Standardized Stimulation Protocol to Predict theLong-Term Success of Radiofrequency Ablation ofPostinfarction Ventricular Tachycardia

DAVID O’DONNELL, JOHN P. BOURKE, and STEVE S. FURNISSFrom the Department of Academic Cardiology, Freeman Hospital, Newcastle-upon-Tyne, England

O’DONNELL, D., ET AL.: Standardized Stimulation Protocol to Predict the Long-Term Success of Radiofre-quency Ablation of Postinfarction Ventricular Tachycardia. Background: The ability to predict the successof radiofrequency ablation (RFA) is an essential step in the management of ventricular tachycardia (VT) inpatients with ischemic heart disease. Methods: This study tested a standardized programmed stimulationprotocol and pre-specified definitions of procedural outcome. Consecutive patients referred for RFA ofdelayed post infarction VT were enrolled. Programmed stimulation was performed at the beginning andthe end of an RFA procedure, and consisted of an 8 beat drive followed by up to 5 extrastimuli. Immediatesuccess was defined as no inducible monomorphic VT, and a modified result was defined as the inducibil-ity of VT with >2 extrastimuli beyond those required at baseline. Procedural failure was defined whenthese criteria were not met. Recurrences of sustained VT and arrhythmic deaths were monitored duringlong-term follow-up. Results: The study enrolled 112 patients. Immediate procedural success was achievedin 38%, a modified result in 34%, and procedural failure in 28% of patients. During a mean follow-upof 78 ± 16 months, recurrent sustained VT was observed in 25 patients. VT recurrence was 3% (3/79) inpatients with a successful or modified result, compared with 67% (22/33) in those who had undergoneunsuccessful procedures (P < 0.001). Conclusions: This standardized stimulation protocol and definitionsof procedural success, enabled us to predict with high accuracy a VT recurrence-free long-term follow-up.This may have implications in recommending devices or other treatments after RFA for postinfarction VT.(PACE 2003; 26[Pt. II]:348–351)

ventricular tachycardia, catheter ablation, myocardial infarction

IntroductionVentricular tachycardia (VT) is a complica-

tion of myocardial infarction (MI) associated withhigh morbidity and mortality.1 Prior studies haveshown that radiofrequency catheter ablation (RFA)of post MI VT can be performed with a satisfactoryprocedural success rate.2−5 One concern has beenthe later recurrence of VT after an initially suc-cessful ablation procedure.4,5 The ability to pre-dict the long-term outcome at the end of the initialprocedure is essential in advising patients and indetermining the need for additional therapy.

Programmed electrical stimulation can pre-dict risk in patients with coronary artery disease.6,7

The commonly used “dual site” protocol involvesa drive from 2 sites, at 2 cycle lengths with upto 3 extrastimuli,6 whereas the Westmead stim-ulation protocol used in this analysis involves adrive from a single site with up to 5 extrastim-uli.7 The sensitivity and specificity of the West-mead protocol as well as the variability of VT in-duction requirements have been clearly defined.8

Address for reprints: David O’Donnell, Department of Aca-demic Cardiology, Freeman Hospital, Heaton Road, Newcastle-upon-Tyne, NE77DN, England. Fax: 44-0191-213-0498; E-mail:odonnell˙research@hotmail.com

By incorporating the level of biologic variability,certain changes in induction requirements repre-sent an antiarrhythmic effect of the interventionrather than chance.

The aim of this study was to apply the West-mead criteria of inducibility to define outcome fol-lowing RFA and in particular to predict which pa-tients would remain free from recurrent VT.

MethodsPatient Population

The study enrolled consecutive patients whounderwent RFA for the delayed development ofpost-MI VT. Prior to RFA all patients underwentcoronary angiography and transthoracic echocar-diography along with investigations to exclude re-versible factors potentially contributing to VT.

Electrophysiological Study and AblationProtocol

Electrophysiological study and RFA wereperformed under local anesthesia and sedation.Femoral arterial pressure, oxygen saturation, andelectrocardiogram were continuously monitoredthroughout. Given the urgency of some clinicalsituations, antiarrhythmic drugs were often con-tinued up to the time of the procedure. During

348 January 2003, Part II PACE, Vol. 26

PREDICTING SUCCESS OF VT ABLATION

the procedure no antiarrhythmic drugs were givenuntil after completion of the final stimulationprotocol.

Ablation sites were determined by a combina-tion of activation and entrainment mapping duringVT as well as pace- and substrate mapping in si-nus rhythm. The ablation procedure consisted ofcreating linear lesions with 4 mm tip, 8 mm tip, orirrigated tip catheters, targeting the common isth-mus zone and exit sites of all inducible VTs.

Stimulation Protocol

Prior to RFA, the stimulation protocolwas per-formed with a quadripolar catheter placed at theright ventricular apex, delivering 2 ms/10 mApulses from a custom-built stimulator. The proto-col consisted of the delivery at the right ventric-ular apex of 8-cycles between 400 and 600 ms,followed by extrastimuli introduced at decreas-ing cycle lengths until refractoriness was reached.The level of stimulation required to induce VT wasrecorded.

At the end of the RFA procedure, the stimula-tion protocol was repeated and completed if the re-fractory period of 5 extrastimuli was reached with-out induction of sustained tachycardia. No furtherRFA was performed after the final stimulation pro-tocol. Procedural success was defined by compar-ing the results of the stimulation protocol beforeand after ablation.

Immediate success—No monomorphic VT in-duced at refractoriness of 5 extrastimuli.

Modified result—Monomorphic VT inducible by 2extrastimuli beyond the number used at base-line.

Failed procedure—Monomorphic VT induciblewith the same protocol as at baseline.

All patients were followed at 1, 6, and 12 monthsfollowing the procedure, and at yearly intervalsthereafter.

Statistical Analysis

Continuous data are expressed as mean ±SD. Individual group comparisons were made bypaired t-tests. Dichotomous variables were exam-ined by Chi-square tests. P < 0.05 was consideredsignificant.

ResultsClinical Characteristics

The population of 112 patients was 56 ± 13of age and included 95 men. The detailed charac-teristics of the group have been published previ-

Table I.

Clinical Characteristics Mean

Age (years) 64 ±13 Range (37–88)Male 95LVEF > 35% 23%LVEF < 35% 42%NYHA Class 1 22%NYHA Class 2, 3 or 4 78%Number of episodes 12.1 ± Range (0-150)

in last MonthAnterior scar 41%Inferior scar 35%

Clinical characteristics of the patients enrolled in the study.LVEF = left ventricular ejection fraction; NYHA = New YorkHeart Association.

ously and are summarized in Table I.9 Follow-upwas complete in all patients, at a mean duration of72 ± 16 months.

Electrophysiogic Characteristics and AblationProcedure

More than 1 morphology of VT was in-duced in 77% of patients (mean = 2.4 morpholo-gies/patient). Hemodynamically unstable VT re-quired mapping and ablation to be performed insinus rhythm in 30% of patients.

Induction Protocols

All patients had VT ongoing or induced atbaseline. VT was incessant in 7 patients, inducedwith a single extrastimulus in 51, with 2 extrastim-uli in 36, and with 3 extrastimuli in 18 patients. Nopatient required >3 extrastimuli to induce VT atbaseline. The final stimulation protocol was com-pleted in 110 of the 112 procedures. In the remain-ing 2 cases the procedure was complicated by car-diac tamponade and urgent pericardiocentesis wasrequired. For the purpose of analysis, both caseswere classified as failed procedures.

According to the definitions of procedural out-come, immediate success was achieved in 38%,a modified result in 34%, and immediate failurein 28% (Table II). The number of extrastimuli re-quired to induce VT increased by an average of 2.6in the group of patients with a modified result, andby an average of 0.4 in patients with a failed pro-cedure. In 11 patients VT was inducible with thesame number of extrastimuli as at baseline, andin 3 patients VT was inducible with one less ex-trastimulus than at baseline (Figs. 1 and 2).

PACE, Vol. 26 January 2003, Part II 349

O’DONNELL, ET AL.

Table II.

Number Free From VT Recurrent VT

Acute Success 42 40 (95%) 2 (5%)Modified 37 36 (97%) 1 (3%)Acute or Modified 79 76 (96%) 3 (4%)Failed 33 11 (33%) 22 (66%)

Comparison of procedural success and long-term outcome.VT = ventricular tachycardia.

Predicting Long-term Outcome from the Resultsof the Stimulation Protocol

VT recurred during follow-up in 25 patients.Of these 25 recurrences, 22 occurred after afailed RFA procedure, 2 after a modified result,and 1 after a procedural success. The VT recur-rence rates differed significantly between the com-plete/modified success group and the proceduralfailure group (P < 0.001, Table II, Fig. 3).

An antiarrhythmic drug was prescribed to58% of patients upon discharge from the hospital,and 14 patients had received an ICD prior to theprocedure. All patients who had undergone un-successful procedures were offered an ICD, and14 patients subsequently had an ICD implanted.When standardized for procedural success, the useof antiarrhythmic drugs or ICDs had no effect onlong-term outcome.

Figure 1. Details of 54 patients who were inducible witha single extrastimulus during baseline testing. Followingablation, 21 patients had no inducible tachycardia, 22patients had VT inducible with 3, 4, or 5 extrastimuli,and were classed as a modified result. The 11 patientsinducible with 1 or 2 extrastimuli were classed as havinghad a failed procedure.

Figure 2. Changes in VT induction requirements in 58patients who had incessant VT or were inducible with2 or 3 extrastimuli at baseline testing. The frequency ofeach of the outcomes is indicated by the thickness of therespective line.

DiscussionThis analysis suggests that, following RFA for

post-MI VT, the long-term recurrence of VT can beaccurately predicted by the procedural endpoint.In previous studies, recurrences of VT after suc-cessful RFA procedures have varied between 15%–40%, depending upon definitions used.2−5,9 The<4% VT recurrence rate in this series is signif-icantly lower than in previous reports. The def-initions of procedural success derived from thestandardized Westmead stimulation protocol are

Figure 3. Kaplan Meier cumulative rates of freedomfrom ventricular tachycardia (VT) recurrence compar-ing patients with successful, modified and failed ab-lation procedures. There was a significantly higher re-currence rate of VT after failed than after successful ormodified procedures.

350 January 2003, Part II PACE, Vol. 26

PREDICTING SUCCESS OF VT ABLATION

the singularity of this series and may explain thehigher predictive accuracy.

Predictive tests are challenged by opposingsensitivity and specificity. We believe it most im-portant to minimize the number of false nega-tives (patients classified as having undergone asuccessful procedure who develop recurrent VT)even if it increases the number of false positives.This stimulation protocol had a negative predic-tive value of 97% and a positive predictive valueof 68%. Despite the importance placed on avoid-ing false positives, the procedural success rateof 72% in this study was comparable with otherreports.2−5

Stimulation Protocol

Previous studies have defined the repro-ducibility of the Westmead stimulation protocol.The induction of VT by programmed stimulationrarely varies by >1 extrastimulus. A change ininduction requirements of ≥2 extrastimuli wouldhave a >95% probability of being beyond the rangeof chance variability.8 These data were behindthe definitions of procedural success used in thisstudy. Although the variability of VT inductionby the dual site protocol has been studied,10 theresults have not been incorporated into the def-initions of the procedural success used by othergroups.

Though several protocols have been validatedafter MI,6,7 none have been assessed randomly af-ter RFA. This study supports the value of the West-mead stimulation protocol in predicting long-termoutcome after RFA.

Definitions of SuccessThe noninducibility of clinical VT is often re-

garded as a marker of success following RFA. How-ever, even after successful ablation of all clinical

VTs, the presence of inducible nonclinical VT isassociated with an increased incidence of recur-rent VT.9 Indeed, following ablation of all clini-cally documented morphologies of VT in our se-ries, further morphologies, which had not beendocumented clinically, were inducible in 83% ofpatients. The induction of nonclinical tachycar-dias may be the result of multiple potential exitsites that surround the critical VT isthmus zone.11

Ablation of the dominant VT exit may force an al-ternative exit with subsequent change in surfaceelectrocardiographic morphology, without neces-sarily affecting the critical components of the VTcircuit. Because of the difficulties of recognitionand definition of clinical VT, the definitions of suc-cess used in this protocol relied on the level ofstimulation required to induce VT rather than themorphology of VT.

Limitations of the StudyThough a high correlation was observed be-

tween findings of the standardized stimulationprotocol and long-term recurrence of VT, the strat-egy of extensive linear substrate ablation targetingall inducible VTs may have contributed to the highpredictive accuracy. This highly effective protocolin our institution needs to be validated by othergroups.

ConclusionA validated stimulation protocol and pre-

specified definitions of RFA procedural outcomeallowed the accurate prediction of long-term free-dom from recurrent VT. Patients with either a com-plete success or a modified result had a <5% riskof recurrent VT during follow-up. If these resultsare confirmed by others, the systematic implan-tation of ICDs after successful RFA may not benecessary.

References1. Kannel WB, Sorlic P, McNamara PM. Prognosis after initial my-

ocardial infarction: The Framingham Study. Am J Cardiol 1979;44:53–59.

2. Harada T, Stevenson WG, Kocovic DZ, et al. Catheter ablation ofventricular tachycardia after myocardial infarction: Relation of en-docardial sinus rhythm late potentials to the reentry circuit. J AmColl Cardiol 1997; 30:1015–1023.

3. Stevenson WG, Friedman PL, Kucovic D, et al. Radiofrequencycatheter ablation of ventricular tachycardia after myocardial in-farction. Circulation 1998; 98:308–314.

4. Gonska BD, Cao K, Schaumann A, et al. Catheter ablation of ven-tricular tachycardia in 136 patients with coronary artery disease:Results and long-term follow-up. J Am Coll Cardiol 1994; 24:1506–1514

5. Rothmann SA, Hsia HH, Cossu SF, et al. Radiofrequencycatheter ablation of postinfarction ventricular tachycar-dia: Long term success and the significance of induciblenon-clinical arrhythmias. Circulation 1986; 73:1239–1247.

6. Buxton AE, Lee KL, DiCarlo L, et al. Electrophysiologic testing to

identify patients with coronary artery disease who are at risk forsudden death. N Engl J Med 2000; 342:1937–1945.

7. Bourke JP, Richards DA, Ross DL, et al. Routine programmed elec-trical stimulation in survivors of acute myocardial infarction forprediction of spontaneous ventricular tachyarrhythmias duringfollow-up: Results, optimal stimulation protocol and cost-effectivescreening. J Am Coll Cardiol 1991; 18:780–788.

8. Cooper MJ, Koo CC, Skinner MP, et al. Comparison of immedi-ate versus day to day variability of ventricular tachycardia induc-tion by programmed stimulation. J Am Coll Cardiol 1989; 13:1599–1607.

9. O’Donnell D, Bourke JP, Anilkumar R, Simeonidou E, Furniss SS.Radiofrequency Ablation for post infarction ventricular tachycar-dia. Eur Heart J In Press.

10. Lombardi F, Stein J, Podrid PJ, et al. Daily reproducibility of elec-trophysiologic test results in malignant ventricular arrhythmias.Am J Cardiol 1986; 57:96–101.

11. de Bakker JCT, Capelle FJL, Janse MJ, et al. Macroreenrty in theinfarcted human heart: Mechanism of ventricular tachycardia witha focal activation pattern. J Am Coll Cardiol. 1991; 18:1004–1008.

PACE, Vol. 26 January 2003, Part II 351