- Email: [email protected]
Indications for antiarrhythmic suppression of ventricular arrhythmias: A definition of life-threatening ventricular arrhythmias
Indications for Antiarrhythmic Suppression of Ventricular Arrhythmias: A Definition of We-Threatening Ventricular Arrhythmias Joel Morganroth,
A surprtsing Rndkg of the Cardiac Arrhythmia SuppressionTrtal(CASl),reportedin 1999,isthat well-tolerated and effecttve antkrrhythmk drugs aswciated with an increase in mormayakobe tality due to arrhythmia. Consequently, attention has been focused on the importance of the beneftt-versus-risk asseswwM of such therapy. The benefKs ofantiarrhythmk therapy are reduction or ellmlnation of arrhythmta-caused symptoms (both hemodywmk and nonhemodynamic) and of the associated risk of death. The risks of such treatment include not only noncardiac adverse effects and organ toxkfty, but also early card& effects (proarrhythmia, heart failure, and conduction defects), as well as the newty recognized potentlal for late proarrhythmia or late arrhythmk death. Unfortunately, as the potential benefits of antiarrhythmk therapy increase in patients with poorer left ventrkular function (owing to their being at greater risk for sudden death), the effectlveness of suppression decreases and the incidence of Itfe-threatenkg complkatkns increases. The impact of this benefK4sk profile is that the indkation for most currently approved antiawhythmic drugs needs to be limited to those patients wtth defbtite life-threatening ventrkular arrhythmias that take the form ofsustainedventrkular tachyawhythmias with assockted hem+ dynamic symptoms. Uroadsnia this indication to indude patients with probable Me-threatenlng or even posskle life-threatenhtg arrhythmias must await the avallabiltty of drugs wtth better beneflt-rtsk proflies. (Am J Cardkl1993;72%-‘IA)
From the Department of Medicine and the Philadelphia Heart Institute of the Presbyterian Medical Center, Philadelphia, Pennsylvania. Address for reprints: Joel Morganroth, MD, Research and Education Development, Presbyterian Medical Center, 39th and Market Streets, Philadelphia, Pennsylvania 19104.
MD
S
ince April 1989, when the interim report of the Cardiac Arrhythmia Suppression Trial (CAST) documented that well-tolerated effective antiarrhythmic drugs could increase late arrhythmic mortality, a more focused consideration of the risks versus benefits of such therapy is essential.’ Before that time, physicians frequently treated patients with ventricular arrhythmias that caused any symptoms (even trivial) or when the arrhythmias were merely prognostically significant.2 With the recognition of the previously unsuspected increased risk of late mortality with antiarrhythmic therapy, the Food and Drug Administration (FDA) since 1989 has approved new antiarrhythmic agents only for the treatment of “life-threatening ventricular arrhythmias or supraventricular arrhythmias.” The definition for arrhythmias especially ventricular arrhythmias, must take account of the balance between the benefits and the risks of such therapy. BENEFIT-MRSUS-RISK
ASSESSMENT
The potential benefits and risks from using antiarrhythmic therapy are detailed in Table I. Antiarrhythmic drugs may provide only 2 benefits to the patient, by reducing or eliminating symptoms caused by the arrhythmia and/or by reducing the associated increased arrhythmic mortality. The risks of the drug therapy include not only the common noncardiac adverse effects such as dizziness and nausea, but also the potential for organ toxic effects, such as agranulocytosis and hepatitis. Cardiac adverse effects include: (1) “early” proarrhythmia, which is defined as the provocation or exacerbation of arrhythmias due specifically to the antiarrhythmic drug; (2) induction or worsening of congestive heart failure; and (3) induction of important conduction defects, such as high-grade atrioventricular block or “sick sinus syndrome.” Since CAST, physicians have become more aware that the most important risk of antiarrhythmic therapy may be the potential for the antiarrhythmic drug to increase late arrhythmic mortality, A SYMPOSIUM:
ANTIARRHYTHMIC
DRUG THERAPY
3A
TABLE I Antiarrhythmic Assessment
1
Drug Therapy Benefit-Versus-Risk
-I
Benefit Reduces or eliminates symptoms and/or increased mortality Risks Noncardiac side effects Organ toxicity Cardiac side effects Early proarrhythmia Congestive heart failure Conduction defects Increased mortality (late proarrhythmia)
TABLE II Ventricular Arrhythmias: Types of Symptoms None Nonhemodynamic Palpitations Light-headedness, dizziness, some presyncope Anxiety (may be debilitating) Hemodynamically important Definite presyncope Angina/heartfailure Syncope Cardiac collapse
which has been termed the “late proarrhythmic effect.“3 The type of symptoms that may be produced by ventricular arrhythmias are presented in Table II. By using ambulatory long-term (Holter) electrocardiographic monitoring, clinicians have found that most spontaneous ventricular arrhythmias are not associated with any presenting symptoms.4 Frequently, palpitations and other such symptoms that are usually thought to be due to a ventricular arrhythmia are found to be present in the patient diary without any arrhythmia correlation on the Holter recording. Patients perceive symptoms from ventricular arrhythmias as a “tip of the iceberg” phenomenon, in that only a small percentage of the actual ventricular arrhythmias are detected symptomatically by the patient. Ventricular premature complexes and episodic nonsustained ventricular tachycardia (NSVT) are often associated with what we term nonhemodynamically important symptoms, such as palpitations, dizziness, and, in some cases, near “presyncope.” These symptoms are ill-defined, are intermittent, and are difficult for the patient and physician to quantify. What is important is that these symptoms are not due to a hemodynamic effect from the ventricular arrhythmia, which is defined as the induction by the arrhythmia of relative hypotension leading to cerebral hypoperfusion. Their physiologic basis is uncertain. These nuisance symptoms often cause anxiety, which to some patients may be so debilitating that their quality of life and u
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
72
functional productivity deteriorate markedly despite physician reassurance or even tranquilization. Some patients’ anxiety may be so great that they complain of feeling faint, which may be interpreted as “presyncope” despite the lack of any evidence of cerebral hypoperfusion. This must be distinguished from the important and uncommon occurrence of frank presyncope due to cerebral hypoperfusion that may lead to frank syncope due to the ventricular arrhythmia. If this hypotension is significant enough in patients with advanced heart disease, it may induce unstable angina, worsen or cause heart failure symptomatology, or cause hemodynamic instability resulting in frank syncope, cardiac collapse, seizures, or cardiac arrest. Obviously, compared with the presence of nuisance nonhemodynamic symptoms, the presence of hemodynamic symptoms substantially elevates the degree of benefit of eliminating a causal ventricular arrhythmia and allows the taking of more risk from therapy. The degree of benefit can be assessed as it relates to the risk of mortality from the arrhythmias. Such risk has been well described, and relates to the degree of left ventricular dysfunction and the frequency and type of ventricular arrhythmias, both of which act independently.5-7 It is clear that there is increasing risk of arrhythmic mortality in patients with worsening left ventricular dysfunction, as there is in patients with higher degrees of ventricular arrhythmias, particularly when repetitive forms such as NSVT and sustained ventricular tachycardia are present. Unfortunately, although the potential benefit from suppressing the causal ventricular arrhythmias increases as left ventricular function deteriorates, the effectiveness of the antiarrhythmic drug to suppress the arrhythmia decreases.8 The degree of the risks associated with the use of oral antiarrhythmic drugs for treating ventricular arrhythmias is detailed in Table III. These risks are often drug-specific, but in many cases can be related to a particular Vaughan Williams drug classification or subclassification and also to the degree of left ventricular function.9J0 The incidence of early proarrhythmia, worsening congestive heart failure, and the occurrence of important new conduction defects are often related to the degree of left ventricular function and/or the Vaughan Williams drug classification. For example, early proarrhythmia is particularly prevalent in patients treated with subclass Ic antiarrhythmic drugs when the left ventricular function deteriorates.rO Drugs such as disopyramide and flecainide are
AUGUST
12, 1993
particularly noted for their marked negative inotropit effects. Clinically relevant risks from antiarrhythmic drugs that are not often related to the severity of left ventricular dysfunction include the presence of noncardiac adverse effects such as dizziness and nausea. Frequently, the drug class can be used for predicting the prevalence of these nuisance effects, which are particularly noted with drugs in Vaughan Williams subclasses Ia and Ib and are particularly uncommon with drugs in subclass Ic, moricizine, l3 blockers (class II), and sotalol (a drug with class II and class III actions). Organ toxic effects such as agranulocytosis and hepatitis are particularly noted with drugs such as quinidine, procainamide, and tocainide, and most notably with amiodarone. The potential for causing increased late arrhythmic mortality has been best demonstrated in CAST, in which encainide and flecainide were shown to be associated with a 2- to 3-fold increase in the arrhythmic mortality rate over that of placebo despite the fact that these drugs were well tolerated and effective in suppressing ventricular arrhythmia in early open-label use.l This “late proarrhythmic mortality” appeared to be present continuously as a risk over the approximately 10 months of follow-up that was available before CAST was prematurely stopped. These results have prompted the reevaluation, using the metaanalysis technique on previously conducted and reported trials, of other antiarrhythmic drugs. Although the class Ic agents encainide and flecanide were shown in CAST to increase mortality, it cannot be assumed that other antiarrhythmics are free from this effect. Two meta-analyses have suggested that quinidine may increase late mortality,i1J2 and at least one meta-analysis13 has suggested this for mexiletine. One meta-analysis of quinidine, reported by Coplen et al,” included data from studies in patients with paroxysmal atria1 fibrillation. However, because of the limitations of this study, such as including noncardiovascular deaths in the quinidine-treated patients in the analysis, this meta-analysis should be interpreted with caution. The other meta-analysis of quinidine data was from studies of short-to-medium treatment in patients with benign or potentially lethal ventricular arrhythmias in whom mixed degrees of left ventricular dysfunction were present.12 Moricizine has recently been shown in CAST II to increase cardiac mortality, and the Data and Safety Monitoring Board has recommended early termination of the study. l4 At this time, l3 blockers are the only agents used in the treatment of arrhythmias that have generally shown an improved risk of
TABLEIII Degree and Type of Risk from Antiarrhythmic Drugs* Especrally High Risk Often related to left ventricular
Especially Low Risk
function and/or drug class
Early proarrhythmia
Class Ic, quinidine
Class lb, II,
Congestive heart failure
Disopyramide, flecainrde
Class lb, quinidine
Conduction defects
-
Not often related to left ventricular
function
Noncardiac adverse effects
Class la, lb
Class Ic, II, sotalol, moricizine
Organ toxicity
Quinidine, procainamide, tocainide, amiodarone
Class Ic, II, sotalol, moricizine
Late proarrhythmia
Encainide, flecainide, quinidine
Class II, sotalol
? mexiletine ? moricizine *Because no head-to-head comparisons of the risks of these agents have been performed, the information in thts table is suggested by the author on the basis of his ClInIcal experience.
mortality, i.e., a decreased risk of sudden and total cardiac mortality.15 Thus, it is clear that the potential for late proarrhythmic mortality is not principally related to the degree of left ventricular dysfunction (in CAST the risk of death imparted by encainide and flecainide was irrespective of left ventricular dysfunction). If late proarrhythmia is a drug-specific effect, then a mortality trial will be needed for each new antiarrhythmic drug to demonstrate whether its late proarrhythmic potential is high or low. To date, there is little evidence that a late proarrhythmic effect occurs in patients treated with class II or III antiarrhythmic drugs. IMPACT OF RISK-BENEFIT
ANALYSIS
When the degree of risks and benefits is analyzed, it is clear that as patients’ left ventricular function deteriorates, the potential benefit of depressing coexisting ventricular arrhythmias is higher, since the risk of mortality and hemodynamic symptoms is increased. Unfortunately, currently available antiarrhythmic drugs appear to be less efficacious in suppressing ventricular arrhythmias in patients with more-compromised left ventricular function and, most important, the risk of lifethreatening adverse effects is increased.8 Because no antiarrhythmic drug has shown a decrease in mortality, the type of ventricular arrhythmia considered to require therapy has shifted markedly after the benefit-risk assessment reported by CAST. Since CAST’ and CAST 1114demonstrated that A SYMPOSIUM:
ANTIARRHYTHMIC
DRUG THERAPY
bA
TMll!
IV Prognostic Classification of Ventricular Arrhythmias
Benign Small to no risk of increased mortality No hemodynamic symptoms Normal lefl ventricular function Potentially lethal or prognostically significant Graded increased risk of mortality due to left ventricular dysfunction and presence of VPCs and/or NSVT Hemodynamic symptoms absent Lethal or malignant Highest risk of sudden cardiac death Most severe left ventricular dysfunction Hemodynamic symptoms present NSVT = complexes.
nonsustained ventricular tachycardia:
VPCs = ventricular
premature
class Ic antiarrhythmic drugs can increase mortality, and other studies have suggested the same result for other classes of agents,“-l3 the only approved indication for antiarrhythmic drug therapy would be for the treatment of ventricular arrhythmias that produce hemodynamically important symptoms. In fact, since 1989, the FDA has approved new antiarrhythmic drugs (namely, propafenone, indecainide, and moricizine) only for the indication of “treatment of life-threatening ventricular arrhythmias” and this agency is actively seeking to limit the labeling of most other antiarrhythmic drugs to this sole indication when ventricular arrhythmias are treated. What, then, is the definition of “life-threatening ventricular arrhythmias”? TOWARD A DEFINITION OF LIFE-THREATENING vENTRlcUlARARRNYniMlAs
The prognostic classification systemof ventricular arrhythmias that has been used since 198416,17 is detailed in Table IV. The classification of ventricular arrhythmias as benign, potentially lethal (prognostically significant), or lethal (malignant) was based primarily on the risk of sudden cardiac death associated with the ventricular arrhythmia and depended less on the actual form of the ventricular arrhythmia. Patients could have ventricular tachycardia, but if it did not produce hemodynamic consequences and was associated with a normal left ventricle, then there would be no evidence of any major increase in the risk of mortality and this arrhythmia would be considered benign.18 Thus, there would be no indication for drug therapy for this ventricular tachycardia, since no benefit could be expected in that there were no associated symptoms and no important increase in the risk of mortality. At the other end of the spectrum are patients with malignant or lethal ventricular arrhythmias, which are defined as arrhythmias that have hemodynamic consequences in patients in 6A
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
72
whom the risk of sudden cardiac death is highest (usually because these patients have the most severely depressed left ventricular function). The most complex part of the spectrum between benign and malignant ventricular arrhythmias comprises patients with so-called potentially lethal ventricular arrhythmias. These patients are a spectrum in themselves.At one end, they are little different from those in the benign ventricular arrhythmia group, other than their having some mild structural heart disease. The risk of sudden cardiac death is increased, however, because of the coexistence of left ventricular dysfunction and the presence of ventricular ectopic activity (ventricular premature complexes and/or NSVIJ5 At the other end of the potentially lethal ventricular arrhythmia spectrum are patients who are not much different from those in the malignant ventricular arrhythmia group save for their not having any arrhythmia-caused hemodynamic symptoms. These patients often have severe left ventricular dysfunction and may have frequent episodes of NSVT but no hemodynamic compromise. Obviously, patients with hemodynamic consequences from their ventricular arrhythmia would be considered to have “life-threatening” arrhythmia, and the type of ventricular arrhythmia usually is sustained ventricular tachycardia and/or ventricular fibrillation. Life-threatening ventricular arrhythmias can be organized into 2 general categories. The first we would call “immediate life-threatening ventricular arrhythmias,” in which the ventricular tachyarrhythmia with hemodynamic consequenceswas present in the patient at the time of physician intervention. Thus, in patients with ventricular fibrillation, torsades de pointes, or sustained monomorphicventricular tachycardia, and possibly in some patients with very severe underlying left ventricular dysfunction and incessant NSVT, the hemodynamic consequences may be such that immediate termination of the arrhythmia is a goal of therapy, in order to reestablish a compensated stable hemodynamic state. Intravenous antiarrhythmic drug therapy with lidocaine, procainamide, or bretylium or possibly with such nonapproved agents as amiodarone and the use of cardioverters or pacing technology are the main treatment options for the intensive care unit physician. In the second category, patients may have what can be termed “chronic life-threatening ventricular arrhythmias,” which we define as a paroxysmal arrhythmia in which the goal of antiarrhythmic drug therapy would be prophylactic, that is, to reduce the incidence or eliminate the occurrence
AUGUST
12, 1993
of the paroxysmal life-threatening ventricular arrhythmia. There is no question that life-threatening ventricular arrhythmia would be the proper designation for any ventricular arrhythmia that produced a hemodynamic consequence. This “definite” form of chronic life-threatening ventricular arrhythmia falls directly into the prognostic classification of malignant or lethal ventricular arrhythmia.16J7 It has recently been shown that the risk of mortality in patients with NSVT without immediate hemodynamic consequenceswho have poor left ventricular function and a coexisting arrhythmogenie substrate defined by the presence of a late potential detected by the signal-averaged electrocardiogram is comparable to the risk of death in those with malignant ventricular arrhythmias.t9 Perhaps the former group of patients might be loosely termed as having “probable or near” lifethreatening ventricular arrhythmia in light of their extremely high risk of mortality despite their not having hemodynamic symptoms. This type of patient would fit into the prognostic classification schema of those patients with the advanced or most severeform of “potentially 1ethal”ventricular arrhythmias. The classification of possible lifethreatening ventricular arrhythmias could include patients with NSVT in whom there is poor left ventricular function but not the presence of a late potential, since there is an increased risk of mortality in these patients. If antiarrhythmic drugs were available that were known to have low toxicity without evidence of late proarrhythmic effect, the inclusion of “probable” or even “possible” designations of life-threatening ventricular arrhythmias might be appropriate, but the long-term safety of the commonly used and currently available antiarrhythmic drugs in the United States has not been established, and therefore the use of these drugs should be restricted to those designations that are considered “definite life-threatening ventricular arrhythmias.” It seemsappropriate for now to limit the definition of “life-threatening ventricular arrhythmias” to only the “definite” type. Clearly, patients with NSVT and left ventricular function that is only mildly depressed or not depressed at all have no indication for antiarrhythmic drug therapy unless their symptomatology is extremely debilitating despite reassurance and possibly tranquilization. Antiarrhythmic drug therapy then might be indicated in such patients, but the drug selected for treatment should be limited to those agents that have demonstrated the least potential for organ toxic effects, late proarrhythmic effects, and the like.
Obviously, CAST has had a major influence on the risk-benefit assessmentand has greatly altered the threshold for the use of antiarrhythmic drugs. It is hoped that the results of other such trials will provide for another shift in the indications and thresholds for using antiarrhythmic drugs. REFERENCES l. The Cardiac Arrhythmia SuppressionTrial (CAST) Investigators. Prelinary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl .I Med 1989;321:406412. 2. Morganroth J, Bier JT Jr, Anderson JL. Treatment of ventricuhu arrhythmias by United States cardiologists: a survey before the Cardiac Arrhythmia SuppressionTrial resultswere available.Am J Cmtfid 1990,6549-48. 3. Morganroth J, Bigger JT Jr. Pharmacologic management of ventricular arrhythmias after the Cardiac Arrhythmia Suppression Trial. Am J Cat&l 1990;65:1497-1503. 4.Zehlii SM. Levine BJ, Michelson EL, Morganroth J. Cardiovascular complaints: correlation with cardiac arrhythmias on 24-hour electrocardiographic monitoring. Chesr 1980;78:456&2. 5. Calii RM, Burks JM, Behar VS, Margolis JR, Wagner GS. Relationships among ventricular arrhythmias, coronary artery disease,and angiographic and electrocardiographic indicators of myocardial fibrosis. C&&ion 1978;57:725732. 6. Bigger JT Jr, Fleiss JL, KIeiger R, Miier JP, Rolnitzky LM, the Multicenter Post-Infarction Research Group. The relationship among ventricular arrhythmias, left ventricular dysfunction, and mortality in the 2 years after myocardii infarction. Ciruhfon 1984$9ZO-258. 7. Mukharji J, Rude RE, Poole WK, Gustafson N, Thomas IJ Jr, StraussHW, Jaffe AS, MulIer JE, Roberts R, Raabe DS, Cmft CH, PassamaniE, Bratmwald E, WiIlerson JT, the MILIS Study Group. Rii factors for sudden death after acme myocardial infarction: two-year follow-up. Am J Cati 1984,54:3136. LPratt CM, Eaton T, Francis M, Wcolbert S, Mahmarian J, Roberts R, Young JB. The inverse relationship between baseline lefi ventricular ejection fraction and outcome of antiarrhythmic therapy: a dangerousimbalance in the risk-benefit ratio. Am Hec~l J 1989;118:433-440. 9. Morganroth J. Ambulatory eIectrocardiographic monitoring in the evaluation of new antiarrhythmic drugs.Cirrularion 1986;73(suppIII):II-92-B-97. 10. Morganroth J, Anderson JL, Gentzkow GD. Classification by type of ventricular arrhythmia predicts frequency of adversecardiac eventsfrom flecainide. JAm Cdl Cardid 1986;8:607-615. ll. Coplen SE, Antman EM, Berlin JA, Hewitt P, Chalmers TC. Efficacy and safety of quinidme therapy for maintenance of sinus rhythm after cardioversion: a meta-analysisof randomized control trials. Circulndon 1990;82:110& 1116. 82. Morganroth J, Gom JE. Quinidine-related mortality in the short-twnediumterm treatment of ventricular arrhythmias: a meta-anaiysis.CinrJotkm 1991;84: 1977-1983. U Anderson JL. Reassessmentof benefit-risk ratio and treatment algorithms for antiarrhythmic drug therapy after the Cardiac Arrhythmia Suppression Trial. J C/in Phannaco11990,30:981-989. l4. The Cardiac Arrhythmia SuppressionTrial II Investigators. Effect of the antiarrhythmic agent moricizine on smvival after myocardial infarction. N Engl J Med 1992;327:227-233.
lS. Yusuf S, Peto R, Lewis J, Collins R, Sleight P. Beta blockade during and after myocardii infarction: an overview of the randomized trials. Rag Can& vast Llis 1’%5;27z335-371.
l.6. Morganroth J. Premature ventricular complexes:diagnosisand indications for therapy.JAMA 19&1,252:673-676. 17. Bigger JT Jr. Identification of patients at high risk for sudden cardiac death.Am J Car&/ 19&1;54(suppl):3D-8D. 18. Kennedy HL Whitlock JA Sprague MK, Kennedy IJ, BuckinghamTA, Goldberg RJ. Long-term follow-up of asymptomatichealthy subjectswith frequent and complexventricular ectopy.N Engl JMed 1985;312:193197. is. Gomes JA, Winters SL, Stewart D, Horowitz S, Milner M, Barreca P. A new noninvasive Index to predict sustainedventricular tachycardia and sudden death in the first year after myocardial infarction: based on signal-averaged electrocardiogram,radionuclide ejection fraction and Hoher monitoring. J Am Cdl Cardid 1987$X4%357.
A SYMPOSIUM: ANTIARRHYTHMIC DRUG THERAPY
7A
A surprtsing Rndkg of the Cardiac Arrhythmia SuppressionTrtal(CASl),reportedin 1999,isthat well-tolerated and effecttve antkrrhythmk drugs aswciated with an increase in mormayakobe tality due to arrhythmia. Consequently, attention has been focused on the importance of the beneftt-versus-risk asseswwM of such therapy. The benefKs ofantiarrhythmk therapy are reduction or ellmlnation of arrhythmta-caused symptoms (both hemodywmk and nonhemodynamic) and of the associated risk of death. The risks of such treatment include not only noncardiac adverse effects and organ toxkfty, but also early card& effects (proarrhythmia, heart failure, and conduction defects), as well as the newty recognized potentlal for late proarrhythmia or late arrhythmk death. Unfortunately, as the potential benefits of antiarrhythmk therapy increase in patients with poorer left ventrkular function (owing to their being at greater risk for sudden death), the effectlveness of suppression decreases and the incidence of Itfe-threatenkg complkatkns increases. The impact of this benefK4sk profile is that the indkation for most currently approved antiawhythmic drugs needs to be limited to those patients wtth defbtite life-threatening ventrkular arrhythmias that take the form ofsustainedventrkular tachyawhythmias with assockted hem+ dynamic symptoms. Uroadsnia this indication to indude patients with probable Me-threatenlng or even posskle life-threatenhtg arrhythmias must await the avallabiltty of drugs wtth better beneflt-rtsk proflies. (Am J Cardkl1993;72%-‘IA)
From the Department of Medicine and the Philadelphia Heart Institute of the Presbyterian Medical Center, Philadelphia, Pennsylvania. Address for reprints: Joel Morganroth, MD, Research and Education Development, Presbyterian Medical Center, 39th and Market Streets, Philadelphia, Pennsylvania 19104.
MD
S
ince April 1989, when the interim report of the Cardiac Arrhythmia Suppression Trial (CAST) documented that well-tolerated effective antiarrhythmic drugs could increase late arrhythmic mortality, a more focused consideration of the risks versus benefits of such therapy is essential.’ Before that time, physicians frequently treated patients with ventricular arrhythmias that caused any symptoms (even trivial) or when the arrhythmias were merely prognostically significant.2 With the recognition of the previously unsuspected increased risk of late mortality with antiarrhythmic therapy, the Food and Drug Administration (FDA) since 1989 has approved new antiarrhythmic agents only for the treatment of “life-threatening ventricular arrhythmias or supraventricular arrhythmias.” The definition for arrhythmias especially ventricular arrhythmias, must take account of the balance between the benefits and the risks of such therapy. BENEFIT-MRSUS-RISK
ASSESSMENT
The potential benefits and risks from using antiarrhythmic therapy are detailed in Table I. Antiarrhythmic drugs may provide only 2 benefits to the patient, by reducing or eliminating symptoms caused by the arrhythmia and/or by reducing the associated increased arrhythmic mortality. The risks of the drug therapy include not only the common noncardiac adverse effects such as dizziness and nausea, but also the potential for organ toxic effects, such as agranulocytosis and hepatitis. Cardiac adverse effects include: (1) “early” proarrhythmia, which is defined as the provocation or exacerbation of arrhythmias due specifically to the antiarrhythmic drug; (2) induction or worsening of congestive heart failure; and (3) induction of important conduction defects, such as high-grade atrioventricular block or “sick sinus syndrome.” Since CAST, physicians have become more aware that the most important risk of antiarrhythmic therapy may be the potential for the antiarrhythmic drug to increase late arrhythmic mortality, A SYMPOSIUM:
ANTIARRHYTHMIC
DRUG THERAPY
3A
TABLE I Antiarrhythmic Assessment
1
Drug Therapy Benefit-Versus-Risk
-I
Benefit Reduces or eliminates symptoms and/or increased mortality Risks Noncardiac side effects Organ toxicity Cardiac side effects Early proarrhythmia Congestive heart failure Conduction defects Increased mortality (late proarrhythmia)
TABLE II Ventricular Arrhythmias: Types of Symptoms None Nonhemodynamic Palpitations Light-headedness, dizziness, some presyncope Anxiety (may be debilitating) Hemodynamically important Definite presyncope Angina/heartfailure Syncope Cardiac collapse
which has been termed the “late proarrhythmic effect.“3 The type of symptoms that may be produced by ventricular arrhythmias are presented in Table II. By using ambulatory long-term (Holter) electrocardiographic monitoring, clinicians have found that most spontaneous ventricular arrhythmias are not associated with any presenting symptoms.4 Frequently, palpitations and other such symptoms that are usually thought to be due to a ventricular arrhythmia are found to be present in the patient diary without any arrhythmia correlation on the Holter recording. Patients perceive symptoms from ventricular arrhythmias as a “tip of the iceberg” phenomenon, in that only a small percentage of the actual ventricular arrhythmias are detected symptomatically by the patient. Ventricular premature complexes and episodic nonsustained ventricular tachycardia (NSVT) are often associated with what we term nonhemodynamically important symptoms, such as palpitations, dizziness, and, in some cases, near “presyncope.” These symptoms are ill-defined, are intermittent, and are difficult for the patient and physician to quantify. What is important is that these symptoms are not due to a hemodynamic effect from the ventricular arrhythmia, which is defined as the induction by the arrhythmia of relative hypotension leading to cerebral hypoperfusion. Their physiologic basis is uncertain. These nuisance symptoms often cause anxiety, which to some patients may be so debilitating that their quality of life and u
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
72
functional productivity deteriorate markedly despite physician reassurance or even tranquilization. Some patients’ anxiety may be so great that they complain of feeling faint, which may be interpreted as “presyncope” despite the lack of any evidence of cerebral hypoperfusion. This must be distinguished from the important and uncommon occurrence of frank presyncope due to cerebral hypoperfusion that may lead to frank syncope due to the ventricular arrhythmia. If this hypotension is significant enough in patients with advanced heart disease, it may induce unstable angina, worsen or cause heart failure symptomatology, or cause hemodynamic instability resulting in frank syncope, cardiac collapse, seizures, or cardiac arrest. Obviously, compared with the presence of nuisance nonhemodynamic symptoms, the presence of hemodynamic symptoms substantially elevates the degree of benefit of eliminating a causal ventricular arrhythmia and allows the taking of more risk from therapy. The degree of benefit can be assessed as it relates to the risk of mortality from the arrhythmias. Such risk has been well described, and relates to the degree of left ventricular dysfunction and the frequency and type of ventricular arrhythmias, both of which act independently.5-7 It is clear that there is increasing risk of arrhythmic mortality in patients with worsening left ventricular dysfunction, as there is in patients with higher degrees of ventricular arrhythmias, particularly when repetitive forms such as NSVT and sustained ventricular tachycardia are present. Unfortunately, although the potential benefit from suppressing the causal ventricular arrhythmias increases as left ventricular function deteriorates, the effectiveness of the antiarrhythmic drug to suppress the arrhythmia decreases.8 The degree of the risks associated with the use of oral antiarrhythmic drugs for treating ventricular arrhythmias is detailed in Table III. These risks are often drug-specific, but in many cases can be related to a particular Vaughan Williams drug classification or subclassification and also to the degree of left ventricular function.9J0 The incidence of early proarrhythmia, worsening congestive heart failure, and the occurrence of important new conduction defects are often related to the degree of left ventricular function and/or the Vaughan Williams drug classification. For example, early proarrhythmia is particularly prevalent in patients treated with subclass Ic antiarrhythmic drugs when the left ventricular function deteriorates.rO Drugs such as disopyramide and flecainide are
AUGUST
12, 1993
particularly noted for their marked negative inotropit effects. Clinically relevant risks from antiarrhythmic drugs that are not often related to the severity of left ventricular dysfunction include the presence of noncardiac adverse effects such as dizziness and nausea. Frequently, the drug class can be used for predicting the prevalence of these nuisance effects, which are particularly noted with drugs in Vaughan Williams subclasses Ia and Ib and are particularly uncommon with drugs in subclass Ic, moricizine, l3 blockers (class II), and sotalol (a drug with class II and class III actions). Organ toxic effects such as agranulocytosis and hepatitis are particularly noted with drugs such as quinidine, procainamide, and tocainide, and most notably with amiodarone. The potential for causing increased late arrhythmic mortality has been best demonstrated in CAST, in which encainide and flecainide were shown to be associated with a 2- to 3-fold increase in the arrhythmic mortality rate over that of placebo despite the fact that these drugs were well tolerated and effective in suppressing ventricular arrhythmia in early open-label use.l This “late proarrhythmic mortality” appeared to be present continuously as a risk over the approximately 10 months of follow-up that was available before CAST was prematurely stopped. These results have prompted the reevaluation, using the metaanalysis technique on previously conducted and reported trials, of other antiarrhythmic drugs. Although the class Ic agents encainide and flecanide were shown in CAST to increase mortality, it cannot be assumed that other antiarrhythmics are free from this effect. Two meta-analyses have suggested that quinidine may increase late mortality,i1J2 and at least one meta-analysis13 has suggested this for mexiletine. One meta-analysis of quinidine, reported by Coplen et al,” included data from studies in patients with paroxysmal atria1 fibrillation. However, because of the limitations of this study, such as including noncardiovascular deaths in the quinidine-treated patients in the analysis, this meta-analysis should be interpreted with caution. The other meta-analysis of quinidine data was from studies of short-to-medium treatment in patients with benign or potentially lethal ventricular arrhythmias in whom mixed degrees of left ventricular dysfunction were present.12 Moricizine has recently been shown in CAST II to increase cardiac mortality, and the Data and Safety Monitoring Board has recommended early termination of the study. l4 At this time, l3 blockers are the only agents used in the treatment of arrhythmias that have generally shown an improved risk of
TABLEIII Degree and Type of Risk from Antiarrhythmic Drugs* Especrally High Risk Often related to left ventricular
Especially Low Risk
function and/or drug class
Early proarrhythmia
Class Ic, quinidine
Class lb, II,
Congestive heart failure
Disopyramide, flecainrde
Class lb, quinidine
Conduction defects
-
Not often related to left ventricular
function
Noncardiac adverse effects
Class la, lb
Class Ic, II, sotalol, moricizine
Organ toxicity
Quinidine, procainamide, tocainide, amiodarone
Class Ic, II, sotalol, moricizine
Late proarrhythmia
Encainide, flecainide, quinidine
Class II, sotalol
? mexiletine ? moricizine *Because no head-to-head comparisons of the risks of these agents have been performed, the information in thts table is suggested by the author on the basis of his ClInIcal experience.
mortality, i.e., a decreased risk of sudden and total cardiac mortality.15 Thus, it is clear that the potential for late proarrhythmic mortality is not principally related to the degree of left ventricular dysfunction (in CAST the risk of death imparted by encainide and flecainide was irrespective of left ventricular dysfunction). If late proarrhythmia is a drug-specific effect, then a mortality trial will be needed for each new antiarrhythmic drug to demonstrate whether its late proarrhythmic potential is high or low. To date, there is little evidence that a late proarrhythmic effect occurs in patients treated with class II or III antiarrhythmic drugs. IMPACT OF RISK-BENEFIT
ANALYSIS
When the degree of risks and benefits is analyzed, it is clear that as patients’ left ventricular function deteriorates, the potential benefit of depressing coexisting ventricular arrhythmias is higher, since the risk of mortality and hemodynamic symptoms is increased. Unfortunately, currently available antiarrhythmic drugs appear to be less efficacious in suppressing ventricular arrhythmias in patients with more-compromised left ventricular function and, most important, the risk of lifethreatening adverse effects is increased.8 Because no antiarrhythmic drug has shown a decrease in mortality, the type of ventricular arrhythmia considered to require therapy has shifted markedly after the benefit-risk assessment reported by CAST. Since CAST’ and CAST 1114demonstrated that A SYMPOSIUM:
ANTIARRHYTHMIC
DRUG THERAPY
bA
TMll!
IV Prognostic Classification of Ventricular Arrhythmias
Benign Small to no risk of increased mortality No hemodynamic symptoms Normal lefl ventricular function Potentially lethal or prognostically significant Graded increased risk of mortality due to left ventricular dysfunction and presence of VPCs and/or NSVT Hemodynamic symptoms absent Lethal or malignant Highest risk of sudden cardiac death Most severe left ventricular dysfunction Hemodynamic symptoms present NSVT = complexes.
nonsustained ventricular tachycardia:
VPCs = ventricular
premature
class Ic antiarrhythmic drugs can increase mortality, and other studies have suggested the same result for other classes of agents,“-l3 the only approved indication for antiarrhythmic drug therapy would be for the treatment of ventricular arrhythmias that produce hemodynamically important symptoms. In fact, since 1989, the FDA has approved new antiarrhythmic drugs (namely, propafenone, indecainide, and moricizine) only for the indication of “treatment of life-threatening ventricular arrhythmias” and this agency is actively seeking to limit the labeling of most other antiarrhythmic drugs to this sole indication when ventricular arrhythmias are treated. What, then, is the definition of “life-threatening ventricular arrhythmias”? TOWARD A DEFINITION OF LIFE-THREATENING vENTRlcUlARARRNYniMlAs
The prognostic classification systemof ventricular arrhythmias that has been used since 198416,17 is detailed in Table IV. The classification of ventricular arrhythmias as benign, potentially lethal (prognostically significant), or lethal (malignant) was based primarily on the risk of sudden cardiac death associated with the ventricular arrhythmia and depended less on the actual form of the ventricular arrhythmia. Patients could have ventricular tachycardia, but if it did not produce hemodynamic consequences and was associated with a normal left ventricle, then there would be no evidence of any major increase in the risk of mortality and this arrhythmia would be considered benign.18 Thus, there would be no indication for drug therapy for this ventricular tachycardia, since no benefit could be expected in that there were no associated symptoms and no important increase in the risk of mortality. At the other end of the spectrum are patients with malignant or lethal ventricular arrhythmias, which are defined as arrhythmias that have hemodynamic consequences in patients in 6A
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
72
whom the risk of sudden cardiac death is highest (usually because these patients have the most severely depressed left ventricular function). The most complex part of the spectrum between benign and malignant ventricular arrhythmias comprises patients with so-called potentially lethal ventricular arrhythmias. These patients are a spectrum in themselves.At one end, they are little different from those in the benign ventricular arrhythmia group, other than their having some mild structural heart disease. The risk of sudden cardiac death is increased, however, because of the coexistence of left ventricular dysfunction and the presence of ventricular ectopic activity (ventricular premature complexes and/or NSVIJ5 At the other end of the potentially lethal ventricular arrhythmia spectrum are patients who are not much different from those in the malignant ventricular arrhythmia group save for their not having any arrhythmia-caused hemodynamic symptoms. These patients often have severe left ventricular dysfunction and may have frequent episodes of NSVT but no hemodynamic compromise. Obviously, patients with hemodynamic consequences from their ventricular arrhythmia would be considered to have “life-threatening” arrhythmia, and the type of ventricular arrhythmia usually is sustained ventricular tachycardia and/or ventricular fibrillation. Life-threatening ventricular arrhythmias can be organized into 2 general categories. The first we would call “immediate life-threatening ventricular arrhythmias,” in which the ventricular tachyarrhythmia with hemodynamic consequenceswas present in the patient at the time of physician intervention. Thus, in patients with ventricular fibrillation, torsades de pointes, or sustained monomorphicventricular tachycardia, and possibly in some patients with very severe underlying left ventricular dysfunction and incessant NSVT, the hemodynamic consequences may be such that immediate termination of the arrhythmia is a goal of therapy, in order to reestablish a compensated stable hemodynamic state. Intravenous antiarrhythmic drug therapy with lidocaine, procainamide, or bretylium or possibly with such nonapproved agents as amiodarone and the use of cardioverters or pacing technology are the main treatment options for the intensive care unit physician. In the second category, patients may have what can be termed “chronic life-threatening ventricular arrhythmias,” which we define as a paroxysmal arrhythmia in which the goal of antiarrhythmic drug therapy would be prophylactic, that is, to reduce the incidence or eliminate the occurrence
AUGUST
12, 1993
of the paroxysmal life-threatening ventricular arrhythmia. There is no question that life-threatening ventricular arrhythmia would be the proper designation for any ventricular arrhythmia that produced a hemodynamic consequence. This “definite” form of chronic life-threatening ventricular arrhythmia falls directly into the prognostic classification of malignant or lethal ventricular arrhythmia.16J7 It has recently been shown that the risk of mortality in patients with NSVT without immediate hemodynamic consequenceswho have poor left ventricular function and a coexisting arrhythmogenie substrate defined by the presence of a late potential detected by the signal-averaged electrocardiogram is comparable to the risk of death in those with malignant ventricular arrhythmias.t9 Perhaps the former group of patients might be loosely termed as having “probable or near” lifethreatening ventricular arrhythmia in light of their extremely high risk of mortality despite their not having hemodynamic symptoms. This type of patient would fit into the prognostic classification schema of those patients with the advanced or most severeform of “potentially 1ethal”ventricular arrhythmias. The classification of possible lifethreatening ventricular arrhythmias could include patients with NSVT in whom there is poor left ventricular function but not the presence of a late potential, since there is an increased risk of mortality in these patients. If antiarrhythmic drugs were available that were known to have low toxicity without evidence of late proarrhythmic effect, the inclusion of “probable” or even “possible” designations of life-threatening ventricular arrhythmias might be appropriate, but the long-term safety of the commonly used and currently available antiarrhythmic drugs in the United States has not been established, and therefore the use of these drugs should be restricted to those designations that are considered “definite life-threatening ventricular arrhythmias.” It seemsappropriate for now to limit the definition of “life-threatening ventricular arrhythmias” to only the “definite” type. Clearly, patients with NSVT and left ventricular function that is only mildly depressed or not depressed at all have no indication for antiarrhythmic drug therapy unless their symptomatology is extremely debilitating despite reassurance and possibly tranquilization. Antiarrhythmic drug therapy then might be indicated in such patients, but the drug selected for treatment should be limited to those agents that have demonstrated the least potential for organ toxic effects, late proarrhythmic effects, and the like.
Obviously, CAST has had a major influence on the risk-benefit assessmentand has greatly altered the threshold for the use of antiarrhythmic drugs. It is hoped that the results of other such trials will provide for another shift in the indications and thresholds for using antiarrhythmic drugs. REFERENCES l. The Cardiac Arrhythmia SuppressionTrial (CAST) Investigators. Prelinary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl .I Med 1989;321:406412. 2. Morganroth J, Bier JT Jr, Anderson JL. Treatment of ventricuhu arrhythmias by United States cardiologists: a survey before the Cardiac Arrhythmia SuppressionTrial resultswere available.Am J Cmtfid 1990,6549-48. 3. Morganroth J, Bigger JT Jr. Pharmacologic management of ventricular arrhythmias after the Cardiac Arrhythmia Suppression Trial. Am J Cat&l 1990;65:1497-1503. 4.Zehlii SM. Levine BJ, Michelson EL, Morganroth J. Cardiovascular complaints: correlation with cardiac arrhythmias on 24-hour electrocardiographic monitoring. Chesr 1980;78:456&2. 5. Calii RM, Burks JM, Behar VS, Margolis JR, Wagner GS. Relationships among ventricular arrhythmias, coronary artery disease,and angiographic and electrocardiographic indicators of myocardial fibrosis. C&&ion 1978;57:725732. 6. Bigger JT Jr, Fleiss JL, KIeiger R, Miier JP, Rolnitzky LM, the Multicenter Post-Infarction Research Group. The relationship among ventricular arrhythmias, left ventricular dysfunction, and mortality in the 2 years after myocardii infarction. Ciruhfon 1984$9ZO-258. 7. Mukharji J, Rude RE, Poole WK, Gustafson N, Thomas IJ Jr, StraussHW, Jaffe AS, MulIer JE, Roberts R, Raabe DS, Cmft CH, PassamaniE, Bratmwald E, WiIlerson JT, the MILIS Study Group. Rii factors for sudden death after acme myocardial infarction: two-year follow-up. Am J Cati 1984,54:3136. LPratt CM, Eaton T, Francis M, Wcolbert S, Mahmarian J, Roberts R, Young JB. The inverse relationship between baseline lefi ventricular ejection fraction and outcome of antiarrhythmic therapy: a dangerousimbalance in the risk-benefit ratio. Am Hec~l J 1989;118:433-440. 9. Morganroth J. Ambulatory eIectrocardiographic monitoring in the evaluation of new antiarrhythmic drugs.Cirrularion 1986;73(suppIII):II-92-B-97. 10. Morganroth J, Anderson JL, Gentzkow GD. Classification by type of ventricular arrhythmia predicts frequency of adversecardiac eventsfrom flecainide. JAm Cdl Cardid 1986;8:607-615. ll. Coplen SE, Antman EM, Berlin JA, Hewitt P, Chalmers TC. Efficacy and safety of quinidme therapy for maintenance of sinus rhythm after cardioversion: a meta-analysisof randomized control trials. Circulndon 1990;82:110& 1116. 82. Morganroth J, Gom JE. Quinidine-related mortality in the short-twnediumterm treatment of ventricular arrhythmias: a meta-anaiysis.CinrJotkm 1991;84: 1977-1983. U Anderson JL. Reassessmentof benefit-risk ratio and treatment algorithms for antiarrhythmic drug therapy after the Cardiac Arrhythmia Suppression Trial. J C/in Phannaco11990,30:981-989. l4. The Cardiac Arrhythmia SuppressionTrial II Investigators. Effect of the antiarrhythmic agent moricizine on smvival after myocardial infarction. N Engl J Med 1992;327:227-233.
lS. Yusuf S, Peto R, Lewis J, Collins R, Sleight P. Beta blockade during and after myocardii infarction: an overview of the randomized trials. Rag Can& vast Llis 1’%5;27z335-371.
l.6. Morganroth J. Premature ventricular complexes:diagnosisand indications for therapy.JAMA 19&1,252:673-676. 17. Bigger JT Jr. Identification of patients at high risk for sudden cardiac death.Am J Car&/ 19&1;54(suppl):3D-8D. 18. Kennedy HL Whitlock JA Sprague MK, Kennedy IJ, BuckinghamTA, Goldberg RJ. Long-term follow-up of asymptomatichealthy subjectswith frequent and complexventricular ectopy.N Engl JMed 1985;312:193197. is. Gomes JA, Winters SL, Stewart D, Horowitz S, Milner M, Barreca P. A new noninvasive Index to predict sustainedventricular tachycardia and sudden death in the first year after myocardial infarction: based on signal-averaged electrocardiogram,radionuclide ejection fraction and Hoher monitoring. J Am Cdl Cardid 1987$X4%357.
A SYMPOSIUM: ANTIARRHYTHMIC DRUG THERAPY
7A