Jonathan J. Langberg MD1
David B. DeLurgio MD2
1Professor of Medicine, Cardiology Division, Emory University School of Medicine, Director, Cardiac Electrophysiology Laboratory, Emory University Hospital
2Assistant Professor of Cardiac Electrophysiology, Emory University School of Medicine, Electrophysiologist, Emory Crawford Long Hospital
The authors serve as clinical investigators for Medtronic, Inc, Guidant Corporation, and St. Jude Medical, Inc.
Ventricular tachyarrhythmias characteristically are sudden in onset, unpredictable, and transitory. Consequently, their assessment and treatment present extraordinary challenges to the clinician. Moreover, the prognosis for patients with these arrhythmias is quite variable. In some patients, ventricular ectopic activity may be benign and without sequelae, but in other patients, comparable ectopy is a harbinger of ventricular fibrillation and sudden cardiac death.1 This chapter summarizes the practical aspects of evaluation and treatment of patients with ventricular arrhythmias.
Ventricular tachyarrhythmias are mediated by one of three basic mechanisms: reentry, abnormal automaticity, and triggering. Although causation cannot be directly determined in individual patients, experimental and clinical observations make it possible to infer the mechanism underlying many of the ventricular arrhythmia syndromes encountered in practice.
VENTRICULAR TACHYCARDIA CAUSED BY REENTRY
Reentrant arrhythmias (also called circus-movement tachycardias) are produced by a continuous circular or looping pattern of myocardial activation. Reentry can occur around lines of anatomic or functional block or occur as spinning wavefronts or rotors that lack a fixed anatomic path. When reentry occurs around lines of block, two features must be present for reentry to occur: (1) a barrier around which the wavefront circulates, either a fixed region of inexcitability caused by scarring or a dysfunctional region resulting from local refractoriness, and (2) unidirectional block at the entrance of the circuit. If activation spreads down both sides of the barrier, the impulses will collide distally and reentry will not occur; however, if propagation is blocked in one limb and proceeds in an anterograde direction over the other, the activation wavefront may be capable of retrograde invasion of the initially blocked pathway, thereby initiating sustained reentry.
In patients with structural heart disease, most symptomatic ventricular arrhythmias are mediated by reentry.2,3 Sustained monomorphic ventricular tachycardia often occurs after transmural myocardial infarction (MI). The arrhythmia usually arises in the border zone of the scar [see Figure 1]. The larger the extent of this heterogeneous border zone, the greater the probability of a circuit capable of mediating reentrant ventricular tachycardia. This is consistent with the observation that the risk of malignant ventricular arrhythmias is proportional to the volume of the scar and the severity of left ventricular dysfunction after MI.4
Figure 1. Reentrant Ventricular Tachycardia
Reentrant ventricular tachycardia usually arises as the result of reentry within the border zone of a myocardial infarction. This region consists of strands of viable myocytes interspersed with inexcitable fibrous tissue. Reentry begins when a wavefront of activation (1) encounters a bifurcation and blocks in one of the two pathways around an obstacle (2). The activation wavefront then conducts exclusively through the orthodromic pathway (3) and encounters a region of relatively slow conduction within the tachycardia circuit (4). The activation wavefront may exit from the tachycardia circuit at a site quite different from the entrance point (5). Although the anterograde limb of the circuit is initially refractory, it recovers excitability by the time it is depolarized by the reentrant wavefront (6). The activation wavefront reenters the orthodromic limb of the circuit, and the circus movement is established.
Ventricular fibrillation is also a reentrant phenomenon.5 Unlike ventricular tachycardia, during which a single activation wavefront circulates around a fixed barrier, ventricular fibrillation is caused by multiple simultaneous impulses that travel around functional barriers of refractory tissue, moving continuously throughout the myocardium to create very rapid, irregular, and ineffective activation. Alternatively, in some patients, ventricular fibrillation may be initiated by very early ectopic beats in the specialized conduction system.6
Like postinfarction arrhythmias, the ventricular tachycardia in patients with nonischemic cardiomyopathy is often the result of reentry in a zone of patchy fibrosis. However, in patients with left ventricular dilatation and slowed conduction in the specialized conduction system, the tachycardia may be mediated by bundle branch reentry: anterograde conduction over the right bundle branch, activation of the septum, and retrograde conduction over the left bundle branch [see Figure 2].7 Although an infrequent cause of ventricular tachycardia, bundle branch reentry is of interest to cardiac electrophysiologists because it can be cured by selective destruction of either the right or the left bundle branch by use of radiofrequency catheter ablation [see 1:VII Pacemaker Therapy].
Figure 2. Left Bundle Branch Disease
Ventricular tachycardia resulting from bundle branch reentry usually occurs in patients with dilated cardiomyopathy and left bundle branch disease. Anterograde activation proceeds down the right bundle branch in a retrograde direction, with the activation wavefront proceeding through the left bundle branch and the bundle of His and finally reentering the right bundle branch. Intracardiac recording during bundle branch reentry reveals retrograde activation of the left bundle branch shortly after the QRS complex, followed by activation of the His bundle and right bundle branch.
VENTRICULAR TACHYCARDIA MEDIATED BY ABNORMAL AUTOMATICITY
Normal ventricular myocytes maintain a steady transmembrane resting potential of -80 to -90 mV, depolarizing only when stimulated by an activation wavefront. Extrinsic factors, such as electrolyte imbalance and ischemia, or intrinsic disease may reduce the resting potential and produce simultaneous diastolic (phase 4) depolarization [see Figure 3].
Figure 3. Resting Transmembrane Potential
The resting transmembrane potential of the myocardial cell is created by active maintenance of sodium and potassium gradients. The cell is depolarized (phase 0) by an electrical stimulus that allows a sudden influx of sodium (Na+). Repolarization, phases 1 through 3, requires an early rapid chloride influx, a plateau phase mediated by calcium currents, and reestablishment of the resting transmembrane potential via potassium (K+) efflux. Between action potentials, the resting potential is designated as phase 4. In cells with automaticity, depolarization mediated by calcium (Ca2+) and Na+ currents may occur during phase 4, resulting in spontaneous generation of the next action potential. In normal ventricular myocytes, the resting potential during electrical diastole (phase 4) remains in the region of -80 to -90 mV. The rate of automatic firing is determined by the resting potential, the slope of phase 4, and the threshold potential.
Unlike reentry, which can usually be induced and terminated by premature beats, automatic rhythms tend not to be influenced by pacing. Changes in heart rate at the onset of ventricular tachycardia may also provide insight into the arrhythmia mechanism. Reentrant tachycardias are usually stable because of a fixed conduction time around the circuit. In contrast, automaticity often shows warm-up, with progressive acceleration during the first few seconds of the tachycardia.
Abnormal automaticity may play a role in a number of clinical arrhythmia syndromes. An accelerated idioventricular rhythm (60 to 100 beats/min) or episodes of slow ventricular tachycardia (100 to 140 beats/min) occur in approximately 20% of patients who are monitored after transmural MI.8 These slow-fast rhythms are probably the result of abnormal automaticity in ischemic Purkinje fibers.
More rapid ventricular tachycardia is also a frequent complication of acute ischemia, reperfusion, or both. These arrhythmias are often polymorphic, characterized by QRS complexes that change in amplitude and cycle length, with heart rates that may approach 300 beats/min. Abnormal automaticity in ischemic myocardium probably causes many of these episodes.
Ventricular tachycardia occasionally occurs in patients without apparent structural heart disease.9 This idiopathic arrhythmia generally originates in the right ventricular outflow tract, just beneath the pulmonary valve. A number of observations suggest that it, too, is sometimes mediated by abnormal automaticity. It can develop spontaneously in response to increased adrenergic tone and, as a rule, cannot be induced or terminated by pacing. It may occur as a pattern of recurrent short bursts of tachycardia interspersed with equally short interludes of sinus rhythm, a pattern more consistent with automaticity than reentry.10
VENTRICULAR TACHYCARDIA CAUSED BY TRIGGERING
Triggered activity, defined as premature activation caused by one or more preceding impulses, is the result of afterdepolarizations that occur either during (early afterdepolarization) or just after (delayed afterdepolarization) completion of the repolarization process [see Figure 4]. Factors that slow the heart rate tend to prolong the duration of depolarization, which is identified by a lengthened QT interval on the electrocardiogram, often sufficiently to bring early afterdepolarizations to threshold. Thus, triggered ventricular tachycardia that results from early after-depolarizations is characteristically bradycardia dependent or pause dependent.
Figure 4. Ventricular Tachycardia Caused by Triggering
In ventricular tachycardia caused by triggering, prolongation of the action potential (and the QT interval) results in depolarization during phase 3. Such early afterdepolarizations are manifested as positive deflections at the end of the phase 2 plateau or during the phase 3 rapid repolarization of the action potential. If this deflection exceeds the threshold potential, one or more triggered beats will occur. Bradycardia-dependent torsade de pointes is an example of an arrhythmia caused by early afterdepolarizations. The electrocardiogram of a patient with quinidine intoxication reveals an extrasystole and polymorphic ventricular tachycardia.
Early afterdepolarizations have been produced experimentally under a variety of conditions, including ischemia, hypo ka le mia, and antiarrhythmic drug toxicity. The arrhythmias seen in these studies are bradycardia dependent and, typically, are both rapid and polymorphic. Slowing of the tachycardia rate just before spontaneous termination is another characteristic feature of early afterdepolarization-mediated ventricular tachycardia.
Although it is difficult to prove, it seems likely that early afterdepolarizations mediate a variety of clinical arrhythmias. Prolongation of the QT interval—whether congenital or acquired as a result of drugs (class IA antiarrhythmic agents or, more commonly, other drugs such as haloperidol or erythromycin) or electrolyte depletion—increases the risk of a polymorphic ventricular tachycardia. As in the experimental situation, patients with QT prolongation tend to develop polymorphic ventricular tachy cardia as a result of slowing of the heart rate, heart rate pauses, or sudden surges in adrenergic tone. Unlike rhythms mediated by automaticity or reentry, ventricular tachycardia in the setting of QT prolongation is almost always polymorphic, sometimes with the twisting pattern that characterizes torsade de pointes.
Arrhythmias mediated by delayed afterdepolarization are distinctly different from those associated with early afterdepolarization and appear to be caused by abnormal accumulation and oscillation of cytosolic calcium concentration. The amplitude of these arrhythmias is augmented by acceleration rather than slowing of the heart rate. Delayed afterdepolarizations have been implicated in the genesis of ventricular tachycardia in patients with digitalis toxicity, and in some patients with ventricular tachycardia who have no apparent structural heart disease. Verapamil may be therapeutic in this subset of patients.11 Although these arrhythmias have been recorded from surviving Purkinje fibers and infarcted canine myocardium, their role in clinical arrhythmias during and after MI is less well established.
Delayed afterdepolarizations are induced at a critical heart rate range, which is patient specific, either spontaneously or during atrial or ventricular pacing. As with reentrant arrhythmias, tachycardia resulting from delayed afterdepolarizations is often terminated by overdrive pacing, although it will frequently persist for several cycles after cessation of pacing.
Asymptomatic Ventricular Ectopy
Ventricular ectopy is recorded in more than half of normal persons undergoing ambulatory electrocardiographic monitoring. Complex ectopy (multifocal premature ventricular complexes and nonsustained ventricular tachycardia) is less frequent but is still observed in 5% to 10% of healthy persons with no apparent heart disease.12
The prognostic significance of ventricular ectopy depends on the severity of left ventricular dysfunction. In the absence of structural heart disease, asymptomatic ventricular ectopic activity is benign, with no demonstrable risk of sudden death, even in the presence of ventricular tachycardia. In patients with structural heart disease, however, ventricular ectopic activity is associated with an increased risk of sudden cardiac death. This risk is markedly increased with progressive left ventricular dysfunction.13 For example, post-MI patients with a left ventricular ejection fraction (LVEF) greater than 40% who experience fewer than 10 ventricular premature complexes (VPCs) an hour after MI have a mortality of 5% to 7% a year. Those patients who experience more than 10 VPCs an hour, however, have a mortality of 12% to 18%. The combination of an LVEF of less than 40% and more than 10 VPCs an hour raises the annual mortality to between 27% and 40%.
The presence of frequent ventricular premature beats 7 to 10 days after MI is associated with a fivefold increase in the risk of symptomatic or fatal arrhythmias during follow-up.4 Because many patients with frequent ectopy do not develop malignant ventricular arrhythmias, the positive predictive accuracy of this finding is only 16%. Conversely, because the majority of patients without frequent ectopy remain free of fatal arrhythmias, its absence is associated with a negative predictive accuracy of 82%. The occurrence of nonsustained ventricular tachycardia (fewer than three consecutive rapid beats over a period of less than 30 seconds) during monitoring appears to confer an even greater risk than does the presence of frequent isolated ventricular premature beats.4,12,13
The association between ambient ventricular ectopy and the risk of arrhythmic death is less well established in patients with nonischemic (i.e., valvular, hypertensive, or idiopathic) cardio myopathy. However, most reports in the literature do suggest that the presence of high-grade ventricular arrhythmias, defined as multifocal VPCs or nonsustained ventricular tachycardia, confers an increased risk of sudden death that is independent of the severity of left ventricular dysfunction.14,15
Because the significance of ventricular ectopy depends on the degree of ventricular function impairment, cardiac imaging should be part of the initial evaluation. Echocardiography is the most versatile test; it provides information regarding regional wall motion abnormalities and valvular lesions as well as the LVEF. Radionuclide ventriculography also gives precise information regarding ejection fraction and may be of value in patients whose heart disease is already well characterized. If ventricular function is normal or close to normal, reassurance or treatment with beta blockers or calcium channel blockers to suppress bothersome symptoms is appropriate. In contrast, if patients have evidence of significant ventricular dysfunction or other significant structural heart disease such as hypertrophic cardiomyopathy, further evaluation and therapy may be appropriate. This evaluation may include additional tests that can help define the risk of a sustained arrhythmic event and the need for a prophylactic implantable cardioverter-defibrillator (ICD).
Signal-averaged electrocardiography may be useful for estimating risk in patients with heart disease and ventricular ectopy. This noninvasive test detects signals from areas of slow conduction in the arrhythmogenic regions on the periphery of an MI. The surface ECG is recorded for approximately 250 beats, and the signal is averaged by a computer and filtered, resulting in dramatic reduction of the signal-to-noise ratio. This allows detection of low-amplitude, high-frequency late potentials that result from the activation of zones of slow conduction just after the offset of the QRS complex.
Low-amplitude, high-frequency late potentials are recorded in about one third of patients after MI. These patients have a 20% incidence of life-threatening ventricular arrhythmias during the first year after infarction, compared with a 3% incidence in patients without late potentials.16 Signal-averaged ECG findings are independently predictive of adverse events after MI and provide additional information regarding risks in patients with frequent ventricular premature contractions and impaired left ventricular function. A limitation of signal-averaged ECG is that it cannot be used in patients with bundle branch block or atrial fibrillation.
MICROVOLT T WAVE ALTERNANS
Another screening test that may be useful for assessing risk of sudden cardiac death in patients with left ventricular dysfunction is microvolt T wave alternans.17 In this technique, signal processing is used to detect minute beat-to-beat variation in T wave amplitude that takes place during low-level exercise. Like signal-averaged ECG, microvolt T wave alternans has been approved by the Food and Drug Administration,18although currently it has little role in the selection of patients for ICDs. Unlike signal-averaged ECG, microvolt T wave alternans appears to have prognostic value in nonischemic as well as ischemic cardiomyopathy. A limitation of the test is that patients must be able to exercise for 5 minutes to a heart rate of at least 110 beats a minute.
Electrophysiologic study can be used to assess the inducibility of sustained ventricular arrhythmias in patients with structural heart disease.4 Electrode catheters are introduced percutaneously into the venous system, usually via the femoral vein, and advanced under fluoroscopic guidance into the right ventricle. Programmed electrical stimulation is performed in an attempt to elicit ventricular tachycardia or fibrillation. This usually consists of a drive train at a constant paced cycle length followed by one, two, or three extra stimuli (premature beats). The stimuli are introduced at progressively more premature coupling intervals until tachycardia is induced or the stimuli fail to capture as the result of local refractoriness [see 1:VII Pacemaker Therapy].
Programmed stimulation can induce sustained monomorphic ventricular tachycardia in about 20% of patients with reduced left ventricular function after MI and can induce ventricular fibrillation in an additional 10% to 15% of such patients. During follow-up, arrhythmic events occur in 5% of the noninducible patients, in 10% of patients with inducible ventricular fibrillation, and in 50% of patients with inducible ventricular tachycardia.
Although electrophysiologic study has reasonable sensitivity for prediction of subsequent arrhythmic events, the positive predictive value of the test is probably no better than that of the signal-averaged ECG, T wave alternans testing, or both, especially when such tests are combined with measurements of left ventricular systolic function and quantification of ambient ectopy. Electrophysiologic study is invasive and relatively expensive. Moreover, there is no evidence to suggest that treatment of this group of patients with antiarrhythmic drugs improves survival. Thus, it is difficult to justify routine electrophysiologic testing in asymptomatic patients after MI. The role of invasive electrophysiologic study for risk stratification in asymptomatic patients after MI has diminished with the increasing use of ICDs for primary prevention of sudden cardiac death.
Electrophysiologic testing is of uncertain value for stratification of risk in patients with nonischemic cardiomyopathy and asymptomatic ventricular ectopy. In this population, induction of sustained monomorphic ventricular tachycardia is infrequent and does not appear to be predictive of subsequent sudden cardiac death.
Currently, electrophysiologic studies are perhaps most useful in patients who have an LVEF of 30% to 40% and evidence of nonsustained ventricular tachycardia. In this select group, electrophysiologic testing can help determine whether the risk of sudden cardiac death is sufficiently high to merit prophylactic implantation of an ICD [see The Implantable Cardioverter-Defibrillator, below].
Syncope and Ventricular Arrhythmias
Syncope, defined as transient loss of consciousness, is a common phenomenon, accounting for about 3% of all emergency room visits.19Because the spells usually resolve by the time the patient is initially evaluated, determination of the cause of loss of consciousness is difficult but extremely important, because prognosis depends on the nature of the episode. If ventricular arrhythmias are detected during subsequent monitoring, additional evaluation should be undertaken to determine whether the syncope was produced by a paroxysm of ventricular tachycardia.
HISTORY AND PHYSICAL EXAMINATION
A thorough history may provide important clues to the diagnosis of ventricular tachycardia. The onset of syncope mediated by ventricular tachycardia is usually abrupt, with only a brief prodrome of light-headedness or no premonitory symptoms at all. The absence of rapid heartbeat does not exclude the diagnosis, because only about one half of patients with documented sustained ventricular tachycardia experience this symptom. The duration of unconsciousness is brief, rarely lasting longer than several minutes. Because of the abrupt onset, traumatic injury is common.
Spontaneous movements during syncope often cause confusion and misdiagnosis. Cerebral hypoperfusion from any cause, including ventricular tachycardia, may produce one or more clonic jerks of the extremities. However, syncopal episodes differ from seizure activity in three respects: (1) the movements in syncopal episodes are not reciprocating (tonic-clonic), (2) they are much briefer in duration, and (3) bladder or bowel incontinence rarely occurs.
Historical information regarding the patient's condition after awakening is frequently overlooked but may be very helpful. Patients typically recover quickly from ventricular tachycardia-mediated syncope. Postictal confusion lasting longer than 5 minutes suggests a grand mal event rather than an arrhythmic one. Similarly, persistent residual malaise, nausea, and weakness are characteristic of a faint produced by the vasodepressor syndrome rather than arrhythmic syncope.
Ventricular tachycardia of sufficient rate or duration to produce loss of consciousness is rare in patients with normal ventricular function. Thus, patients in whom ventricular arrhythmias are identified after a syncopal episode must be thoroughly evaluated for structural heart disease. The presence of severe left ventricular dysfunction in these patients is associated with an ominous prognosis.
Patients with coronary artery disease, syncope, or ventricular arrhythmias require evaluation of myocardial ischemia with a functional study (e.g., thallium scintigraphy), coronary angiography, or both, in addition to quantification of ventricular function. Acute ischemia may precipitate rapid ventricular tachycardia that is sufficient to cause loss of consciousness. In such cases, exercise treadmill testing may induce ventricular ectopy, thereby suggesting the diagnosis, especially if premonitory symptoms are reproduced.
On occasion, findings on a 12-lead ECG will suggest the cause of the loss of consciousness in a patient with unexplained syncope. A prolonged QT interval can indicate congenital long QT syndrome; ST segment elevation in lead V1 can indicate the Brugada syndrome; and a short QT interval may indicate short QT syndrome [see Heritable Ventricular Arrhythmias, below].
Signal-averaged electrocardiography plays a limited but important role in the evaluation of patients with syncope and ventricular arrhythmias. The positive predictive accuracy of this test is inadequate to confirm the diagnosis of an arrhythmic event. However, a negative result makes the possibility of sustained ventricular tachycardia unlikely enough that additional, more invasive studies are probably not justified.
Ambulatory electrocardiography is useful in selected patients with a history of syncope and ventricular arrhythmias. The yield of 24-hour or 48-hour Holter monitoring is low in patients with infrequent arrhythmic episodes, however. In such patients, a transtelephonic loop recorder is more likely to provide diagnostic information. This device is worn by the patient for 4 to 6 weeks, continuously recording and storing several minutes of the ECG in an endless loop. Immediately after presyncope or a syncopal spell, the patient presses the event button on the device to stop the recording and store the preceding ECG in memory. The output of the device is then transmitted over the telephone to a receiving station. This system has been shown to be more cost-effective than Holter monitoring and is preferable unless symptoms are present on a daily basis.
Electrophysiologic testing can be useful in determining whether an episode of loss of consciousness was produced by ventricular tachycardia.20 Assessment of sinus node function and atrioventricular conduction should be performed during electrophysiologic testing even when ventricular tachycardia is suspected, because episodic bradyarrhythmias may produce spells with very similar symptoms.
The induction of sustained monomorphic ventricular tachycar dia during programmed stimulation increases the probability that the patient's spontaneous episode was mediated by ventricular tachycardia and increases the likelihood that therapy will be effective. Several studies have shown a lower rate of recurrent syncope in patients whose therapy is based on results of electrophysiologic testing, compared with those in whom the study was unrevealing or for whom no effective treatment could be found.20,21
Evaluation of the Patient Rescued from Cardiac Arrest
In 80% to 90% of patients who develop out-of-hospital cardiac arrest, the precipitating event is either primary ventricular fibrillation or a rapid ventricular tachycardia that degenerates into ventricular fibrillation. Bradyarrhythmic events occur occasionally, but when asystole is recorded as the initial rhythm, it is usually indicative of a prolonged downtime interval and is associated with a very poor prognosis.
The majority of patients who sustain cardiac arrest have structural heart disease. In industrialized societies, this is most often the result of coronary atherosclerosis. Studies of both victims and survivors of cardiac arrest show significant coronary obstruction in 75% to 80% of patients. Unfortunately, sudden cardiac death is the initial manifestation of coronary artery disease in 10% to 20% of patients, making it the most common cause of mortality in adults younger than 65 years.22
Despite the close association between coronary artery disease and sudden cardiac death, acute MI is an infrequent cause of cardiac arrest. Only about 20% of patients rescued from an episode of ventricular fibrillation have evidence of an evolving MI during their subsequent hospitalization.23 The prognosis is favorable for cardiac arrest survivors in whom the event can be clearly linked to acute myocardial ischemia, with a recurrence rate of only 2% during the subsequent year. In contrast, patients with ventricular fibrillation not related to an ischemic event have an annual recurrence rate of greater than 20%, presumably because they have a chronic substrate capable of mediating malignant ventricular arrhythmias.22,23
All patients rescued from cardiac arrest require serial ECGs and enzyme measurements to determine whether the event was a consequence of acute MI. Coronary angiography should be performed in all patients as well, except those in whom the precipitating factor has already been unequivocally identified.
Laboratory evaluation of patients rescued from cardiac arrest should be directed at the identification of specific reversible causative factors. As in patients who have experienced syncope, the post resuscitation ECG may provide important information. A prolonged QT interval suggests the possibility of drug-induced torsade de pointes or the congenital long QT syndrome. A short PR interval and slurring of the QRS onset (a delta wave) are manifestations of the Wolff-Parkinson-White (WPW) syndrome [see 1:V Supraventricular Tachycardia]. Patients with WPW syndrome have an accessory connection linking the atrium and ventricle across either the mitral or the tricuspid annulus. A subset of patients with the WPW syndrome are capable of very rapid anterograde conduction over the accessory connection. If these patients develop atrial fibrillation, the ventricular response may be in excess of 300 beats/min and can degenerate into ventricular fibrillation.
The initial evaluation of serum electrolytes is sometimes revealing, because severe depletion of serum potassium, serum magnesium, or both may precipitate ventricular arrhythmias. Such depletions are characteristic of patients with chronic heart failure who are maintained on long-term diuretic therapy with inadequate electrolyte supplementation.
Electrophysiologic study was once an important part of the evaluation of cardiac arrest survivors in whom a reversible cause cannot be identified. With ICD therapy becoming commonplace for such patients, however, the usefulness of electrophysiologic testing has become limited to patients in whom the exact nature of the arrhythmia that precipitated the arrest remains uncertain. In a study of electrophysiology testing in 572 patients with ventricular fibrillation, ventricular tachycardia with syncope, or sustained ventricular tachycardia in the setting of left ventricular dysfunction, 67% of patients had inducible sustained ventricular tachycardia or ventricular fibrillation, but inducibility of these arrhythmias did not predict death or arrhythmia recurrence. These investigators concluded that electrophysiologic testing may not be worth the risks and costs of the procedure in this patient population, particularly in those patients likely to receive an ICD.24
Heritable Ventricular Arrhythmias
Alterations in the duration of the QT interval are most often acquired, typically from drugs.25 In rare cases, however, ventricular arrhythmias result from genetic disorders that alter ventricular repolarization. These disorders include long QT syndrome, Brugada syndrome, and short QT syndrome.
LONG QT SYNDROME
A familial disorder with distinct clinical features, the congenital long QT syndrome usually presents as syncope (or, in rare instances, as cardiac arrest) during childhood or the teenage years, mediated by recurrent bouts of rapid, polymorphic ventricular tachycardia. Many patients are incorrectly diagnosed with a grand mal seizure disorder. Loss of consciousness characteristically occurs with a sudden surge in adrenergic tone caused by abrupt physical, emotional, or auditory stimulation. There is often a family history of unexplained syncope or premature sudden cardiac death.
The hallmark of this disorder is abnormal prolongation of the QT interval on the ECG. Prolongation is present if the heart rate-corrected QT interval (QT/RR interval) exceeds 0.47 in children, 0.46 in men, or 0.48 in women. Other depolarization abnormalities are often present in the long QT syndrome. The T wave is flattened and may have a bifid, or double-hump, appearance. In addition, a prominent U wave may be seen. About one third of patients will have a resting heart rate of less than 60 beats/min.
Congenital long QT syndrome has two principal phenotypes. The originally described Jervell and Lange-Nielsen syndrome, an autosomal recessive disorder with associated deafness, has proved to be quite rare.26 The more common Romano-Ward syndrome is an autosomal dominant disorder and is not associated with hearing loss. Genomic studies in families with congenital long QT syndrome have shown that the disorder is produced by mutations of membrane ion channel proteins. To date, more than 300 of these mutations have been identified in seven genes, accounting for approximately 70% of affected patients.27 Interestingly, the different mutations seem to produce slightly different ECG appearances.
Evaluation of a patient for the long QT syndrome should include screening of all first-degree relatives. A careful history regarding unexplained syncope and a 12-lead ECG should be obtained. A point system has been developed that combines ECG findings, clinical history, and family history (e.g., unexplained sudden death at a young age in an immediate family member) into a score that indicates the likelihood of disease.28 Genetic testing for long QT syndrome is now commercially available and includes analysis of five major cardiac ion channel genes. The sensitivity of this test is approximately 70%, and its role in the management of affected patients and their families has not been established. Treatment of long QT syndrome is with beta blockers, ICD placement in high-risk patients, and left thoracic sympathectomy in selected cases.29 Treatment of asymptomatic family members should be considered if screening uncovers a prolonged QT interval.
Brugada syndrome is an inherited disorder that is manifested by syncope or sudden cardiac death. It is characterized by an ECG that shows an incomplete right bundle branch block and ST segment elevation in leads V1 through V3.30 However, these ECG findings also occur in many patients who do not have Brugada syndrome. Because of the low specificity of the ECG characteristics, the diagnosis should not be made on the basis of the ECG alone.31 A set of diagnostic criteria that includes history, ECG and electrophysiologic test results, and family history has been proposed.32 As with the long QT syndrome, genetic testing for Brugada syndrome is commercially available, but it is of uncertain utility. ICD placement is recommended for patients with Brugada syndrome who have experienced symptoms; recommendations for ICD placement are less well established for patients with Brugada syndrome who are asymptomatic or have inducible arrhythmias.
SHORT QT SYNDROME
Short QT syndrome is an inherited disorder characterized by a family history of sudden death (perhaps including sudden infant death syndrome), an abnormally short QT (QTc < 300 msec), and inducible ventricular fibrillation.33,34 The syndrome has been traced to mutations in the cardiac ion channel genes.35 Because short QT syndrome has only recently been recognized, its incidence remains uncertain. It is important to remember that electrolyte and drug effects that cause QT shortening (e.g., hypercalcemia and digitalis) need to be excluded before this diagnosis is entertained.
As a result of changes in the medical care system, more primary care practitioners bear direct responsibility for treatment decisions in patients with cardiac arrhythmias. The use of antiarrhythmic drugs in patients with ventricular arrhythmias presents a growing challenge, especially given that the medical literature contains reports of real and potential harm associated with the use of antiarrhythmic drugs.
CLASSIFICATION AND MECHANISMS OF ANTIARRHYTHMIC DRUGS
Antiarrhythmic drugs directly alter the electrophysiologic properties of myocardial cells. Therefore, an understanding of basic cellular electrophysiology is critical for an informed use of these compounds [see Figure 5].36
Figure 5. Electrophysiological Characteristics of Antiarrhythmic Drugs
The electrophysiologic hallmark of class I antiarrhythmic drugs is inhibition of the fast Na+ channel, which results in a decrease in the slope and amplitude of phase 0 of the cardiac action potential. Class IA agents (quinidine, procainamide, and disopyramide) also prolong the action potential duration, whereas class IB agents (lidocaine and mexiletine) may shorten the action potential duration, particularly in ischemic tissue. Class IC agents (flecainide and propafenone) have little effect on action potential duration.
The most widely accepted classification of antiarrhythmic drugs, originally proposed by Vaughan Williams in 1970, involves four main classes of drugs, with the first class further divided into three subgroups [see Table 1].37 This classification is based primarily on the ability of the drug to control arrhythmias by blocking ionic channels and currents. Few drugs demonstrate pure class effects, however, and other characteristics, such as influence of the drug on autonomic tone, contractility, and adverse effects, may be more important clinically and will be discussed as they pertain to individual drugs.
Table 1 Classification of Antiarrhythmic Drugs47
Class I agents inhibit the fast Na+ channel during depolarization (phase 0) of the action potential, with resultant decreases in depolarization rate and conduction velocity [see Figure 5]. Agents in class IA (quinidine, procainamide, disopyramide, and moricizine) significantly lengthen both the action potential duration and the effective refractory period (and therefore the QT interval) through a combination of the class I effect of Na+ channel inhibition and the lengthening of repolarization by K+ channel blockade, a class III effect.
Class IB drugs (lidocaine, mexiletine, and phenytoin) are less powerful Na+ channel blockers and, unlike class IA agents, shorten the action potential duration and refractory period in normal ventricular tissue, probably by inhibition of a background Na+ current during phase 3 of the action potential.38,39 In ischemic tissue, lidocaine may also block an adenosine triphosphate (ATP)-dependent K+ channel, thus preventing ischemically mediated shortening of depolarization.40
Class IC drugs (flecainide and propafenone), the most potent Na+ channel blockers, markedly decrease phase 0 depolarization rate and conduction velocity. Unlike other class I agents, they have little effect on the action potential duration and the effective refractory period in ventricular myocardial cells, but they do shorten the action potential of the Purkinje fibers.41,42 This inhomogeneity of depolarization combined with marked slowing of conduction may contribute to the proarrhythmic effects of this class of drugs.
Class II agents are the beta-adrenergic antagonists. The efficacy of these drugs in the reduction of arrhythmia-related morbidity and mortality has become more evident in recent years, but the precise ionic bases for their salutary effects have not been fully elucidated. Beta-adrenergic antagonism has been shown to decrease spontaneous phase 4 depolarization and, therefore, to decrease adrenergically mediated automaticity, an effect that may be of particular importance in the prevention of ventricular arrhythmias during ischemia and reperfusion. Beta blockade also results in the slowing of heart rate and decreased oxygen consumption, effects long recognized as desirable in MI patients.43 Effects on the cardiac action potential differ in atrial, ventricular, and specialized conduction tissues. For example, conduction velocity is slowed most profoundly in specialized conduction tissue, resulting in prolongation of the PR interval, whereas action potential duration in ventricular myocardium is generally not affected.
The primary actions of class III agents (amiodarone, sotalol, and dofetilide) are prolongation of depolarization, the action potential duration, and the effective refractory period by K+ channel blockade. These effects may prevent arrhythmias by decreasing the relative proportion of the cardiac cycle during which the myocardial cell is excitable and therefore susceptible to a triggering event. Reentrant tachycardias may be suppressed if the action potential duration becomes longer than the cycle length of the tachycardia circuit and if the leading edge of the wavefront suddenly impinges on inexcitable tissue. Class III agents have proven efficacy and an incidence of proarrhythmia lower than that seen with class IA agents.
Class IV agents act by inhibiting the inward slow Ca2+ current, which may contribute to late afterdepolarizations and therefore to ventricular tachycardia. These Ca2+ channel blockers reduce afterdepolarizations and are useful in the treatment of idiopathic ventricular tachycardia.11,44,45 They have no appreciable effect on conduction velocity or repolarization and tend to evoke sympathetic activation. Thus, their role in the treatment of ventricular tachycardia in the setting of structural heart disease is limited.
Antiarrhythmic drugs in clinical use today have activity in multiple classes. For example, in addition to its class III effects, amiodarone also exhibits prominent Na+ channel blockade (class I), beta blockade (class II), and Ca2+ channel blockade (class IV). Sotalol is a racemic mixture of d and l isomers, which have similar class III effects, whereas the l-isomer is essentially a beta blocker. d-Sotalol has been shown to increase mortality in patients with left ventricular dysfunction and recent MI.46 The lower incidence of proarrhythmia seen with amiodarone or racemic sotalol therapy may be related to beneficial class II effects.
Proarrhythmia refers to the worsening of an existing arrhythmia or the induction of a new one by an antiarrhythmic drug. Three types of proarrhythmia have been described: torsade de pointes (the most common), incessant ventricular tachycardia, and extremely wide complex ventricular rhythm.
Torsade de Pointes
Torsade de pointes is triggered by early afterdepolarizations in a setting of delayed repolarization and increased dispersion of refractoriness. Class IA and class III drugs, which prolong refractoriness (and thus the QT interval) by K+ channel blockade, provide the milieu for torsade de pointes. Drug-induced torsade de pointes is often pause dependent or bradycardia dependent, because the QT interval is longer at slower heart rates and after pauses. Exacerbating factors, such as hypokalemia, hypomagnesemia, and the concomitant use of other QT-prolonging drugs, are particularly important in this type of proarrhythmia.
Incessant Ventricular Tachycardia
Incessant ventricular tachycardia may be induced by drugs that markedly slow conduction (class IA and class IC) sufficiently to make the patient's own ventricular tachycardia continuous.47,48 The arrhythmia is generally slower because of the drug effect, but it may become resistant to drugs or cardioversion, with potentially disastrous consequences in the presence of hemodynamic instability. This proarrhythmia is rarely associated with class IB drugs, which affect weaker Na+ channel blockades.
Extremely Wide Complex Ventricular Rhythm
Extremely wide complex ventricular rhythm is usually associated with class IC agents, also in the setting of structural heart disease, and has been linked to excessive plasma drug levels or a sudden change in dose. The arrhythmia is not thought to represent a preexisting reentrant tachycardia and easily degenerates to ventricular fibrillation.
EFFICACY AND OUTCOMES OF ANTIARRHYTHMIC DRUG USE
Suppression of ambient ventricular ectopy by an antiarrhythmic agent does not prevent future life-threatening arrhythmias. In fact, patients effectively treated with class IC agents in the Cardiac Arrhythmia Suppression Trial (CAST) had a greater risk of sudden cardiac death than those who received placebo, a finding that underlines the proarrhythmic potential of these agents.49 Conversely, beta blockers, which typically do not suppress ambient ectopy, appear to reduce the risk of malignant ventricular arrhythmias. A retrospective analysis of the CAST data showed that mortality related to arrhythmias, as well as from all causes, was reduced in patients who received beta blockers. The Electrophysiologic Study versus Electrocardiographic Monitoring (ESVEM) trial compared seven antiarrhythmic drugs and found that the risk of arrhythmia recurrence and cardiac mortality was greater with the class I agents than with sotalol.
As mentioned, patients with a history of MI and ventricular arrhythmias have an increased risk of fatal arrhythmias during follow-up. Meta-analysis of 138 trials involving 98,000 patients showed increased mortality with class I drugs.50 Beta blockers have been conclusively associated with short-term and long-term survival in this population.51 Therefore, all such patients should receive a beta blocker unless it is specifically contraindicated. In contrast, evidence that class IA and class IC agents increase mortality suggests that these drugs should be avoided in MI patients. Class IV agents have shown neither benefit nor harm.
Amiodarone is not associated with a significant survival benefit in MI patients, nor does it seem to be associated with an increased risk of sudden death. For example, the randomized, double-blind, placebo-controlled Canadian Amiodarone Myocardial Infarction Arrhythmia Trial (CAMIAT) was conducted in 1,202 MI patients with frequent or repetitive ventricular premature depolarizations. Resuscitated ventricular fibrillation or arrhythmic death occurred in 6.9% of patients in the placebo group and in 4.5% of those in the amiodarone group.52
Treatment of ventricular arrhythmias in patients with chronic heart failure is particularly challenging. The presence of a reduced ejection fraction and ventricular ectopy significantly increases the risk of sudden death. No antiarrhythmic drug has been shown to produce a significant survival benefit in this population. The proarrhythmic and negative inotropic effects of class IA and class IC drugs preclude their use in these patients. Amiodarone, overall, appears to be neutral in its effects. The Survival Trial of Antiarrhythmic Therapy in Congestive Heart Failure did not show significantly greater improvement in survival in patients treated with amiodarone than in those who received placebo, despite an antiarrhythmic effect.15 Similarly, in the European Myocardial Infarct Amiodarone Trial (EMIAT), a randomized, double-blind, placebo-controlled trial conducted in 1,486 MI survivors with an LVEF of 40% or less, neither all-cause mortality nor cardiac mortality differed between the amiodarone and the placebo groups. The investigators noted, however, that the 35% risk reduction in arrhythmic deaths in the amiodarone group support the use of amiodarone in patients for whom antiarrhythmic therapy is indicated.53 Therefore, the only indication for the use of amiodarone in patients with chronic heart failure appears to be to suppress symptoms from frequent ectopy and nonsustained ventricular tachycardia. Improvement in survival requires ICD therapy.
SURGERY AND CATHETER ABLATION OF VENTRICULAR TACHYCARDIA
Surgical techniques for the treatment of ventricular tachycardia after MI were introduced in the late 1970s. However, these procedures are associated with relatively high perioperative mortality and require ventriculotomy, which can further compromise an already damaged ventricle. For these reasons, along with the increased simplicity of ICD implantation, surgical treatment is now rarely performed.
Radiofrequency catheter ablation has a role in selected patients with idiopathic ventricular tachycardia. Ablation is also useful for palliation in patients who have had an ICD implanted and are experiencing frequent shocks.
THE IMPLANTABLE CARDIOVERTER-DEFIBRILLATOR
The ICD automatically detects ventricular tachycardia or fibrillation and terminates the arrhythmia by overdrive pacing, high-energy shocks, or both. Since the first implantation of an ICD in a human, in 1980, the device has been utilized in hundreds of thousands of patients worldwide, and its use is growing exponentially.
All ICD systems contain three elements: the generator, rate-sensing leads, and electrodes to deliver high-energy shocks. In the early ICDs, defibrillating shocks were delivered via wire-mesh patch electrodes applied directly to the epicardial surface, and the generator was implanted subcutaneously in the abdomen. The implantation procedure required a thoracotomy and was associated with considerable morbidity and a perioperative mortality of 3% to 5%.54 Current ICD models use transvenous leads, and the generator is implanted in a subcutaneous pocket in the anterior chest wall [see Figure 6].55 ICD implantation is simple and safe, with a median duration of less than an hour and a median postoperative stay of 24 hours or less. The incidence of surgical complications is less than 2%—similar to that with routine pacemaker implantation.55 As with modern pacemakers, the current generation of ICDs are multiprogrammable, microprocessor-based devices capable of automatically detecting ventricular tachycardia or fibrillation on the basis of timing information. The heart rate and duration of a tachycardia episode that will trigger overdrive pacing or shock therapy can be programmed. Additional detection enhancements can be used to reduce the probability that inappropriate pacing or shock will be delivered during episodes of sinus tachycardia or atrial fibrillation that exceed the programmed rate cutoff. The device can also be programmed to initiate therapy only if the heart rate increases abruptly during one cycle and only if the rate variability during the episode is less than a specified amount.
Figure 6. Transvenous Cardioverter-Defibrillator
An implantable cardioverter-defibrillator (ICD) consists of a pulse generator and one or more leads for cardioversion and defibrillation. The pulse generator is usually installed in a subcutaneous pocket in the pectoral region. It comprises a battery, capacitors, memory chips, integrated circuits and microprocessors, and a telemetry module, which are sealed within a titanium casing. A transvenous defibrillating lead from the pulse generator is inserted into the subclavian vein and advanced into the apex of the right ventricle. When a persistent ventricular tachyarrhythmia with a rate faster than the programmed rate cutoff is detected by the rate-sensing electrode in the lead's tip, the device charges and delivers a high-voltage defibrillating shock. For this purpose, the shock coil in the right ventricle serves as the cathode, whereas the proximal shock coil in the superior vena cava portion of the lead, plus the metal casing of the generator, serve as the anode. In older ICD models, the metal casing alone serves as the anode.
The ICD's output can also be tailored to suit patients' individual needs. For patients with a history of primary ventricular fibrillation, the ICD is programmed to deliver high-energy shocks when it detects tachycardia. Patients with a history of stable monomorphic ventricular tachycardia may benefit from overdrive pace termination. Cardioverting shocks will be delivered only if the specified number of pacing trains fails to terminate or if pacing accelerates the arrhythmia. Because overdrive pacing is associated with little or no discomfort, the device may be considered in patients with recurrent episodes of tolerated ventricular tachycardia.
The ICD also functions as a ventricular demand pacemaker, obviating a second device in patients with symptomatic brady arrhythmias. This feature is also useful for prevention of the transitory bradycardia that sometimes occurs after delivery of a defibrillating shock.
The ICD has the capability of recording individual arrhythmia episodes. When tachycardia is detected, the device stores the electrograms in memory that can then be played back through the programmer at the time of a follow-up visit. This Holter function provides valuable diagnostic information regarding arrhythmia frequency, duration, rate, and response to therapy.
ICDs were initially used for secondary prevention in survivors of cardiac arrest and in patients with documented life-threatening ventricular arrhythmias. Three large randomized trials have compared ICD therapy with pharmacologic treatment for the prevention of death in survivors of ventricular fibrillation or sustained ventricular tachycardia: the Antiarrhythmics vs Implantable Defibrillator (AVID) study,56 the Cardiac Arrest Study Hamburg (CASH),57 and the Canadian Implantable Defibrillator Study (CIDS).58 A meta-analysis of the three trials showed consistent benefit from ICDs: patients who received ICDs had a significant reduction in death from any cause, with a summary hazard ratio (ICD:amiodarone) of 0.72; for arrhythmic death, the hazard ratio was 0.50.59 Furthermore, 11-year follow-up of a subset of CIDS patients found that the benefit of the ICD over amiodarone increases with time; eventually, most amiodarone-treated patients develop side effects, experience recurrences of arrhythmia, or die.60
More recently, ICDs have been used for the primary prevention of sudden death. The first Multicenter Automatic Defibrillator Implantation Trial (MADIT I) compared ICD therapy with conventional medical therapy in MI patients with reduced ejection fraction, nonsustained ventricular tachycardia, and inducible nonsuppressible ventricular tachycardia on electrophysiologic testing. MADIT I showed that compared with conventional therapy, ICD therapy saved lives, with an ICD to non-ICD hazard ratio of 0.46.61 The magnitude of the survival benefit increased with the severity of cardiac dysfunction.62
To study the role of ICDs in primary prevention, MADIT II enrolled MI patients with advanced left ventricular dysfunction (LVEF, 30% or less) who did not necessarily have manifest or inducible ventricular tachycardia. ICD implantation also increased survival in this population: over 20 months of follow-up, the ICD to non-ICD hazard ratio for death from any cause was 0.69.63
Additional trials have been performed to determine whether prophylactic ICD implantation is beneficial in all patients with chronic heart failure of any cause, ischemic or nonischemic. In the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT), patients were evenly divided between those with ischemic and those with nonischemic cardiomyopathy. Patients who were receiving conventional treatment for heart failure were randomly assigned to supplemental therapy with amiodarone, placebo, or an ICD. Amiodarone did not increase survival, but simple, shock-only ICDs decreased mortality by 23%. The protective effect of the device was independent of the cause of the heart failure.64The Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure (COMPANION) trial was designed to compare medical therapy with a biventricular pacemaker and with a ventricular ICD in patients with advanced chronic heart failure and a wide QRS. Cardiac resynchronization with a biventricular pacemaker improved outcome, compared with medical therapy. Compared with biventricular pacing, biventricular ICD had an additional 21% survival benefit.65
Indications for ICD Implantation
Guidelines for the selection of patients for ICD implantation have been developed by the American College of Cardiology, the American Heart Association, and the National Association for Sport & Physical Education [see Table 2].66 Since the 2002 revision of these guidelines, the indications for ICD implantation have expanded to include patients with nonischemic cardiomyopathy and an LVEF of less than 30%, even in the absence of symptomatic arrhythmias.64
Table 2 Recommendations for Implantable Cardioverter-Defibrillator Therapy66
AUTOMATED EXTERNAL DEFIBRILLATORS
An automated external defibrillator (AED) is a compact, easily portable device that can automatically analyze a patient's cardiac rhythm and, if it detects ventricular fibrillation, direct the rescuer to apply a shock. AEDs require minimal training to operate and are achieving widespread distribution.
AEDs can now be found in many public places, such as airports, stadiums, casinos, and large office buildings. Preliminary data suggest that these devices may confer a survival benefit,67 although cost-effectiveness is difficult to calculate. It seems safe to say that the availability of public-access AEDs will result in increased numbers of patients successfully resuscitated from cardiac arrest, who will then require follow-up treatment.
The FDA has approved several AED models for consumer use in the home, without a prescription, and these devices are now being marketed directly to the public for this purpose. The utility of home AEDs is uncertain, but the patients for whom these devices should be considered are those who meet the criteria for prophylactic ICD therapy but either have declined the implantation procedure or have comorbidities that make the implantation procedure inadvisable. The cost-effectiveness of these devices will be difficult to measure, and the potential medicolegal liability issues involved may be complex.
Wearable Automatic Defibrillators
An automatic defibrillator that is worn as a vest has been approved by the FDA.68 This device is typically worn by patients who are awaiting heart transplants or who recently experienced an MI or underwent coronary revascularization. At our institution, we have used the device to provide temporary prophylaxis for a patient who required removal of an ICD because of site infection.
Figures 1, 2, and 7 Joseph Bloch, CMI.
Figures 3, 4, 5, and 6 Marcia Kammerer.
Editors: Dale, David C.; Federman, Daniel D.