Comprehensive Heart Failure Practice Guideline 

Diagnosis. The clinical diagnosis of HF depends on the presence of commonly accepted signs and symptoms. Preserved LVEF may be shown by quantifying LVEF and LV volumes or dimensions through imaging techniques such as echocardiography, radionuclide ventriculography, contrast ventriculography, or cardiac magnetic resonance imaging. Among these, echocardiography is the most commonly used and has several advantages, including availability and the ability to provide information about LV wall thickness, filling patterns, cardiac anatomy, and valvular function.

Confirmation of increased LV wall stress by documenting elevation of B-type natriuretic peptide (BNP or NT-proBNP) may be useful when dyspnea may be due to noncardiac causes.²⁷ Increased BNP or NT-proBNP identifies patients with elevation of the LV end-diastolic pressure, but does not differentiate patients with preserved versus reduced LVEF.²⁸ HF with reduced LVEF tends to be associated with greater elevation of BNP than does HF with preserved EF, but BNP is above normal in both categories of HF.²⁹ There is some overlap with the normal range.^27,28,30

Differential Diagnosis. LV hypertrophy (LVH), diagnosed by echocardiography or electrocardiography, is present in the most prevalent forms of HF with preserved LVEF. Doppler echocardiography frequently demonstrates abnormalities in LV diastolic filling.

Classification by the presence or absence of LVH has been based on the most common presentation of the disorders listed in Table 11.1. Restrictive myopathies may also be divided on the basis of myocardial disorders (noninfiltrative, infiltrative, or storage disorders) and endomyocardial disorders. In the presence of hypertrophy, the most prevalent form of HF with preserved LVEF is hypertensive-hypertrophic cardiomyopathy. The echocardiogram is more sensitive than the electrocardiogram for the diagnosis of LVH.³¹ In addition to chronic systemic hypertension, LVH may be due to other causes of LV pressure overload, such as aortic stenosis or aortic coarctation.

Detecting LVH in the absence of an obvious cause for LV pressure overload supports the diagnosis of hypertrophic cardiomyopathy. This condition is usually regional (eg, septal, apical), but may be global. It is usually familial and genetically mediated.^32,33 Increased wall thickness by echocardiography, coupled with low voltage on the electrocardiogram, strongly supports the diagnosis of an infiltrative cardiomyopathy. Among the most common infiltrative disorders is amyloidosis,³⁴ a disorder with a very poor prognosis.³⁵ In addition to low voltage, pseudo-infarction Q waves may be present. In the absence of hypertrophy, other infiltrative processes include sarcoidosis and Gaucher's disease. Sarcoid nodules in the myocardium rarely cause LV restrictive physiology, but pulmonary sarcoidosis may commonly cause pulmonary hypertension and right HF.³⁶

Less common storage disorders include hemochromatosis. Rare disorders include Fabry disease and glycogen storage diseases. Hemochromatosis has several etiologies (familial, idiopathic, and acquired) and is manifested primarily as a dilated cardiomyopathy with reduced systolic performance, but occasionally as a non-dilated, restrictive cardiomyopathy.³⁷ Fabry disease may be associated initially with normal LV mass, but later with hypertrophy. Restrictive disorders are rare, and may be associated with either LVH or normal LV mass.³⁸ Endomyocardial disorders include endomyocardial fibrosis (usually in tropical climates); the hypereosinophilic syndrome, which may or may not be related to endomyocardial fibrosis; and carcinoid.

In the absence of aortic or mitral regurgitation, LV volume overload denotes a high cardiac output because of ventricular septal defect, patent ductus arteriosus or other arteriovenous shunt, chronic anemia, thyrotoxicosis, or chronic liver disease.

It is essential to clarify the diagnosis of pericardial constriction versus restrictive disorders. In the absence of substantial pericardial fluid, the diagnosis of pericardial disease may require invasive hemodynamics, computerized tomography, or magnetic resonance imaging to identify pericardial thickening.³⁹

In contrast to the rarer forms of restrictive and infiltrative cardiomyopathies and to pericardial disease, ischemic heart disease with transient LV dysfunction is much more common. It is considered here and in other sections, particularly Section 13.

Right ventricular (RV) dysfunction is most commonly caused by LV dysfunction. In such conditions, there is pulmonary hypertension. Other causes of pulmonary hypertension, such as pulmonary thromboembolic disorders and intrinsic lung disease, may also precipitate RV dysfunction. Occasionally severe RV dysfunction may follow RV infarction. This is usually transient, but occasionally chronic RV dysfunction can cause LV dysfunction resulting from ventricular interaction, a situation in which RV pressure-volume overload may deform and displace the interventricular septum toward the LV, increasing LV diastolic pressure even as LV volume remains constant or decreases. Such conditions reduce LV compliance.

In summary, there is a broad differential diagnosis of HF with relatively preserved LVEF, and this must be kept in mind during the initial evaluation of such patients. After other disorders have been eliminated, hypertensive LVH is the most common cause of HF with relatively preserved LVEF. In analyzing HF in such patients, most emphasis has centered on LV diastolic dysfunction. Nevertheless, at the present state of knowledge, one must consider that hypertension with abnormal vascular-ventricular interaction may play a significant, causative role in the pathophysiology of the HF by severely increasing the LV diastolic pressure.

Section 13 provides a detailed approach to the diagnosis of ischemic heart disease in patients with HF by noninvasive stress imaging and by cardiac catheterization. Ischemic mitral regurgitation, acute or chronic, may aggravate HF with normal systolic performance.

In conditions with LVH, restrictive or constrictive physiology, a small decrease in intravascular volume may be associated with significant reduction in LV preload, resulting in decreased cardiac output. Orthostatic changes and prerenal azotemia provide evidence for excessive preload reduction.⁶ Acutely, in addition to diuretics, nitrates may have a role in diminishing pulmonary venous pressure and clinical congestion. Chronically, the effects may be similar, but one must be alert to the possibility of excess reduction in LV preload.

A trial of the ARE, candesartan, in patients with HF and preserved LVEF showed a trend toward reduction in the primary endpoint of cardiovascular death or hospitalization (unadjusted hazard ratio 0.89, CI 0.77-1.03, P = .118; adjusted hazard ratio 0.86, CI 0.74-1.00, P = .051).⁴⁰ At enrollment, approximately 20% of patients were receiving ACE inhibitors and 55% were receiving β-adrenergic blocking drugs. There was no subset analysis of the combination of these drugs in these specific patients, but the candesartan group showed a reduction in both hospitalizations and blood pressure.

Studies supporting the use of ACE inhibitors in patients with HF and preserved LVEF did not enroll patients with known HF. A secondary endpoint of the Heart Outcomes Prevention Evaluation (HOPE) trial was progression to HF in the following high risk patients: those older than age 55 years with either documented vascular disease or multiple cardiac risk factors, one of which was diabetes.⁴¹ In this randomized study, 9297 patients received double-blind placebo or ramipril 10 mg daily and were followed for 4.5 years. The annual risk for development of HF was approximately 2.5%, which was reduced by 23% with the ACE inhibitor. The risk reduction was independent of multiple covariates. The presence of a subsequent MI during the study increased the risk of developing HF more than eightfold. Treatment with ramipril was associated with a 33% reduction in the development of HF in those with a baseline systolic pressure above the median of 139 mm Hg versus only 9% in those whose systolic blood pressure was below the median (P = .024).

The European Trial on Reduction of Cardiac Events with Perindopril in Stable Coronary Artery Disease (EUROPA) trial studied high dose ACE inhibitor therapy versus placebo in patients older than age 18 with documented coronary artery disease.⁴² A mean follow-up of 4.2 years showed that perindopril reduced total mortality by 14% (from 6.9% to 6.1%), recurrent MI by 22% (6.2% to 4.8%), and hospital admission for HF by 39%. All findings were statistically significant, were consistent in all predefined subgroups, were independent of coexistent β-blocker therapy, and were seen in the setting of aggressive treatment of vascular disease, as determined by the high rate of antiplatelet (92%), antilipid (58%), and β-blocker (62%) usage.

No large-scale studies to date have demonstrated improvement in clinical outcomes from β-blockers specifically in patients with HF and preserved LVEF. However, as with ACE inhibitors, large subsets of this population fall into one or another category for which β-blockers have either proven beneficial or are highly likely to achieve clinical benefit.

In failing hearts, rapid rates are associated with progressively reduced contractile force and increased resting tension. The increased resting tension is related to incomplete relaxation due to incomplete reuptake of calcium to storage sites in the sarcoplasmic reticulum.⁹ In a noninvasive study of hypertrophic cardiomyopathy, β-adrenergic blocking drugs prolonged diastolic filling time, suggesting better LV filling.⁴³ In the presence of coronary artery disease, tachycardia is associated with a prompt increase in LV diastolic pressure.⁴⁴ Thus reducing the heart rate with β-adrenergic blocking drugs should be beneficial for LV filling and a reduction in the LV end-diastolic pressure. Furthermore, retrospective studies have suggested substantial benefit of adequate rate control on systolic function in patients with atrial fibrillation with a rapid ventricular response.^45,46 Patients with sinus tachycardia may benefit from a reduction in heart rate; however, because the tachycardia may reflect an inability to increase stroke volume, care must be taken in using β-blockade.

Although controlled clinical trial data are lacking, several properties of the calcium channel blocking drugs (eg, verapamil, diltiazem), suggest they may benefit patients with HF and preserved LVEF. Beyond these circumstances, calcium channel blockers are not routinely recommended, despite small studies showing hemodynamic benefit in select patients.

An important effect of these drugs is slowing heart rate. This effect should enhance calcium removal from the myocyte and calcium reuptake in the sarcoplasmic reticulum.^6,9,12 This should lower end-diastolic pressure¹² and improve passive ventricular filling.⁴⁷ Improved passive ventricular filling is associated with long-term improvement in exercise capacity in patients with hypertrophic obstructive cardiomyopathy, a clinical condition which, like HF with preserved LVEF, may be associated with significant abnormalities in myocardial relaxation.⁴⁷ Numerous studies have shown benefit from verapamil or diltiazem in chronic stable angina pectoris, although the patients likely did not have HF with preserved EF.⁴⁸ Verapamil has been shown to acutely reduce arterial stiffness in elderly normal subjects. The improvement is due to improved arterioventricular interaction, and this reduction in arterial stiffness has been related to improved exercise performance.⁴⁹

In patients with atrial fibrillation or flutter who remain symptomatic after adequate rate control, it is reasonable to consider restoration of sinus rhythm. Because studies comparing rhythm control to rate control in patients with atrial fibrillation have generally excluded symptomatic patients, there are no randomized clinical trials for guidance. Nevertheless, retrospective evaluation of studies of patients with HF suggest that in the subset of patients with atrial fibrillation both amiodarone and dofetilide increased conversion to sinus rhythm and maintenance of sinus rhythm.^50-52

These trials also demonstrated the safety of these drugs in patients with HF. Early experience suggests that catheter ablation of atrial fibrillation may also be considered in patients with HF to improve symptoms.⁵³