Diabetes is an important, yet perhaps under-recognized, comorbid condition in heart failure. In clinical trials, diabetes is present in roughly one-quarter of stable outpatients with heart failure and almost one-half of patients hospitalized for heart failure (Table 39-3) (5). Thirty percent of stable, elderly heart failure patients have diabetes, and the prevalence of diabetes rises to 40% in elderly patients hospitalized with heart failure (3,6,7). Conversely, diabetes is frequently complicated by the presence of heart failure, present in 12% of type II diabetics, compared to only 5% of nondiabetics (8).

Several studies have clearly established diabetes as a risk factor for developing heart failure. After myocardial infarction, diabetic patients are more likely than nondiabetic patients to have complicated hospital courses, including the development of heart failure (9). The Framingham Heart Study found that diabetic men had a twofold elevated risk of developing heart failure while diabetic women had a fourfold elevated risk (10). More recent population-based studies have found similarly elevated risk attributable to diabetes (11,12,13); for example, heart failure incidence was reported at 31 per 1,000 person-years in diabetics compared to 13 per 1,000 person-years in nondiabetics (12). The risk of heart failure is magnified in diabetics with poor glycemic control (14); for each 1% increase in the glycosylated hemoglobin (HbA1c) level, risk of heart failure increases by 8% (15). Diabetes-related risk factors such as hypertension, CAD, and left ventricular hypertrophy all
independently contribute to the development of heart failure (16,17). Further increasing the risk of heart failure are pathophysiological processes associated with diabetes and insulin resistance, including increased central sympathetic nerve activity (18), endothelial dysfunction (16), and preferential myocardial utilization of fatty acids, which may depress myocardial contractility and increase myocardial susceptibility to ischemic injury (19).

Not only does diabetes increase risk of heart failure, but heart failure itself increases the risk of insulin resistance and diabetes (7,20,21,22). According to one study, the odds of developing diabetes is increased threefold by the presence of heart failure (7) and an additional study found that the risk of diabetes increased with heart failure severity (22). Underlying pathophysiological mechanisms that may directly promote the development of insulin resistance in heart failure include sympathetic nervous system activation and elevated circulating free fatty acids (23).

CAD, hypertension, and a specific diabetic cardiomyopathy are the most common etiologies of heart failure in diabetics, termed by Bell the cardiotoxic triad (24). Hypertension and CAD have consistently been ranked as two of the most important risk factors for development of heart failure in epidemiologic studies (10,11). Diabetes also promotes development of cardiomyopathy independent of CAD or hypertension, via direct effects of hyperglycemia and insulin resistance on the heart. Advanced glycation end-products in the myocardium lead to fibrosis and altered calcium homeostasis, leading to diastolic and systolic dysfunction. Additional cardiotoxic effects of hyperglycemia and insulin resistance include cardiac hypertrophy, endothelial dysfunction, inflammation, and lipotoxicity (25). The relative contribution of diabetes and associated comorbid conditions such as hypertension or CAD is variable.

Diabetes not only predicts development of heart failure but also predicts poor outcomes in patients with established heart failure. Analyses from the Studies of Left Ventricular Dysfunction (SOLVD) trial have shown that diabetes is a risk factor for progression from asymptomatic left ventricular dysfunction to symptomatic heart failure [relative risk (RR) = 1.6] as well as a risk factor for all-cause mortality (RR = 1.4), although this increased risk was observed only in patients with ischemic etiology of heart failure (26,27). Although data from the Beta-blocker Evaluation of Survival Trial (BEST) also linked the diabetes-related mortality risk to ischemic disease (28), others studies have not found the risk to be specific to ischemic heart failure (29,30,31). In the Danish Investigations on Arrhythmia and Mortality on
Dofetilide (DIAMOND), diabetes predicted mortality (independent of heart failure etiology) in both systolic and diastolic heart failure. The increased mortality risk associated with diabetes has been observed in additional heart failure cohorts, including elderly patients with chronic and new-onset heart failure and advanced heart failure patients in a transplant referral center (3,29,30).

Diabetic heart failure patients derive benefit from standard, life-prolonging heart failure medical therapy. A meta-analysis of six randomized controlled trials of angiotensin-converting enzyme inhibitors (ACEIs) in systolic heart failure, including 2,398 diabetic and 10,188 non-diabetic subjects, found ACEI therapy was associated with a 14% and 15% decreased mortality risk, respectively (Table 39-4) (32). Likewise, subgroup analysis of the Assessment of Treatment with Lisinopril and Survival (ATLAS) study documented a mortality risk reduction of 14% and 6% in those with and without diabetes, respectively (33).

There is also abundant clinical trial evidence supporting the use of beta-adrenergic antagonists to reduce morbidity and mortality in diabetic patients with systolic heart failure (32,34,35,36). A pooled analysis of three randomized controlled trials, which included 1,883 diabetic and 7,042 nondiabetic heart failure patients, demonstrated a similar magnitude of benefit from beta-blocker therapy in the two cohorts (32). In the Carvedilol or Metoprolol European Trial (COMET), the predefined subgroup of diabetic heart failure patients had improved risk reduction with carvedilol compared to metoprolol, on a similar scale as the nondiabetic cohort (RR 0.85 and 0.82 for diabetics and nondiabetics, respectively) (36).

Beta-blockers may be underutilized in diabetic patients with heart failure or other cardiovascular disease, due to legitimate concerns about worsening of insulin resistance and dyslipidemia, development of hypoglycemic unawareness, or exacerbation of erectile dysfunction. However, severe hypoglycemia is extremely rare in type II diabetes. Furthermore, these undesirable effects are much more common in cardioselective beta-blockers such as metoprolol or atenolol, compared to the nonselective alpha-, beta-blocker carvedilol (37,38). Beneficial effects of the newer vasodilating beta-blockers include improved insulin sensitivity, decreased triglycerides, improved renal blood flow, and reduction of microalbuminuria. These benefits were exemplified in the recent GEMINI trial, which compared carvedilol with metoprolol for hypertension control in 1,235 subjects with type II diabetes mellitus. Although reductions in heart rate and blood pressure were similar, HbA1c and triglycerides were both increased in the metoprolol group compared to unchanged HbA1c and decreased triglycerides in the carvedilol group (39).

Aldosterone antagonists are the third standard, life-prolonging class of drugs essential to heart failure management (40,41). The Randomized Aldactone Evaluation Study (RALES) trial studied spironolactone in 1,663 patients with NYHA Class III-IV systolic heart failure, finding a 30% reduced mortality risk. However, a diabetic subgroup analysis was not reported (41). The subsequent Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS) study of 3,319 post-myocardial infarction patients with left ventricular dysfunction documented a 15% mortality risk reduction with eplerenone, and risk reduction was consistent in the diabetes (n = 2,122) and nondiabetes subgroups (40). Although the risk of developing hyperkalemia in these trials was low (2% to 5.5%), the frequency of hyperkalemia in patients with diabetes, especially those with renal dysfunction, may be substantially higher (42,43). Prevention of hyperkalemia in higher-risk patients involves lower initiation and maintenance dosing, frequent monitoring of electrolytes and renal function, concomitant loop diuretic therapy, and restricted dietary potassium (43).

Device therapy for heart failure—implantable cardio-verter defibrillators and cardiac resynchronization therapy—has now been established to reduce morbidity and mortality in subsets of heart failure patients. The major device trials have not included analyses stratified by presence of diabetes, yet since there is no evidence to suggest a differential efficacy in diabetes, device therapy is indicated for diabetic heart failure patients who meet established criteria.

The optimal medical regimen for glucose control in diabetics with heart failure is uncertain. Although poor glycemic control has been shown to increase risk of developing heart failure (15), the issue of glycemic control in patients with established heart failure has not been prospectively studied. However, improved glycemic control has potential benefits in heart failure, including improvement of myocardial glucose utilization and decrease of free fatty acids, which may be cardiotoxic (24).

The major classes of diabetes medications have possible adverse effects in heart failure. In a study of 554 advanced heart failure patients, diabetics treated with insulin had a fourfold to fivefold increased mortality risk, while noninsulin-treated diabetics were at similar risk to nondiabetics (29). Although insulin-associated increased risk has not been consistently observed (44), in a large study of elderly post-myocardial infarction patients, insulin-treated diabetics were also at particularly high mortality risk (45). Insulin therapy may merely reflect more advanced disease, but there are also plausible biological mechanisms by which insulin could directly contribute to worsening of heart failure, including increase in sympathetic nervous system activation, promotion of ventricular hypertrophy, and impairment of endothelial function (16,18,46).

Decisions regarding oral antihyperglycemic therapy in heart failure are also not straightforward. According to prescribing information, metformin therapy is contraindicated in heart failure, due to concerns about lactic acidosis (47). Regarding thiazolidinediones (TZDs), the American Heart Association/American Diabetes Assocation consensus statement has recommended that this class of medications be used with caution in NYHA Class I-II heart failure and not used at all in NYHA Class III-IV heart failure, as TZDs have been implicated as causing weight gain, edema, and worsened heart failure (48). However, fluid retention seems to resolve quickly with cessation of TZD therapy (49) and, importantly, TZDs do not adversely affect cardiac structure and function (50). Despite warnings, TZDs and metformin (so-called insulin-sensitizing medications) are used in roughly one-quarter of diabetics with heart failure (51). A recent, large study of Medicare patients has indicated that metformin and TZDs do not worsen heart failure outcomes but, rather, are independently associated with decreased mortality risk (13%). Although metformin did not increase risk of lactic acidosis in this cohort, the use of TZDs did increase heart failure hospitalizations (44). Further investigation into medical management of diabetes in heart failure is needed.

Only recently has anemia been recognized as a prevalent condition in patients with heart failure. The reported prevalence of anemia in heart failure has varied, depending on how anemia is defined and the severity of heart failure in the population studied. Anemia was present in 4% of patients with asymptomatic or mild to moderate heart failure (52) in 17% of a community cohort with new-onset heart failure (53); in 30% of patients with chronic, advanced heart failure (54); and in 49% of patients with acute decompensated heart failure (55).

There are several features of heart failure that have potential to contribute to the development of anemia. Renal insufficiency and hemodilution are present in some but not all heart failure patients with anemia (56,57). The inflammatory cytokine activation of heart failure may engender epogen resistance (58,59). Most recently, ACEIs have been associated with anemia in heart failure (60). The frequency of iron or other deficiencies in heart failure has been reported as low (53,56,61). Understanding the pathophysiology behind anemia in heart failure will require further investigation.

Anemia in heart failure is associated with increased severity of disease; patients with lower hemoglobin (Hb) levels have higher NYHA class, lower exercise capacity, decreased left ventricle ejection fraction (LVEF), an impaired hemodynamic profile (54), as well as increased B-type natriuretic peptide (BNP) and cardiac troponin levels (62). Anemia in heart failure also predicts increased left ventricular mass, as quantified by magnetic resonance imaging, and hypertrophy regresses over time as Hb level improves (63).

In the Framingham study, lower hematocrit (Hct) was associated with increased risk of developing heart failure (64), and anemia in end-stage renal disease has been associated with increased heart failure risk (65). In diverse populations of patients with established heart failure, the presence of anemia has clearly been identified as a significant prognosticator of poor outcomes. In a study of 1,061 advanced systolic heart failure patients at a single university referral center, the annual mortality risk increased by 13% for each 1g/dL decrease in Hb, independent of age, gender, CAD, renal function, and hemodynamics. Markedly increased risk was seen with relatively mild degrees of anemia (Hb <12.3 g/dL, Fig. 39-1) (54). No elevation of risk was observed at the very highest Hb levels, despite this finding in a subsequent study (66). In an analysis of SOLVD results, with each 1% decrease in Hct, adjusted mortality risk increased by 2.7% (52). Likewise, in acutely decompensated heart failure, an analysis of 906 patients demonstrated a 12% increased risk of death or rehospitalization at 60 days per 1g/dL decrease in Hb (55). In over 12,000 new-onset heart failure patients, anemia conferred an adjusted relative risk of mortality of 1.34 (53), although a smaller study did not find anemia to be a significant prognostic factor in this population (67). Lastly, in heart failure with preserved systolic function (diastolic heart failure), anemia appears to occur with similar frequency and to predict similarly reduced survival (68).