Outcome of patients with nonsustained ventricular tachycardia and severely impaired ventricular function who have negative electrophysiologic studies

Twenty-nine consecutive patients with a prior myocardial infarction, severely reduced left ventricular ejection fraction (26% _ 8%), and asymptomatic nonsustained ventricular tachycardia were enrolled in a prospective trial. After a negative programmed electric stimulation study (3 extrastimuli at 2 sites with 2 drive trains), the 26 men and 3 women (mean age 71) were monitored for a mean of 13 months without antiarrhythmic drug therapy. Five patients died suddenly or had sustained ventricular tachycardia; three others had a cardiac, nonarrhythmic death. Events occurred in the first 13 months of the surveillance period. Clinical factors associated with a poor outcome included congestive heart failure and lack of/~-blocker therapy. In addition, patients with events tended to have lower ejection fractions than those without (21% vs 28%, p not significant). Thus a negative programmed electric stimulation study does not necessarily imply a benign outcome in patients with a prior infarction and nonsustained ventricular tachycardia if they also have severe left ventricular dysfunction and a history of heart failure. These data have important implications for the design and conduct of contemporary clinical trials. (AM HEART J 1995;129:492-6.)

Maribel Hern~ndez, MD, Joan Taylor, LPN, Roger Marinchak, MD,

Seth Rials, MD, PhD, Andrew Rubin, MD, and Peter Kowey, MD Wynnewood and Philadelphia, Pa.

The clinical approach to patients with nonsustained ventricular tachycardia (NSVT) after myocardial infarction has attracted much interest, in part be- cause of the evidence that treatment has significant liability. 1-2 Nevertheless, because NSVT is associ- ated with an increased risk of sudden and cardiac death, at tempts to select patients at highest risk have intensifiedP -4 A logical tool is electrophysiologic study. It seems intuitive that the ability to induce a sustained arrhythmia in this cohort might indicate that there is a greater chance that the arrhythmia would occur spontaneously. 5 Unfortunately, this hy-

From the Division of Cardiovascular Diseases, Lankenau Hospital and Medical Research Center Wynnewood; and the Department of Medicine, Thomas Jefferson University, Philadelphia.

Received for publication April 1, 1994; accepted May 20, 1994.

Supported in part by the Mary L. Smith Charitable Leed Trust, Philadel- phia. Reprint requests: Peter R. Kowey, MD, Division of Cardiovascular Diseases, Lankenau Hospital and Medical Research Center, 100 Lancaster Avenue, Room 128, MRB, Wynnewood, PA 19096.

Copyright ® 1995 by Mosby-Year Book, Inc. 0002-8703/95/$3.00 + 0 4 /1 /60428

pothesis has been difficult to prove, in part because a large number of patients have been treated on the basis of the results of the electrophysiologic study or despite them. 6 In addition, many studies have in- cluded diverse populations of patients or were them- selves retrospective surveys. 711

To obtain a clearer answer, we carried out a prospective study in which all patients with prior in- farction and NSVT underwent programmed electric stimulation. Of 81 patients studied over a 2-year pe- riod, 29 had no arrhythmia induced and were moni- tored without antiarrhythmic drug therapy. This is a report of the results of that longitudinal experience.

METHODS

The protocol was reviewed and approved by the Institu- tional Review Board for clinical studies of the Lankenau Hospital and Medical Research Center. All patients signed

a consent form for the procedure and agreed to be con- tacted periodically to gather follow-up information. Those patients who had had a documented myocardial infarction and evidence of NSVT on ambulatory monitoring were candidates for the study. NSVT was defined as three or

492

Volume 129, Number 3

American Heart Journal Hern~ndez et al. 493

* 29 1 O0

90

80 2 % 7o

~ 60 LIJ

5O

~ 40

~ 30

20

10

0 ~ 0 2

2~28 ~27 2 21

15 12 9

I I F J ! _ I ! I I ~ I I

4 6 8 10 12 14 16 18 20 22 24

Months

* Number of patients

Fig. 1. Kap•an-Meiersurviva•curvei••ustratespercentage•fpatientswh•remainedeventfree•verc•urse of follow-up. Numbers on curve indicate number of patients actually available for follow-up at each time point.

more ventricular beats in succession occurring at a rate >100 beats/min in the absence of coincidental antiarrhyth- mic drug therapy. Patients may have had palpitations with their arrhythmia, but those with more severe symptoms were excluded. Also excluded were those with a recent (<1 month) myocardial infarction, a history of syncope, or a terminal illness. Each enrolled patient was evaluated non- invasively by, at the minimum, noninvasive testing to quantitate left ventricular function and to exclude active myocardial ischemia. We then carried out a standard pro- grammed electric stimulation study. Up to three sequential extrastimuli were delivered decrementally to refractoriness during two ventricular drive cycle lengths at each of two right ventricular sites. All studies also included burst right ventricular pacing. Those patients who had no more than 15 beats of any form of ventricular tachycardia despite completion of the entire stimulation protocol were enrolled and monitored without antiarrhythmic drug therapy. No patients in the series had a history of symptomatic brady- cardia, nor did they have evidence of severe distal conduc- tion disease at their baseline electrophysiologic study. The follow-up program consisted of either periodic office or telephone contacts, during which symptoms were assessed and events tallied. We also inquired as to changes in med- ical therapy and the reasons for intervention. In cases in which the patient/family could not be contacted, we con- ferred with the referring physician. Adequate follow-up information was obtained in every case.

For the purposes of the analysis, sudden death was de- fined as death occurring precipitously with a witness present or during sleep without antecedent clinical deteri- oration. Data in text in table represent mean _+ SD. Inter- group comparisons were carried out with Fisher's exact test. Graphic representation of the follow-up data were il- lustrated with a Kaplan-Meier plot.

RESULTS

The s tudy populat ion had a mean age of 71 years and included 26 men and 3 women. Mean left ventr icular ejection fract ion was 26 % +_ 8 %. All pa- t ients had had myocardial infarct ion a mean of 48 months before the study; nonsusta ined ventr icular tachycardia was documented by ambula tory moni- toring in 15 pat ients and by in-hospital t e lemet ry in 14 patients. After a negative-result electrophysio- logic study, all pat ients were moni tored wi thout an- t ia r rhythmic drug the rapy for 3 through 30 months (mean 13 months) . Eight pat ients had a clinical event during the follow-up period. Events were sudden dea th in 4, sustained ventr icular tachycardia in 1, and hear t failure dea th in 3 others. A Kaplan-Meier curve of event-free survival is i l lustrated in Fig. 1. Note tha t a r rhy thmic events occurred in the first 13 months of follow-up. Clinical informat ion comparing data in pat ients who did and did not have an a r rhy thmic event af ter their negative electrophysiologic s tudy is presented in Table I. Note tha t there were no signif- icant differences between these groups in global left ventr icular ejection fraction, al though pat ients who did not have events tended to have slightly be t te r left ventr icular funct ion than those who did. We did not discern any difference between groups in the length of the runs of nonsusta ined V T al though the small number of pat ients l imited this analysis.

All eight pat ients with events had congestive hear t failure compared~to 57 % of the no-event group, None of the pat ients who had events had positive exercise tests, suggesting tha t ischemia was not the explana-

March 1995

494 Herndndez et al. American Heart Journal

Table I. Clinical data grouped by outcome

E v e n t N o even t (n = 8) (n = 21)

LVEF 21% 28 % CHF* 8/8 12/21 Positive SAECG (3/3 criteria) 5/6 9/16 Positive ETT result 0/8 5/21 3-Vessel CAD 3/5 13/19

CAD, Coronary artery disease; CHF, congestive heart failure; LVEF, left ventricular ejection fraction; ETT, exercise treadmill test; SAECG, signal- averaged electrocardiograms. *p < 0.05.

Table Ih Cardiovascular medication use in total cohort and grouped by outcome (n -- 29)

Total E v e n t No even t even ts (n = 8) (n = 21)

fl-Blocker 7 0 7 ACE inhibitor 20 8 12 Calcium blocker 4 1 3 Nitrates 11 3 8 Digoxin 8 3 5

No significant intergroup difference for any comparisons. ACE, Angioten- sin-converting enzyme.

tion for their deaths. In Table II we have listed car- diovascular medications used by the two groups. None of the event-positive patients received ~-block- ers, but all were receiving angiotensin-converting enzyme inhibitors.

DISCUSSION

Because NSVT is a marker for an adverse outcome in patients who have had a myocardial infarction and because recent studies have indicated a distinct dis- advantage of suppressive antiarrhythmic drug ther- apy, the clinician has turned increasingly to tech- niques to quantitate risk in individual patients. 12-1u So far, we have learned that well-preserved left ven- tricular function and the absence of a late potential on a signal averaged electrocardiogram bode well. 1415 The true dilemma comes in the approach to the pa- tients who do not have the good fortune to have these findings. Some physicians have elected to subject these patients to programmed stimulation on the premise that~oninducibil i ty might portend a better outcomeJ 6 In fact, this assumption may not be appropriate. In reviewing the results of major trials, there is no consensus on this issue. Although most investigators have found that noninducibility is as- sociated with fewer events in follow-up, not all have.

When the major studies were subjected to a metaanal- ysis, there was a statistically significantly greater number of events in the patients who had been found inducible compared to those who were not (18% vs 7 %; p < 0.001). 17 The positive predictive accuracy was only 18%, but the negative predictive was a very good 93 %. However, the studies had the usual flaws of mixed groups, varied protocols, and frequent use of antiarrhythmic drugs. In addition, the clinical sig- nificance of the findings could seriously be ques- tioned; is it really worth the cost, risk, and inconve- nience to subject this large group of patients to a test that has marginal ability to discriminate patients who might have an arrhythmic event, especially be- cause there is little guidance as to what to do with those patients who do have inducible arrhythmia? is

Clearly there has been a need for prospective stud- ies of pure patient populations in which decisions re- garding treatment are made a priori. This is a report of such a study. We enrolled patients who had ischemic cardiomyopathy and NSVT and monitored those patients who had a negative programmed elec- tric stimulation study result. Our enrollment in- cluded mostly patients with severe left ventricular dysfunction. We managed to monitor all of these pa- tients without antiarrhythmic drug therapy. To our surprise, 5 of the 29 patients had an arrhythmic event during the follow-up. This 17% annual event rate is substantially higher than the 4.7% figure reported in the literature metaanalysis for noninducible pa- tients. There are several possible reasons for this dif- ference. The literature metaanalysis contained pa- tients with a variety of forms of heart disease stud- ied under diverse protocols. In addition, 13 % of the noninducible patients were treated with antiarrhyth- mic drugs. Because the patients with the worst arrhythmias may have been selected for therapy, their outcome could have been altered favorably. The 17 % figure for noninducible patients in the present study is quite consistent with the 15% annual event rate reported in the largest available study in the lit- erature, which provided information on 205 people who had had a myocardial infarction and manifested NSVT. 6

Our results stand in direct contradiction to the study published by Wilber et alJ 9 in 1990. The patient populations appear comparable, as do the stimulation protocols. They reported a 2 % annual arrhythmic event rate in patients who did not have inducible sustained VT. We strongly suspect that our patients were sicker in that all had a history of heart failure and none were receiving ~-blockers, although we had encouraged such therapy. In fact, we do not

Volume 129, Number 3

American Heart Journal Hern~ndez et al. 495

know the exact mechanism for collapse in our pa- tients. People with poor left ventricular function and congestive failure are susceptible to malignant bradyarrhythmias and tachyarrhymias. 2° In any case, a history of congestive heart failure has been a pow- erful predictor of total mortality in several patient populations, including those who have asymptomatic ventricular arrhythmia after a myocardial infarc- tion. 21 Our results are not substantially different from the reported mortality rates in patients who have had manifest heart failure in other clinical set- tings. ~2 Angiotensin-converting enzyme inhibitor therapy was used in a large percentage of our cohort, including all of the patients who later had events. Angiotensin-converting enzyme inhibition, although having an impact on total mortality in such patients, has not been associated with a reduction in sudden death incidence in postinfarction trials. 23

Our study did not include patients with inducible arrhythmias. The pressure to treat an inducible pa- tient in our practice is substantial because the patients are usually referred with the expectation that a positive test may be acted on. The true signif- icance of our findings remains unknown without this information. Fortunately, multicenter trials are un- derway to enroll patients into a follow-up program in which treatment is randomized and which will in- clude a nontreatment arm. Early data from these tri- als confirm that a very large number of patients must be recruited and screened to complete such a project. Our study may help investigators designing larger trials. It is clear that large patient populations need to be selected and described carefully and that all patients need to be monitored quite carefully over a long time period. The lack of a systematic in-person follow-up program will limit definite conclusions be- cause transient risk factors such as electrolyte ab- normalities or worsened heart failure can contribute to electric destabilization, especially in patients who have severely impaired cardiac function. For now, we conclude that it is not safe to assume that noninduc- ibility is necessarily a prelude to an event-free follow-up.

Conclusion. Patients with NSVT and reduced left ventricular function are now being routinely sub- jected to programmed stimulation on the assumption that noninducibility augers well for their outcome. This recommendation has been based on studies in the literature that have been inconsistent at best. We recommend a reconsideration of these advise- ment until the results of larger trials are available. In addition, we urge that the design of these multicenter efforts include rigorous follow-up of all patients, in-

cluding those with negative electrophysiologic study results.

We t h a n k Rose Marie Wells for her usua l pa t i en t he lp in the prepara t ion of the m a n u s c r i p t and the referr ing phys ic ians who pe rmi t t ed the enro l lment of their pa t ients .

REFERENCES

1. The Cardiac Arrhythmia Suppression Trial Investigators. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989;321:406-12.

2. Furberg CD. Effect of antiarrhythmic drugs on mortality after myocar- dial infarction. Am J Cardiol 1983;52:32C-36C.

3. The Multicenter Post-Infarction Research Group. Risk stratification and survival after myocardial infarction. N Engl J Med 1983;309:331-6.

4. Mukharje J, Rude RE, Poole WK, Gustafson N, Thomas L, Strauss HW, Jaffe A, Muller J, Roberts R, Raabe D, Croft L, Passamani E, Braun- ward E, Willerson J, the MILIS Study Group. Risk factors for sudden death after myocardial infarction: two-year follow-up. Am J Cardiol 1984;54:31-6.

5. Podrid P, Bumio F, Fogel R. Evaluating patients with ventricular arrhythmia: role of signal averaged electrocardiogram, exercise test, ambulatory electrocardiogram and electrophysiologic studies. Cardio- vasc Clin 1992;10:371-95.

6. Kowey PR, Waxman H, Greenspon A, Greenberg R, Poll D, Kutalek S, Gessman L, Muenz L, the Philadelphia Arrhythmia Group. Value of electrophysiologic testing in patients with previous MI and nonsus- rained ventricular tachycardia. Am J Cardiol 1990;65:594-8.

7. Klein RC, Macheil C. Use of electrophysiologic testing in patients with nonsustained ventricular tachycardia: prognostic and therapeutic im- plications. J Am Coll Cardiol 1989;14:155-61.

8. Hammill SC, Trusty JM, Wood DL, Bailey KR, Vatteratt P J, Osborn MJ, Holmes DR, Gersh BJ. Influence of ventricular function and pres- ence or absence of coronary artery disease on results of electrophysio- logic testing for asymptomatic nonsustained ventricular tachycardia. Am J Cardiol 1990;65:722-8.

9. Zheutlin TA, Roth H, Chua W, Steinman R, Summers C, Lesch M, Ke- hoe RF. Programmed electrical stimulation to determine the need for antiarrhythmic therapy in patients with complex ventricular ectopic activity. AM HEART J 1986;111:860-7.

10. Suplizi AM, Friehling TD, Kowey PR. Value of electrophysiologic test- ing in patients with nonsustained ventricular tachycardia. Am J Cardiol 1987;59:841-5.

11. Kadish A, Schmaltz S, Caikins H, Morady F. Management of nonsus- tained ventricular tachycardia guided by electrophysiological testing. PACE 1993;16:1037-50.

12. Marchlinski FE, Buxton A, Waxman H, Josephson M. Identifying pa- tients at risk for sudden death after myocardial infarction: value of re- sponse to programmed stimulation, degree of ventricular ectopic activ- ity and severity of left ventricular dysfunction. Am J Cardiol 1983; 52:1190-6.

13. Mintz R, Marinchak RA, Rials SJ, Kowey PR. Clinical management of patients with nonsustained ventricular tachycardia: what do we really know? Cardiology 1990;77:166-80.

14. Bigger JT, Fleiss JL, Klieger R, Miller TP, Rolnitzky L, the Multi-cen- ter Post-Infarction Research Group. Relationship among ventricular arrhythmias, LV dysfunction and mortality in two years after myocar- dial infarction. Circulation 1984;69:250-8.

15. Kuchar DL, Thorburn CW, Samuel NL. Late potentials detected after myocardial infarction: natural history and prognostic significance. Cir- culation 1986;74:1280-9.

16. Manelis AS, Estes NAM. Value of programmed ventricular stimulation in the evaluation and management of patients with nonsustained ven- tricular tachycardia associated with coronary artery disease. Am J Car- dial 1990;65:201-5.

17. Kowey PR, Taylor JE, Marinchak RA, Rials SJ. Does programmed stimulation really help in the evaluation of patients with nonsustained

March 1995

Ferrick et al. American Heart Journal

ventricular tachycardia? Results of a meta-analysis. AM HEART J 1992;123:481-3.

18. Mitra RL, Buxton AE. The clinical significance of nonsustained ventricular tachycardia. J Cardiovasc Electrophysiol 1993;4:490-6.

19. Wilber DJ, Olshansky B, Moran JF, Scanlon PJ. Electropbysiologic testing and nonsustained ventricular tachycardia: use and limitations in patients with coronary artery disease and impaired ventricular function. Circulation 1990;82:350-8.

20. Luu M, Stevenson WG, Stevenson LW, Baron K, Waldon J. Diverse mechanisms of unexpected cardiac arrest in advance heart failure. Cir- culation 1989;80:1675-80.

21. Francis GS. Should asymptomatic ventricular arrhythmias in patients with congestive heart failure be treated with antiarrhythmic drugs? J Am Coll Cardiol 1988;12:274-6.

22. Packer M. Sudden unexpected death in patients with congestive heart failure: a second frontier. Circulation 1985;75:681-5.

23. Pfeffer MA, Braunwald E, Moye LA, Basta L, Brown EJ, Cuddy TE, Davis BR, Geltman EM, Goldman S, Flaker GC, Klein M, Lamas GA, Packer M, Rouleau J, Rouleau JL, Rutherford J, Wertheimer JH, Haw!~ins CM, on behalf of the SAVE Investigators. Effect of captopriI on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 1992;327:669-77.

Prediction of electrophysiologic study results in patients treated with amiodarone

To identify whether electrophysiologic study results during early-phase amiodarone therapy can be predicted by previous electrophysiologic study, we reviewed the electrophysiologic data of 50 patients with inducible sustained ventricular arrhythmias who underwent 4.3 _+ 1.3 drug trials before being given amiodarone. Study results during testing with agents of the modified Vaughan Williams la classification were compared with data obtained after 2 weeks of amiodarone therapy. Partial response by electrophysiologic study was defined as well-tolerated ventricular tachycardia <150 beats/rain associated with a blood pressure -->90 mm Hg. Significant slowing in the rate of induced ventricular tachycardia was seen during therapy with both la agents and amiodarone, although there was a trend toward greater slowing during amiodarone treatment (180 _+ 45 beats/min vs 164 _+ 65 beats/min; p = 0.09). Two of three patients with noninducible ventricular tachycardia during amiodarone showed profound ventricular tachycardia slowing during la therapy. Thirty-eight of 50 patients demonstrated concordance of electrophysiologic study results with regard to achieving partial response criteria. Twenty patients died during a mean follow-up period of 37 _+ 29 months; 7 of the 10 sudden deaths occurred in patients who did not meet partial response criteria. We conclude that patients with inducible sustained ventricular arrhythmias failing serial drug testing with la agents only rarely have their ventricular tachycardia suppressed during amiodarone therapy. Partial response criteria are often concordant between testing on agents of the la classification and amiodarone, and there was no significant difference in survival in patients based on their partial response status. (AM HEART J 1995;129:496-501.)

Kevin J. Ferrick, MD, Sardul Singh, MD, James A. Roth, MD, Soo G. Kim, MD, and

John D. Fisher, MD Bronx, N. Y.

Electrophysiologic testing has been demonstrated to be of use in reducing arrhythmia recurrence and at- rhythmic mortality in patients with aborted sudden

From the Arrhythmia Service, Division of Cardiology, Montefiore Medical Center, Albert Einstein College of Medicine.

Received for publication April 5, 1994; accepted May 23, 1994. Reprint requests: Kevin J. Ferrick, MD, Arrhythmia Service, Montefiore Medical Center, 111 East 210 St., Bronx, NY 10467.

Copyright © 1995 by Mosby-Year Book, Inc. 0002-8703/95/$3.00 + 0 4 /1 /60420

cardiac death and recurrent sustained ventricular tachycardia) -5 Nevertheless, these studies are time consuming, costly, and often involve prolonged hos- pitalization. In an era of increasing cost conscious- ness and limited resources, length of hospital stay could be minimized if serial antiarrhythmic drug testing could be performed so as to minimize the to- tal number of ineffective drug trials. Efficiency would be further maximized if the efficacy of agents with long half-lives of accumulation could be predicted by previous drug testing. We therefore reviewed the

496