagen and raised serum levels of advanced glycation end products result in increased myocardial stiffness. Further, at the cellular level, impairment of calcium homoeostasis is frequently seen in diabetic cardiomyocytes. These changes have been extensively studied in animal models. Derangements in calcium homoeostasis lead to decreased rates of release and reuptake of calcium into the sarcoplasmic reticulum. Changes at the receptor level are seen as decreased expression of sarcolemmal sodium–calcium exchanger. Preferential use of free fatty acids by the diabetic heart over time lead to lipid accumulation in the myocardium, resulting in cell damage and destruction. Thus, there is evidence for development of a distinct cardiomyopathy in diabetes independent of coexisting conditions such as coronary disease and hypertension and explains why diabetic subjects develop heart failure even in the absence of epicardial coronary artery disease.

Histomorphometric and electron microscopic studies of cardiomyocytes from cardiac biopsy samples taken from these subjects with different forms of heart failure have shown differences in myocyte diameter, myofibrillar density and response to administration of protein kinase A.13 The relevance of the distinction between the two types of heart failure becomes somewhat redundant when you consider that there is often overlap of both forms of failure in the same subject. It may be impossible to quantify the relative contribution of each form of failure to the overall clinical presentation in a subject with advanced systolic dysfunction. It is not difficult to see why the overlap will be especially common in diabetes. The prevalence of obesity, hypertension, autonomic dysfunction and ischaemic heart disease is markedly high in diabetic subjects and all of these are associated with the development of diastolic dysfunction.

DIAGNOSIS AND PROGNOSIS

Diabetic hearts often show overlapping diastolic and systolic dysfunction,14 even though by traditional definitions of ejection fraction the patient may be considered to have “pure” systolic or “pure” diastolic dysfunction. While echocardiography is well established as a modality for assessing systolic function,15 technical advances have made it possible for echocardiography to comprehensively assess left ventricular diastolic function.16 Traditional measures of diastolic dysfunction have been reduced early diastolic filling, prolongation of isovolumic relaxation time and increased atrial filling. Using newer techniques that are less load dependent such as tissue Doppler imaging, Boyer and coworkers reported a diastolic dysfunction prevalence of 75% in asymptomatic, normotensive diabetic subjects.17 The tremendous impact of diastolic function on mortality is shown in the study by Redfield and coworkers, where even mild diastolic dysfunction conferred a hazard ratio of 8.31 for mortality compared with subjects with normal diastolic function.18 Data from the Strong Heart study19 and the study of Fang et al20 show the independent effect of diabetes on indices of left ventricular systolic function. Compared with non-diabetic subjects, those with diabetes have greater left ventricular mass, and lower left ventricular fractional shortening after adjusting for confounding covariables. Follow-up studies show that diabetic subjects have more rapid decline in systolic function than those without diabetes.21

Tribouilloy and coworkers reported the markedly increased mortality in patients with heart failure with preserved ejection fraction among diabetic compared with non-diabetic subjects.22 This was independent of the presence of clinical coronary disease. Despite the acknowledged limitations of the study, most notably the undertreatment of coronary disease and risk factors in both groups, the remarkably high mortality among diabetic patients (with preserved left ventricular ejection fractions) is very disconcerting.

FUTURE DIRECTIONS

What can we do about the high mortality in diabetic patients with heart failure? Are there conditions specific to diabetes that merit special attention? Is it possible that treatment aimed at heart failure in the general population may be less effective in diabetic subjects? Specific answers to these questions will need more research and clinical trials. There are certain leads, however, which may potentially be investigated. Several investigators have reported a link between poor glucose control and increased heart failure morbidity and mortality.23 However, it is not known whether hyperglycaemia has a causal role or is merely a marker for increased events. Either way, there are no data available to show that tight blood sugar control will reduce the cardiac event rate in these subjects and future studies in this area are urgently needed. Diabetes is associated with widespread endothelial dysfunction, which results in reduced availability of nitric oxide. The benefit of supplementing nitric oxide (by using a combination of hydralazine nitrates and standard heart failure treatment) to improve heart failure prognosis was recently shown in the African–American Heart Failure Trial (A-HeFT). This trial gets more interesting when one sees the magnitude of benefit achieved despite having a disproportionately large number of diabetic patients.24 Might a similar strategy work in diabetic subjects with heart failure?

Thus while scientists ponder the mechanisms and differences in pathophysiology and pathogenesis of systolic and diastolic dysfunction in diabetes, clinicians will anxiously await newer breakthroughs in treatment modalities to help to reduce the excess mortality in diabetic patients with heart failure. Such trials are urgently needed.