Introduction

Heart failure is characterized by a failure to pump an adequate volume of blood to meet the metabolic needs of the body. Systolic failure deals with expulsion of blood from the heart, and diastolic failure is defined as an abnormality in relaxation leading to impaired filling. This chapter, because of limited data on isolated diastolic heart failure, will provide only a comprehensive discussion of systolic heart failure, with the consideration that most patients with systolic dysfunction have concomitant diastolic dysfunction.

The etiologies of systolic heart failure are multi-factorial and include associated coronary artery disease, valve pathology, primary myopathy (toxins, infection, etc.), and related tachyarrhythmias.

Both atrial and ventricular arrhythmias can not only lead to heart failure but can also occur as a consequence of heart failure. Arrhythmias commonly seen in heart failure include sinus tachycardia, atrial fibrillation (AF), premature ventricular contractions (PVCs), nonsustained ventricular tachycardia (NSVT), sustained ventricular tachycardia (VT), torsades de pointes (TdP), and ventricular fibrillation (VF) (Box 56-1).

Mechanisms

The mechanisms of arrhythmias in heart failure include re-entry, triggered activity, abnormal automaticity, and altered stretch (Box 56-2).

Re-entry is often seen in heart failure and may be related to conduction slowing caused by functional or anatomic barriers. The underlying mechanisms include reduced myocardial excitability and increases in resistance.¹ Moreover, downregulation of the gap junction protein connexin 43 promotes cell-to-cell uncoupling, a milieu for re-entrant arrhythmia.²

Patients with heart failure often develop left ventricular hypertrophy, a compensatory mechanism for inadequate cardiac output. Left ventricular hypertrophy is associated with increased action potential duration that provides a potential for increased dispersion of repolarization leading to TdP or VF as a result of triggered activity, especially early afterdepolarization.³ Also of note, because of the downregulation of potassium (K⁺) currents and increased late sodium (Na⁺) current, the action potential duration is further prolonged, with a greater propensity for triggered activity.⁴ Increases in intracellular calcium (Ca²⁺) from enhanced sympathetic stimulation from the heart failure itself or from the use of digitalis may cause triggered arrhythmias by inducing delayed afterdepolarization.^5,6

Abnormal automaticity stems from ischemia and is related to abnormal Ca²⁺ handling, leading to non–re-entrant arrhythmias.⁶

Stretch-activated channels may play an important role in arrhythmia genesis. This observation has been noted mostly in animal studies. Stretch affects both conduction and refractoriness. Any such alterations may predispose to arrhythmias in patients with heart failure. Electromechanical feedback can lead to electrical remodeling with changes in conduction and repolarization, creating heterogeneity and dispersion, with enhancement of atrial and ventricular arrhythmias.^7,8

A reduction in the sinoatrial node I_f current (“funny” current) and its hyperpolarizing cyclic nucleotide (HCN) channel gene leads to sinus bradycardia from sinoatrial node dysfunction. In heart failure, however, increases may occur in atrial and ventricular tissue I_f currents that may lead to arrhythmias.⁹

Atrial Fibrillation

AF is quite commonly seen in patients with heart failure (30%). The tachycardia itself may predispose to heart failure but may occur as a result of heart failure. The risk of emboli is related to etiology and the CHADS 2 (Cardiac Failure, Hypertension, Age, Diabetes, Stroke [Doubled]) score. Valvular AF carries a higher risk of stroke. Compared with sinus rhythm, after adjustments for comorbidities, AF does not seem to have an independent mortality risk in heart failure.¹⁰ Mechanisms of AF in heart failure, which are complex and multi-factorial, include re-entry, triggered activity, automaticity, and activation of stretch receptors. In addition, atrial interstitial fibrosis, increased collagen synthesis, and altered connexin expression predispose to heterogeneities and electrical uncoupling and eventually AF. Of note, experimental heart failure in dogs has been shown to promote AF by causing fibrosis, interfering with local conduction, and not by altering atrial refractory period, refractoriness heterogeneity, or conduction velocity as seen with rapid atrial pacing.¹¹

Ventricular Arrhythmias

PVCs are very common in patients with heart failure, and in those with more than 10 per hour, the incidence of NSVT is about 90%.¹² In patients with prior myocardial infarction (MI), PVCs are associated with an increased risk of death, especially in those with left ventricular dysfunction. The presence of NSVT does not seem add any more risk over PVCs.¹³

Sustained VT and VF may account for 50% of all deaths in patients with heart failure and is often classified as sudden cardiac death (SCD). However, all SCDs may not be tachyarrhythmia related. Death from bradycardia or electromechnical dissociation is common, especially in severe heart failure.^14,15 Bundle branch re-entry is a form of sustained VT that may be reproduced in the electrophysiology laboratory and is important to recognize, since this arrhythmia can be successfully treated with ablation (see below).

TdP is seen with antiarrhythmic drug toxicity. Drugs that prolong the Q-T interval may predispose to this potentially lethal arrhythmia, especially in those with heart failure.

Evaluation

Numerous tests have been used to evaluate the excessive risks of patients with heart failure (Box 56-3). The electrocardiogram is extremely useful in determining etiologies and important prognostic factors. Sinus bradycardia in response to β-blocker therapy has been shown to be beneficial in patients with heart failure. However, sinus bradycardia may be profibrillatory in the atrium because of increased dispersion (atrial torsades) or in the ventricle, especially in the presence of class III (K⁺ channel blockers) antiarrhythmic drugs (see below).¹⁶ Wide QRS or bundle branch block is an independent risk factor for premature death, that is, independent of ejection fraction (EF).^17,18 A fragmented QRS has been shown to be predictive of mortality and SCD in both ischemic and nonischemic cardiomyopathy. An acquired prolonged Q-T interval is also quite important and may be a harbinger of TdP, especially when associated with K⁺ channel blockers, bradycardia, female gender, and electrolyte imbalance.

T-wave alternans, that is, a beat-to-beat variation in T-wave morphology, although subtle, has been associated with potentially lethal arrhythmias. A more sensitive technique is microvolt T-wave alternans (MTWA). A positive test carries a poor prognosis compared with a negative test, as shown in Multicenter Automatic Defibrillator Implantation Trial II (MADIT II), in which patients with prior MI and left ventricular dysfunction were randomized to implantable cardioverter-defibrillator (ICD) or not.¹⁹ However, in the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) substudy, a positive MTWA was not predictive of excess mortality in patients with heart failure.²⁰ Of note, a wide QRS and the use of β-blockers may affect the results. It appears that MTWA testing should not be used to make decisions on implantation of an ICD.

Baroreflex sensitivity (BRS), a measure of autonomic nervous system activity that is independent of β-blocker usage, has been used to identify patients at high risk, especially those with previous MI. A low BRS is associated with a poor outcome. In the Autonomic Tone and Reflexes after Myocardial Infarction (ATRAMI) trial, a depressed BRS predicted cardiac mortality after MI.²¹

The role of the electrophysiology study (EPS) in nonischemic cardiomyopathy has not been established, and the chance of inducing a sustained ventricular arrhythmia is quite low.²² VT caused by bundle branch re-entry is sometimes seen in patients with nonischemic cardiomyopathy, and ablation of the right bundle, if electrophysiologically induced, is curative.²³ Patients with ischemic cardiomyopathy and NSVT in whom sustained VT can be provoked at EPS carry a high risk of SCD compared with those in whom the arrhythmia is not inducible. However, because of a low EF, noninducibility still carries a substantial risk.²⁴ Thus the role of EPS is limited. Left ventricular ejection fraction is the most powerful predictor of outcome and is used frequently to guide therapy, especially nonpharmacologic therapy.

Therapy

Box 56-4 provides a brief outline of the various treatment modalities for the patient with heart failure and left ventricular dysfunction.