Management of Nonsustained VentricularTachycardia Guided By ElectrophysiologicalTesting

ALAN KADISH, STEPHEN SCHMALTZ, HUGH CALKINS, andFRED MORADY

From the Division of Cardiology. Department of Internal Medicine, University of MichiganMedical Center. Ann Arbor, Michigan

KADISH, A., ET AL.: Management of Nonsustained Ventricular Tachycardia Guided By Electrophysiologi-cal Testing. Two hundred eighty patients with spontaneous nonsusfained ventricular tachycardia weretreated based on the results of electrophysioJogical testing. Seventy-nine patients had no evidence 0/structural heart disease, 134 had coronary artery disease, 43 had idiopathic dilated cardiomyopathy,and 24 patients had miscellaneous types of heart disease. Sustained monomorphic ventricular tachycar-dia was induced during electrophysiological testing in the drug free state in 52 of 280 patients fl9%).Ventricular tachycardia was induced more frequently in patients with coronary artery disease {32%} thanin any of the other groups (P < 0.001}. The patients with inducible sustained monomorphic ventriculartachycardia underwent a mean of 1.9 ± 1.3 drug trials. Twenty-five patients had the induction of ventricu-lar tachycardia suppressed by pharmacological therapy and were treated with the drug judged to beeffective during electropharmacological testing. Twenty-seven patients continued to have inducible sus-tained monomorphic ventricular tachycardia despite antiarrhythmic therapy and were discharged onthe drug that made induced ventricular tachycardia best tolerated. Forty-five of 280 patients (16.1%)died during a mean follow-up period of 19.6 ± 14.4 months, There were 15 sudden cardiac deaths, 21nonsudden cardiac deaths, 6 noncardiac deaths, and 3 deaths that could not be classified. Suddencardiac death mortality was lowest in the patients without structural heart disease (0% at 2 years),intermediate in the patients with coronary artery disease and miscellaneous heart disease (4% al 2 years),and highest in the patients with idiopathic dilated cardiomyopathy (13% at 2 years; P < 0.01 for pairwisecomparisons). No patient treated with a drug that had suppressed the induction of sustained ventriculartachycardia died suddenly during the follow-up period whereas four of 27 patients who were dischargedon "ineffective antiarrhythmic drugs" and 11 of 228 patients without inducible sustained ventriculartachycardia experienced sudden cardiac death during the foUow-up period. By muJtivariate analysis,ejection fraction and inducibJe ventricular tachycardia during the predischarge eiectrophysiological testwere independent predictors of sudden cardiac death. In conclusion, in patients with spontaneous non-sustained ventricular tachycardia: (lj Arrhythmia inducibility varies depending on the underlying heartdisease. Ventricular tachycardia is most often inducible in patients with coronary artery disease andleast often in patients without structural heart disease; (2) With the exception of patients with idiopathicdilated cardiomyopathy, management of patients with nonsustained ventricular tachycardia guided hyelectrophysiologicaJ testing appears to result in a low incidence of sudden cardiac death although effectson total mortality are less impressive; and (3) Patients with idiopathic dilated cardiomyopathy andpatients with other heart diseases who continue to have inducible ventricular tachycardia despite antiar-rhythmic drug therapy are at substantial risk of sudden cardiac death. (PACE. Vol. 16, May, Part 11993)

sudden cardiac death, ejection fraction, ventricular tachycardia

Supported in part by Grant 5MO1-RR00042-25 from the Na- 524, 250 East Superior Street, Chicago. IL 60611. Fax: {312)tional Institute of Health, Bethesda, Maryland. Dr. KadLsh is a 908-4753.recipient of an NIH First Award Number HL40667. Received January 10, 1992; revision September 24, 1992; revi-Address for reprints: Alan Kadish, M.D., Northwestern Memo- sion November 6, 1992; accepted November 9, 1992.rial Hospital, Section of Cardiology, Wesley Pavilion. Suite

PACE, Vol. 16 May 1993. Part I 1037

KADISH. ET AL.

IntroductionNonsustained ventricular taGhycardia is asso-

ciated with an adverse prognosis in certain pa-tients.'"' Several studies have examined the roleof ejection fraction, Holter monitoring, the signal-averaged electroGardiogram, and electrophysio-logical testing in defining prognosis in patientswith nonsustained ventricular tachycardia.'"''However, the use of these diagnostic tests and themost appropriate form of therapy in patients withvarious types of heart disease and nonsustainedventricular tachycardia are as yet unclear.

The purpose of this study was to descrihe theoutcome in a large cohort of patients with varioustypes of heart disease who had nonsustained ven-tricular tachycardia and whose management wasguided hy electrophysiological testing.

MethodsPatient Population

The present investigation was a retrospectivestudy of 280 patients consecutive with asymptom-atic, nonsustained ventricular tachycardia whounderwent electrophysiological testing in the drugfree state. Patients with nonspecific palpitationswere not considered symptomatic for the purposes

of this study. Patients who had suffered a cardiacarrest or experienced sustained uniform ventricu-lar tachycardia were not entered into the study.Nonsustained ventricular tachycardia was definedas having a duration of three heats to 30 secondsand was documented by amhulatory monitoring(240 patients) or telemetry recordings. Seventy-nine patients had no evidence of structural heartdisease as determined by cardiac catheterizationor the combination of exercise testing and two-dimensional echocardiography. One hundred thir-ty-four patients had coronary artery disease docu-mented by cardiac catheterization or the presenceof a Q wave infarction, and 43 patients had anidiopathic dilated cardiomyopathy documentedby cardiac catheterization. Twenty-four patientshad other types of heart disease, primarily valvu-lar heart disease (Table I). One hundred twenty-five of the 134 patients with coronary disease hadevidence of prior myocardial infarction. Demo-graphic characteristics of these patients are de-scribed in Table I.

Electrophysiological Study Protocol

Electrophysiological studies were performedwhile patients were fasting and not sedated. In-formed consent was ohtained from all patientsprior to electrophysioiogicai testing. Antiarrhyth-

Heart DiseaseNumber of ptsMean Age (years)Gender

Mean EFNYHA Class 1

IIIIIIV

Demographic

1

None79

52 ± 1546 male33 female

0.62 ± 0.06611620

Table 1.

Characteristics of

Group

2

CAD134

62 ± 11118 male16 female

0.36 ± 0.1545473210

the Patient Population

Number

3

CM43

58 + 1337 male6 female

0.26 ± 0.146

15166

4

Other'24

54 + 1614 male10 female

0.43 + 0.1910743

All

28058 + 13

215 male65 female

0.37 + 0.17122855419

CAD = coronary artery disease; CM = idiopaihic dilated cardiomyopathy: EF - ejection fraction.* Other Heart Disease: valvular heart disease 13, hypertensive cardiovascular disease 3; idiopathic hypertrophic cardiomyopathy7; coronary artery spasm 1.

1038 May 1993, Part I PACE, Vol. 16

ELECTROPHYSIOLOGICAL TESTING IN NONSUSTAINED VENTRICULAR TACHYCARDIA

mic therapy was discontinued at least five half-lives hefore testing. Three quadripolar catheterswere inserted into a femoral vein and positionedin the high right atrium, at the right ventricularapex, and in the right ventricular outflow tract.Leads VI, 1, and 3 and the intracardiac elec-trograms were displayed on an oscilloscope andrecorded on a Siemens-Elema Mingograf 7 re-corder (Siemens-Elema, Solna, Sweden).

Ventricular stimulation was performed usinga programmable stimulator (Bloom Associates,Narbeth, PA, USA). Rectangular stimuli 2 msec induration at twice diastolic threshold were used.Up to 3 extrastimuli were introduced using at leasttwo basic drive cycle lengths at the right ventricu-lar apex and at the right ventricular outflow tractor upper septum.^ Diastole was scanned as previ-ously described and coupling intervals were lim-ited to 200 msec." The endpoint of the stimulationprotocol was the induction of sustained monomor-phic ventricular tachycardia. Nonsustained poly-morphic ventricular tachycardia was not an end-point of the stimulation protocol. If sustainedpolymorphic ventricular tachycardia or ventricu-lar fibrillation was induced once, stimulation gen-erally was continued in an attempt to induce sus-tained monomorphic ventricular tachycardia ex-cept if multiple countershocks were required toterminate ventricular fibrillation or if the patientrefused further testing.'* Only sustained monomor-phic ventricular tachycardia was considered apositive result.

If sustained monomorphic ventricular tachy-cardia was induced during the haseline electro-physiological test, the patient underwent serialdrug testing in an attempt to suppress the induc-tion of ventricular tachycardia. Sustained mono-morphic ventricular tachycardia was also the end-point used for assessing drug efficacy. If the induc-tion of ventricular tachycardia was suppressed byan antiarrhythmic agent, the patient was treatedlong term with that drug. If the induction of ven-tricular tachycardia was not suppressed hy antiar-rhythmic therapy, patients were treated with thedrug that made induced ventricular tachycardiahest tolerated, except for two patients who refusedtherapy. Tolerance of ventricular tachycardia wasdetermined by blood pressure and/or symptomsduring sustained ventricular tachycardia. All pa-tients who failed drug testing (except one who re-

fused further testing) had undergone at least twoelectrophysiological tests on different antiarrhyth-mic drugs.

The following definitions were used with re-gard to the electrophysiological study: Sustainedventricular tachycardia was defined as lasting >30 seconds or requiring termination because of he-modynamic compromise. Nonsustained ventricu-lar tachycardia was defined as lasting from threebeats to 30 seconds. Monomorphic ventriculartachycardia had a stable QRS morphology whenviewed simultaneously in three leads. Polymor-phic ventricular tachycardia was defined as hav-ing a variahle QRS morphology.

Patient Follow-Up

Follow-up information was ohtained from thepatient, patient's family, and/or the patient's phy-sician. Eleven (3.8%) patients who met the inclu-sion criteria were lost to follow-up and the remain-ing 280 patients form the study population. Infor-mation was obtained on the occurrence ofsustained ventricular tachycardia or cardiac arrestand on the causes of death.

Deaths were classified as sudden cardiacdeath, other cardiac death, and noncardiac death.For the purposes of this study, sudden cardiacdeath was defined as instantaneous death or un-witnessed death with no evidence of a nonarrhyth-mic cause of death. Patients who experienced pre-monitory symptoms consistent with an arrhyth-mia, e.g., palpitations or presyncope, were alsoclassified as having sudden cardiac death. Thepresence of other premonitory symptoms such aschest pain or shortness of breath led to a classifica-tion of "other cardiac death." In three patients,insufficient information was available to classifythe cause of death. In the analysis of sudden car-diac death and ventricular tachycardia occur-rence, these three patients were censored at thetime of death.

Statistical Methods

Data are expressed as mean ± 1 standard de-viation, Frequencies were compared hy Chi-square analysis. Survival curves were constructedusing the Kaplan-Meier product limit estimate.^°The standard errors of survival curves were calcu-lated using the methods of Kaplan-Meier and mul-

PACE, Vol. 16 May 1993, Part I 1039

KADISH, ET AL.

tiple comparisons between curves were performedusing the Bonferroni correction method.^" A Pvalue of < 0.05 was considered significant. In ex-amining outcome, the three response variablesused were the time until sudden cardiac death,time until death (all causes), and the time untilsustained ventricular tachycardia/ventricular fi-brillation. Sustained ventricular tachycardia/ven-tricular fibrillation occurrence was defined as theoccurrence of documented sustained ventriculartachycardia or sudden cardiac death. The maingoal of the statistical analysis was to assess thedependence of the three response variables on anumber of explanatory variables. These explana-tory variables are listed in Table II. The statisticalmethod used to assess the effect of these explana-tory variables on survival was Cox regression anal-ysis. In the first stage of the analysis, each of theexplanatory variables was used to assess its indi-vidual effect on survival. Explanatory variablesthen were combined using a forward selection pro-cedure to determine the best set of variables re-lated to survival. Because survival times differ de-pending on the underlying heart disease, the anal-ysis was stratified by heart disease. The data werealso separately analyzed for the coronary arterydisease and cardiomyopathy groups. A P value <0.10 was considered significant for the purpose ofthe Cox regression analysis.^"

Instead of selecting a predefined ejection frac-tion to separate low risk and high risk groups, lo-gistic regression was used to estimate the probabil-ity of death as a function of this variable,"' Using

Table II.

Variables Used in Multivariate Analysis

AgeSexEjection FractionNYHA ClassificationPrior anti-arrhythmic drug therapyResults of Baseline electrophysioiogical testingResults of Predischarge electrophysioiogical testingAnti-arrhythmic drug therapy after dischargeDigoxin therapyDiurectic therapyBeta blocker therapy

this logistic regression, the sensitivity and speci-ficity were calculated for different cutoffs of thepredicted probabilities. These predicted probabili-ties were then inverted to yield the resulting cut-offs for the ejection fraction.

Results

Electrophysioiogical Testing

Among the 280 patients who underwent abaseline electrophysioiogical test, 52 (19%) hadinducible sustained monomorphic ventriculartachycardia. The mean cycle length of monomor-phic ventricular tachycardia induced during thebaseline electrophysioiogical test was 286 ± 59msec. A mean of 2.3 ± 0,6 extrastimnli were re-quired to induce sustained monomorphic ventric-ular tachycardia. The inducibility of ventriculartachycardia based on the type of underlying heartdisease is shown in Figure 1. Ventricular tachycar-dia was induced more frequently in patients withcoronary artery disease (32%] than in the othergroups (P < 0.001). The induction of ventriculartachycardia in patients with coronary artery dis-ease was dependent upon the level of left ventricu-lar function. Nine of 50 (18%) patients with a leftventricular ejection fraction of > 0.4 had inducibleventricular tachycardia versus 34 of 84 (40%) pa-tients with a left ventricular ejection fraction <0.4 (P < 0.05). Ventricular tachycardia also wasinduced more frequently in the patients with idio-pathic dilated cardiomyopathy (14%) and withmiscellaneous heart diseases (13%) than in thosewithout structural heart disease (0%; P < 0.01 forboth comparisons). The mean cycle length of in-duced ventricular tachycardia was similar in thevarious heart disease groups: 285 ± 59 msec inthe patients with coronary artery disease, 297 ±65 msec in the patients with idiopathic dilated car-diomyopathy, and 280 ± 71 msec in the patientswith miscellaneous heart disease. The mean num-her of extrastimuli required to induce ventriculartachycardia was also similar regardless of the un-derlying heart disease.

Among the 228 patients considered to havenegative electrophysioiogical tests, 156 patientshad no arrhythmia induced, 59 had nonsustainedventricular tachycardia induced, and 13 had ven-tricular fibrillation induced.

1040 May 1993, Part I PACE, Vol, 16

ELECTROPHYSIOLOGICAL TESTING IN NONSUSTAINED VENTRICULAR TACHYCARDIA

EPS Results140

£] Non-lnducible

• InducJbie

' - p<0,001 vs other gps

# - p<0,01 vs CAD and no HD

Figure 1. Inducibility of ventricukir tachycardia according to type o/underlying heorf disease.The number of patients with inducible monomorphic venfricuiar tachycardia during a baselineelectrophysioiogical test is shown in the solid areas and the number of palieiits without inducibiesustained ventricuiar (achycardia in the shaded areas, Inducibility of ventricular tachycardiawas highest in the patients with coronary artery disease (CADj, intermediate in those wifh idio-pathic dilated cardiomyopathy fCM) and miscellaneous heart disease (Misc HD), and lowest inthose without structuraJ heart disease (No HD) (P < 0.01 for pairwise comparisonsj.

Electropharmacoiogical Testing

The patients with inducible sustained mono-morphic ventricular tachycardia underwent amean of 1,9 ± 1.3 drug trials (range 1-5). Threepatients refused electrophysioiogical testing be-fore having failed at least two antiarrhythmic drugtrials. They were discharged on empiric therapy(one patient) or no therapy (two patients) and wereconsidered to have persistently inducible ventric-ular tachycardia for further analysis. None of thesethree patients died suddenly during the follow-upperiod. No patient underwent implantation of animplantable cardioverter defihrillator or had ar-rhythmia surgery performed.

Among 52 patients who underwent electro-pharmacological testing, 27 continued to have in-ducible sustained monomorphic ventriculartachycardia and 25 did not. During the baselineelectrophysioiogical test, the mean cycle length ofinduced ventricular tachycardia (279 ± 59 vs 292± 59 msec) and the mean number of extrastimulineeded to induce ventricular tachycardia (2.2 ±0.7 vs 2.4 ± 0.5) were similar in the patients whodid and did not respond during drug testing. Inthe patients with coronary artery disease, five of

nine patients with a left ventricular ejection frac-tion > 0.4 had the inducihility of ventriculartachycardia suppressed with antiarrhythmia med-ication. The patients in whom ventricular tachy-cardia induction was suppressed hy an antiar-rhythmic drug wore treated long term with thatdrug; the 27 patients who continued to have indu-cible sustained ventricular tachycardia were dis-charged on the drug that made the induced ven-tricular tachycardia slowest and most hemody-namically stable. Among the 52 patients who hadinducible sustained monomorphic ventriculartachycardia in the baseline state, two were dis-charged on no therapy (at the patient's request),24 received a type IA antiarrhythmic agent, 16 re-ceived amiodarone, 6 received a combination ofamiodarone and a type I antiarrhythmic agent, and4 were treated with a type IC antiarrhythmic agent.

Among 228 patients without inducible sus-tained monomorphic ventricuiar tachycardia dur-ing the baseline electrophysioiogical test, 89 re-ceived empiric antiarrhythmic therapy for nonsus-tained ventricular tachycardia or for atrialarrhythmias and 139 received no drug therapy.Forty patients were treated with a type IA antiar-

PACE, Vol. 16 May 1993, Par. I 1041

KADISH, ET AL.

rhythmic agent, 11 with amiodarone, 3 with acomhination of amiodarone and another agent, 10with verapamil (for nonsustained ventriculartachycardia in the absence of structural heart dis-ease), and 25 with a type IC antiarrhythmic agent.

Outcome

Patients were followed for a mean of 19.6 ±14.4 months (range 1-65). Forty-five of 280 pa-tients (16.1%) died during the follow-up period.There were 15 sudden cardiac deaths, 21 nonsud-den cardiac deaths, 6 noncardiac deaths, and 3deaths that could not be classified. Survival curvesfor deaths from all causes are shown in Figure 2.Two-year survival was 100% in the patients with-out heart disease, 83% in the patients with coro-nary artery disease, 73% in the patients with mis-cellaneous heart diseases, and 68% in the patientswith idiopathic dilated cardiomyopathy. Theprobability of survival was significantly higher inthe patients without structural heart disease (P <0.01) than in all the other groups. In addition, pa-tients with coronary artery disease had a signifi-cantly lower mortality than those with idiopathicdilated cardiomyopathy (P < 0.05),

The incidence of sudden cardiac death at 2years was 0% in the patients without heart disease,

5% in the patients with coronary artery disease,4% in the patients with miscellaneous heart dis-eases, and 23% in the patients with idiopathic di-lated cardiomyopathy. Patients without structuralbeart disease had a significantly lower incidenceof sudden cardiac death than all other groups (P <0,01) and the patients with coronary artery diseasehad a lower incidence than patients with idio-pathic dilated cardiomyopathy (P < 0.02). How-ever, patients with idiopathic dilated cardiomyop-athy had a mean ejection fraction that was lowerthan that in the other groups (Table I).

Survival curves for freedom from sudden car-diac death are shown in Figure 3. Four of the 52patients with inducible sustained ventriculartachycardia during the baseline electrophysiologi-cal test died suddenly during the follow-up pe-riod. The mean cycle length of induced ventriculartachycardia was 305 ± 86 msec in the foiir pa-tients who died suddenly and 284 ± 57 msec inthe 48 patients who did not die suddenly (NS).The number of extrastimuli required to induceventricular tachycardia was also similar in thesetwo groups (2.2 ± 0.5 extrastimuli vs 2.3 ± 0.6extrastimuli).

Twenty-four of 280 patients either died sud-denly or had an episode of sustained ventriculartachycardia during follow-up. One had no under-

O O O O O O P O O O O O O O

a>1in

EO

i " B o o o o o o o o o o o e e > o o o o

5 10 15 20 25Time (months.)

No HD «-» CAD » « CM Other

30 35

Figure 2. Survival curve analysis^j total mortality during follow-upin each of the subgroups. Totalmortality was highest in the pa-tients with idiopathic dilated car-diomyopalhy, intermediate inthose with coronary artery diseaseand miscellaneous heart disease,and lowest in those without struc-tural heart disease. See text/or de-tails.

1042 May 1993. Part I PACE, Vol. 16

ELECTROPHYSIOLOGICAL TESTING IN NONSUSTAINED VENTRICULAR TACHYCARDIA

^^

irvl

vc

_j

inc

Cu

mu

lai

100

95

90

85

80

7570

65

60

55

50

oaaaa0 000000

DDOOODDODaO

Figure 3. Sudden cardiac death ineach of the subgroup.s. The inci-dence of sudden cardiac death washighest in patients with idiopathicdilated cardiomyopathy, interme-diate in patienfs with coronary ar-tery diseases and misceJJaneousheart disease, and lowest in pa-tients without StructuraJ heart dis-ease. See text for detaiJs.

30

NoHD

15 20 25Time (monlhs)

CAD B « CM Other

35

lying heart disease, 13 had coronary artery disease,8 had an idiopathic dilated cardiomyopathy, and2 had miscellaneous heart diseases. Freedom fromsustained ventricular arrhj^hmia occurrence ineach of the groups is shown in Figure 4. Patientswithout structural heart disease were less likely

to experience a sustained arrhythmia occurrenceduring follow-up than were patients with coronaryartery disease or idiopathic dilated cardiomyope-thy (P < 0.05].

Figures 5 through 7 describe patient outcomebased on the results of electrophysioiogical test-

ODOOODGOQDDIi]

^ • • • • • • • • • • • » • » • »

D D D a o-G

10

NoHD

15 20 25 30Time (months)

CAD » « CM Other

35

Figure 4. Survival curve analysisof freedom from arrhythmia recur-rence in each of the subgroups,VentricuJar tachycardia and sud-den cardiac death were highest inpatients with idiopathic diJatedcardiomyopathy, intermediate in(hose wilh coronary artery diseaseand miscelJaneous heart diseases,and lowest in those without struc-tural heart disease. See text/or de-tails.

PACE, Vol. 16 May 1993. Part I 1043

KADISH, ET AL.

280 Patients

inducible VTn = 52

no inducible VTn=228

VT Suppressedtreated with effective

drug therapyn = 25

VT not suppressed27

no auhythmian=23

sustained noVT Airhythmia11=2 n=22

sustainedSCD

n=4n=l

arrhythmia sCD83 n=3

Figure 5. ReJationship of results of eJectrophysioJogicaJ testing to outcome in the totaJ groupof 280 patients.

ing. Frequency data [not actuarial survival) is pre-sented in the figures for simplicity. The relation-ship between the results of electrophysioiogicaltesting and outcome for all patients is shown inFigure 5. Patients who had inducihle ventriculartachycardia during the predischarge electrophysi-

oiogical test had the highest actuarial incidence ofsudden cardiac death during follow-up (11% at 2years]. The incidence of sudden cardiac death at2 years was 0% in the patients treated with an anti-arrhythmic agent that suppressed the induction ofventricular tachycardia, 4% in the patients treated

134 Patients

inducible VTn=43

no indudble VTn=91

VT Suppressedn=21

VT not suppressedn=22

SCDno an-hythmia VT sustained „n=19 n=2

empiric drug Rxn=35

no drug Rxn=56

VT sustained no1=1 arrhythmia

no arrhythmian=18 SCD

11=1

VT sustained SCDn=l n=3

VT sustainedn=l

no arrhythmian=52

Figure 6. Relationship of the results of eiectrophysioJogical testing to outcome in 134 patientswith coronary artery disease.

1044 May 1993. Part I PACE, Vol. 16

ELECTROPHYSIOLOGICAL TESTING IN NONSUSTAINED VENTRICULAR TACHYCARDIA

43 Patients

inducibleVTn=6

no inducibleVTn=37

VT Supressedn=3

noarrhythmia

n=3

VT not suppressedn=3

empiric dnig Rxn=14

noarrhythmia

n=2SCDn=l

SCDn=l

no drug Rxn=22

no no SCDarrhythmia arrhythmia n=5

n=13 n=17

Figure 7. Relationship of the results of electrophysioiogicai testing to outcome in 43 patientswilh idiopathic dilated cardiomyopathy.

empirically with antiarrhythmic agents despitenegative baseline electrophysiological tests, and5% in the untreated patients. Data on the 134 pa-tients with coronary artery disease and on the 43patients with idiopathic dilated cardiomyopathyare shown in Figures 6 and 7. Patients with idio-pathic dilated cardiomyopathy were at high riskfor sudden cardiac death despite a negative elec-trophysiological test. If patients with idiopathicdilated cardiomyopathy are excluded from theanalysis, the incidence of sudden cardiac death at1 year was 0% in the 22 patients who were treatedwith drugs that were effective in suppressing in-ducible ventricular tachycardia, 12% in the 24 pa-tients who did not have inducible ventriculartachycardia suppressed at electrophysiologicaltesting, and 3% in the 191 patients with a negativebaseline electrophysioiogicai test. The incidenceof sudden death was not significantly different inthose who did (2.6%) versus did not (3.4%) receiveempiric therapy.

Predictors of OutcomeResults of the forward selection procedure are

shown in Table III.Using the forward selection procedure, only

lower ejection fraction and inducible ventriculartachycardia during the predischarge electrophysi-ological test correlated with sudden death duringfollow-up. Older age. lower ejection fraction, andhigher New York Heart Association (NYHA] clas-

sification all correlated with total mortality duringthe follow-up period. Sustained ventricular ar-rhythmia occurrence during the follow-up periodwas associated with a lower ejection fraction andinducible sustained ventricular tachycardia dur-ing the predischarge electrophysiological test.

Survival analysis was repeated in twosubgroups: patients with coronary artery diseaseand patients with idiopathic dilated cardiomyopa-thy. Despite the high percentage of sudden cardiacdeath in patients with idiopathic dilated cardio-myopathy, the number of patients was too smallto perform the forward selection procedure. In pa-tients with coronary artery disease, increasing age,lower ejection fraction, and higher NYHA classifi-cation correlated with sudden cardiac death. In-creasing age, higher NYHA classification, and ven-tricular tachycardia induced during the predis-charge electrophysiological test correlated withtotal mortality. No variables correlated with sus-tained ventricular arrhythmia occurrence in thepatients with coronary artery disease.

Using logistic regression, a left ventricularejection fraction of 0.37 provided the best discrim-ination among outcomes [Fig. 8). Patients who hadan ejection fraction > 0.37 had a 0% incidence ofsudden cardiac death at 2 years, whereas thosewho had an ejection fraction < 0.37 had a 25%incidence of sudden cardiac death at 2 years (P <0,001). In addition, patients with a left ventricularejection fraction < 0.37 who had coronary artery

PACE, Vol. 16 May 1993, Part I 1045

KADISH, ET AL,

Results of

Table III.

Multivariate Analysis

Total Group (n =

Variables P

(Significant

280)

Value

Variables)

Coronary Artery

Variables

Disease (n

P

= 134)

Value

Sudden Cardiac Death

Total Survival

Sustained Arrhythmia Occurrences

EFVT last Dx

AgeEF

NYHA Class

EFVT last Dx

0.0010.011

0.0060.0010.036

0.0230.046

EFNYHA Class

EFAge

NYHA ClassVT last Dx

0.0430.023

0.0130.0080.0070.094

EF = ejection fraction; NYHA = New York Heart Association Class; VT last Dx = Inducible VT at predischarge electrophysiologi-cat testing.

disease or miscellaneous heart disease and a nega-tive baseline electrophysiological test had a 2-yearsudden cardiac death mortality of only 6%.

Survival analysis was repeated to determinethe relationship between therapy with specificantiarrhythmic agents and subsequent arrhythmiaevents. By univariate analysis, therapy with typeIA, IC, or III antiarrhythmic agents was not associ-

ated with the occurrence of sudden cardiac deathor ventricular tachycardia.

DiscussionMain Findings

The major finding of this study is that the useof electrophysiological testing and electropharma-

3

in

2Eo

100

95

90

85

80

75

70

60

55

501' '• ' ' I ' ' ' • T • -̂ • • ' 1 ' ' ' ' T •• • ' ' I '

0 5 10 15 20 25Time (months)

EF< 0.368 •^-•EF > 0.368

30 35

Figure 8. Freedom from suddencardiac death in the totaJ patientpopulation stratified by ejectionfraction (EF). An EF of 0.37 had asensitivity of 93% and a specificityof 59% in predicting sudden car-diac death.

1046 Mav 1993, Part I PACE, Vol. 16

ELECTROPHYSIOLOGICAL TESTING IN NONSUSTAINED VENTRICULAR TACHYCARDIA

cological testing to manage patients with nonsus-tained ventricular tachycardia are associated witha low incidence of sudden cardiac death in pa-tients with coronary artery disease, miscellaneousheart diseases, and in patients without structuralheart disease. However, the incidence of suddendeath in patients with idiopathic dilated cardio-myopathy was high [17% at 1 year). The resultssuggest that electrophysiologicai testing and elec-tropharmacological therapy may be useful in man-aging patients with heart diseases other than idio-pathic dilated cardiomyopathy. Excluding thosewho had idiopathic dilated cardiomyopathy, pa-tients with negative baseline electrophysiologicaltests or those treated with antiarrhythmic drugsthat suppressed the induction of ventricular tachy-cardia had a low incidence of sudden cardiacdeath during follow-up. In contrast, patients withpersistently inducible ventricular tachycardia andpatients with idiopathic dilated cardiomyopathyremained at substantial risk of sudden cardiacdeath. However, the results of this study must beregarded as preliminary because of the potentialfor proarrhythmia in the patients treated empiri-cally and because total mortality was not clearlyrelated to the results of electropbysiologicaltesting.

Effect of Underlying Heart Disease

Sudden cardiac death and total mortality ap-peared to be dependent on the underlying heartdisease in this study. Patients without structuralheart disease had a negligible incidence of ventric-ular tachycardia or death during follow-up. Pa-tients with coronary artery disease and miscella-neous heart diseases had a moderately low mortal-ity from sudden cardiac death, whereas patientswith idiopathic dilated cardiomyopathy had ahigh mortality from sudden and nonsudden car-diac death. However, patients with idiopathic di-lated cardiomyopathy had lower left ventricularejection fractions than tbose in other groups andthus the independent risk of the type of underlyingheart disease on outcome cannot be establishedwith certainty.

Electrophysiological testing was not of valuein patients without structural heart disease whohad asymptomatic nonsustained ventriculartachycardia, electrophysiological tests were uni-

formly negative and prognosis was excellent. Inaddition, an ejection fraction > 0.37 was associ-ated with a good prognosis in all groups of patientswhen patients were managed based on the resultsof electrophysiological testing. However, the useof this number as an absolute cutoff for assessingrisk must be approached with caution, because thenumber was retrospectively derived.

The results of this study suggest that electro-physiological testing may be of benefit in risk strat-ification and management of patients with coro-nary artery disease or miscellaneous heart diseaseand left ventricular dysfunction. The inducibilityof ventricular tachycardia during the predischargeelectrophysiological test was an independent pre-dictor of sudden cardiac death. The value of a neg-ative baseline electrophysiological test in thisgroup of patients cannot be determined witb cer-tainty in the absence of a control group not sub-jected to electrophysiological testing. However,this incidence of sudden death of 2% per year islower than in several groups of historical controlswith untreated spontaneous nonsustained ventric-ular tachycardia.^ ' When the analysis was con-fined only to patients with coronary artery disease,the raw incidence of sudden cardiac death waslower in patients with negative predischarge elec-trophysiological tests than those with positiveones [6% vs 14%), but this difference did not reachsignificance perhaps because of the smaller num-ber of patients. A positive electrophysiological testalso appeared to be of value. Patients with a posi-tive baseline electrophysiological test in whomventricular tachycardia was suppressed by antiar-rhythmic therapy had a 0% incidence of suddencardiac death whereas those with persistently in-ducible ventricular tachycardia had a 20% inci-dence of sudden cardiac death at 2 years, suggest-ing that the results of electropharmacological test-ing are predictive of outcome in these patients. Inpatients with idiopathic dilated cardiomyopathy,a negative electrophysiological study had no pre-dictive value. Six of 37 patients with a negativeelectrophysiological test died suddenly during fol-low-up. However, it is possible that a positive elec-trophysiological test was useful in these patients.Although none of the three patients with a positivebaseline electrophysiological test who had ven-tricular tachycardia suppressed by antiarrhythmicmedications experienced sudden cardiac death or

PACE, Vol. 16 May 1993, Part I 1047

KADISH, ET AL.

ventricular tachycardia during follow-up, thisnumber of patients clearly is too small to allowany conclusions.

Results of Prior Studies

The results of several prior studies that in-cluded relatively small numbers of patients havebeen divergent. Some studies have suggested thatelectrophysiological studies are of no value in riskstratification or patient management, while othershave suggested that the noninducibility of ventric-ular tachycardia predicts a low incidence of sud-den death.^^"^^ The majority of studies have con-cluded that the inducibility of sustained ventricu-lar tachycardia portends a poor prognosis.**'̂ '*'̂

Three recent studies of 100 or more patientshave examined the value of electrophysiologicaltesting in patients with nonsustained ventriculartachycardia. Hammill et al.'^ studied 110 patientswith a variety of types of heart disease, and fol-lowed them for a mean of 15 months. In theirstudy, only the ejection fraction and the presenceof congestive heart failure were independent pre-dictors of outcome, and electrophysiological test-ing was not of major benefit in risk stratification.Kowey et al.̂ ^ studied 205 asvmptomatic patientswith coronary disease and remote myocardial in-farction. Only left ventricular function correlatedsignificantly with outcome and electrophysiologi-cal testing did not appear to be of major value.Unexpectedly, 14 of 45 patients (31%) had ventric-ular tachycardia or sudden cardiac death despitereceiving therapy guided by electropharmacologi-cal testing.^^ This incidence of ventricular tachy-cardia and sudden death is inexplicably higherthan noted in historical controls who presentedwith sustained ventricular tachycardia.^"* Wilberet al."* reported the results of electrophysiologicaltesting in 100 consecutive patieuts with ejectionfractions of < 0.40, spontaneous nonsustaiuedventricular tachycardia, and prior myocardial in-farction. The persistence of inducible sustainedventricular tachycardia was the only independentpredictor of sudden cardiac death in that study.

Prior studies have suggested that electrophys-iological testing is uot useful in managing patientswith idiopathic dilated cardiomyopathy.^"'^^ Pa-tients with negative electrophysiological tests hada high risk of sudden cardiac death in studies byKron et al.^° and Das et al.̂ ^ The results of the

present study are in agreement with these find-ings. One potential complicating feature in inter-preting the results iu patients with dilated cardio-myopathy is that a recent study has suggested thatbradycardia may be a common mechanisru of sud-den cardiac death in such patients.^^ Thus, diag-nostic tests and therapeutic maneuvers directed attachyarrhythmias may be less beneficial iu suchpatients.

The reasons for the different findings in thisstudy compared to some other recent studies areunclear. The major difference between this studyand the study of Kowey et al.^^ is the low inci-dence of sudden cardiac death in this study in pa-tieuts who received antiarrhythmic therapyguided by electrophysiological testing. This lowincidence is similar to that described by Klein andMachelP^ and by Wilber et al.'" One potential ex-planation is that iu the multicenter study byKowey et al.,'^ there may have been differences inthe endpoints of electropharmacological testing.The role of electropharmacological testing in themanagement of patients who present with sus-tained tachyarrhythmias is also somewhat contro-versial. Although a large number of studies (re-viewed in reference 23) have suggested thatsuppression of inducible sustained ventriculartachycardia by type I antiarrhythmic drugs por-tends a good prognosis, recent studies such asthose by Poole et al.^^ suggests that this may notbe true. However, in the present study, patientspresenting with nousustained ventricular tachy-cardia who had inducible ventricular tachycardiasuppressed by autiarrhythmic drugs had a 0% in-cidence of sudden cardiac death at 2 years.

The findings of this study are generally inagreement with those of Wilber et al.^^ However,there are two important differences between thetwo studies aside from the larger number of pa-tients in the present study. Rather than selectinga predefined ejection fraction for inclusion iu thestudy, we analyzed data from all patients and per-formed logistic regression prospectively to definean ejection fraction (0.37) that provided an excel-lent separation between patients with and withoutsubsequent cardiac death. In addition, we in-cluded all patients regardless of the nature of theunderlying heart disease, allowing a comparisonamong different heart disease groups.

In contrast to the findings of the CASTstudy,^^ therapy with antiarrhythmic drugs in gen-

1048 May 1993, Part I PACE, Vol. 16

ELECTROPHYSIOLOGICAL TESTING IN NONSUSTAINED VENTRICULAR TACHYCARDIA

eral and with type IC antiarrhythmic drugs in par-ticular did not correlate with the occurrence ofsudden cardiac death or ventricular tachycardia.One potential explanation is that in the presentstudy these drugs were used empirically only inpatients with negative electrophysiological stud-ies. It is possible that patients with inducible ven-tricular tachycardia are at bigher risk of proar-rhythmia from type IC antiarrhythmic agents. Inaddition, a number of patients who received typeIC antiarrhythmic agents in this study had normalleft ventricular function and tbis also may havecontributed to lowering tbe risk of a proarrhyth-mic response to antiarrhythmic drugs.

Limitations

The major limitation of this study is the ab-sence of an untreated or placebo control group.Therefore, it cannot be proven that managementbased on the results of electrophysiological testingimproved the outcome in patients who had indu-cible ventricular tachycardia. The response of in-ducible ventricular tachycardia to antiarrhythmicdrugs may itself be a marker of good outcome. Al-though comparison with historical controls^"^suggests that the incidence of sudden cardiacdeath in patients in this study who had inducibleventricular tachycardia was low, such retrospec-tive comparisons have major limitations. A secondlimitation of this study is that 11 patients werelost to follow-up and that the mechanism of deathscould not be determined with certainty in threepatients. However, this represents a low percent-age of the total number of patients in the study(3.8%) and of total deaths (6.3%) and thus anyerror introduced into the study by missing datashould be small. In addition, 89 of 228 patientswith a negative baseline electrophysiological testwere treated with antiarrhythmic drugs by theirreferring physicians. However, drug therapy wasnot an independent predictor of subsequent eventsin the study and the incidence of sudden cardiacdeath was similar in the patients who received em-piric drug therapy and those who did not. Anotherlimitation of this study is the small number of pa-tients in the group with miscellaneous heart dis-ease. In addition, patients in this group containedvarious diagnoses including hypertrophic cardio-myopathy and valvular heart disease. Althoughthese patients were included in the study so that

consecutive patients could be reported, the num-bers are probably too small to draw any meaning-ful conclusions. In addition, we did not analyzethe occurrence of sustained ventricular tachycar-dia alone as a separate outcome but only in combi-nation with cardiac arrest. This was done to limitthe number of statistical analyses. Finally, the re-sponse rate to antiarrhythmic drugs in this studywas unusually high. This may have in part beensecondary to the fact tbat inducible nonsustainedventricular tachycardia [< 30 sec] was considereda drug success. This unusually high response ratemay limit the applicability of these results to otherpatient populations.

Clinical Implications

The results of this study must be regarded aspreliminary because of the absence of a controlgroup and because of the small number of patientsin some subgroups. However, they suggest thatelectrophysiological testing may be of benefit inrisk stratification and management of some pa-tients with nonsustained ventricular tachycardiadepending on left ventricular function and thetype of heart disease. Among patients who haveleft ventricular dysfunction, those with coronaryartery disease or miscellaneous heart diseases anda negative electrophysiological test had a low inci-dence of sudden cardiac death and may not re-quire antiarrhythmic drug therapy. Patients whohave a positive baseline electrophysiological testand in whom the induction of ventricular tachy-cardia is suppressed by antiarrhythmic drugs alsohad a low incidence of sudden cardiac death, andelectropharmacological testing may be a usefulguide to therapy. However, patients who have apositive electrophysiological test and in whom theinduction of ventricular tachycardia cannot besuppressed by antiarrhythmic drugs have a highincidence of sudden death as do patients with id-iopathic dilated cardiomyopathy regardless of theresults of electrophysiological testing. These pa-tients therefore may be appropriate candidates fornonpharmacological therapy, such as placementof an implantable cardioverter defibrillator. Therisk to benefit ratio of device implantation or othertherapy in these types of patients remains to bedetermined by larger clinical trials using placebocontrols.

PACE, Vol. 16 May 1993, Part t 1049

KADISH, ET AL.

References

1. Bigger IT, Fleiss JL, Kleiger R, et al.. and the Multi-center Post-infarction Research Group. The rela-tionships among ventricular arrhythmias, left ven-tricular dysfunction and mortality in the two yearsafter myocardial infarction. Circulation 1984; 69:250-258.

2. Mukharji I. Rude FE. Poole K, MILIIS Study Group.Risk factors for sudden death following acute myo-cardial infarction: Two-year follow-up. Am I Car-diol 1984; 54:31-36.

3. Moss AI. Davis HT. Decamilla I. et al. Ventricularectopic beats and their relation to sudden and non-sudden cardiac death after myocardial infarction.Circulation 1979; 60:998-1003.

4. Holmes I, Kubo SH, Cody R). et al. Arrhythmias inischemic and nonischemic dilated cardiomyopa-thy: Prediction of mortality by ambulatory electro-cardiology. Am I Cardiol 1985; 55:166-151.

5. Dennis AR, Cody DV, Young AA, et al. Prognosticsignificance of ventricular tachycardia and fibrilla-tion induced at programmed stimulation and de-layed potentials detected on the signal-averagedelectrocardiograms of survivors of acute myocar-dial infarction. Circulation 1986; 74:731-745.

6. Buxton AE, Marchlinski FE, Flores BT, et al. Non-sustained ventricular tachycardia in patients withcoronary artery disease: Role of electrophysiologystudy. Circulation 1987; 75:1178-1185.

7. Morady F, DiCarlo L, Winston S, et al. A prospec-tive comparison of triple extrastimuli and left ven-tricular stimulation in studies of ventricular tachy-cardia induction. Circulation 1984; 70:52-57.

8. Morady F, DiCarlo LA, Baerman J, et al. Compari-son of coupling intervals that induce clinical andnonclinical forms of ventricular tachycardia dur-ing programmed stimulation. Am } Cardiol 1986;57:1269-1273.

9. Kou WH, deBuitleir M, Kadish A. et al. Sequelaeof nonsustained polymorphic ventricular tachy-cardia induced during programmed ventricularstimulation. I Am Coll Cardiol 1989; 64:1148-1151.

10. Kalbfleisch JD, Pretice RL. Statistical Analysis ofFailure Time Data. New York, John Wiley & Sons,Inc., 1980, pp. 70-117.

11. Veltri EP, Platia EV, Griffith LSC, et al. Pro-grammed electrical stimulation and long-term fol-lowup in asymptomatic, nonsustained ventriculartachycardia. Am J Cardiol 1985; 56:309-314.

12. Sulpizi AM, Friehling TD, Kowey PR. Value ofeiectrophysiologic testing in patients with nonsus-tained ventricular tachycardia. Am J Cardiol 1987;59:841-845.

13. Kharsa MH, Gold RL, Moore H, et al. Long-termoutcome following programmed electrical stimula-tion in patients with high-grade ventriculeir ec-topy. PACE 1988; 11:603-609.

14. Gomes JAC, Hariman RI, Kang PS, et al. Pro-

grammed electrical stimulation in patients withhigh grade ventricular ectopy: Electrophysiologicfindings and prognosis for survival. Circulation1984; 70:43-51.

15. Klein RG, Machell G, Use of electrophysiologictesting in patients with nonsustained ventriculartachycardia: Prognostic and therapeutic implica-tions. ) Am Coll Cardiol 1989; 111:860-867.

16. Hammill SC, Trusty JM, Wood DL, et al. Influenceof ventricular function and presence or absence ofcoronary artery disease on results of electrophsio-logic testing for asymptomatic nonsustained ven-tricular tachycardia. Am J Gardiol 1990; 65:722-728.

17. Kowey PR, Waxman HL, Greenspon A, et al., andthe Philadelphia Arrhythmnia Group. Value ofelectrophysiologic testing in patients with previ-ous myocardial infarction and nonsustained ven-tricular tachycardia. Am J Cardiol 1990; 65:594-598.

18. Rae AP, Greenspan AM, Spielman SR, et al. Antiar-rhythmic drug efficacy for ventricular tachyar-rhythmias associated with coronary artery diseaseas assessed hy electrophysiologic studies. Am JGardiol 1985; 55:1494-1499.

19. Wilher DJ, Olshansky B, Moran JF, et al. electro-physiological testing and nonsustained ventriculartachycardia: Use and limitations in patients withcoronary artery disease and impaired ventricularfunction. Circulation 1990; 82:350-358.

20. Kron J, Hart M, Schual-Berke S, et al. Idiopathicdilated cardiomyopathy: Role of programmed elec-trical stimulation and Holter monitoring in pre-dicting those at risk of sudden death. Chest 1988;93:85-90.

21. Das SK, Morady F, DiCarlo L, et al. Prognostic use-fulness of programmed ventricular stimulation inidiopathic dilated cardiomyopathy without symp-tomatic ventricular arrhythmias. Am J Cardiol1986; 58:998-1000.

22. Luu M, Stevenson WC, Stevenson LW, et al. Di-verse mechanisms of unexpected cardiac arrest inadvanced heart failure. Girculation 1989; 80:1675-1680.

23. Gottlieb C, Josephson ME. Programmed stimula-tion in the evaluation of life-threatening or poten-tially life-threatening ventricular arrhythmias. JGardiovasc Drugs Ther 1987; 1:155-159.

24. Poole JE, Mathisen TL, Kudenchuk PJ, et al. Long-term outcome in patients who survive out of hospi-tal ventricular fibrillation and undergo electro-physiologic studies: Evaluation hy electrophysio-logic subgroups. J Am Coll Cardiol 1990; 16:657-665.

25. The Gardiac Arrhythmia Suppression Trial (CAST)Investigators. Effect of encainide and flecainide onmortality in a randomized trial of arrhythmiasuppression after myocardial infarction. N Engl JMed 1989; 321:406-412,

1050 May 1993, Part I PAGE. Vol. 16