Europace (2004) 6, 301e306

by guest on June 1, 2013http://europace.oxfordjournals.org/

Dow

nloaded from

Efficacy and safety of ibutilide vs.transoesophageal atrial pacing for thetermination of type I atrial flutter

Andrea Mazzaa, Maria Stella Feraa, Irma Biscegliaa,Francesca Bettiolb, Giovanni Pulignanoa,), Pietro Tanzib,Carlo Gaudioc, Ezio Giovanninia

aI Division of Cardiology, S. Camillo Hospital, Via G. Livraghil, 00152 Rome, ItalybCentral Service of Cardiology, S. Camillo Hospital, Rome, ItalycCardiology Department, University ‘‘La Sapienza’’, Rome, Italy

Submitted 17 August 2003, and accepted after revision 25 March 2004

Abstract Aims Comparing efficacy and safety of ibutilide vs. transoesophagealatrial pacing (ATP) for the termination of type I atrial flutter (AFL).Methods and results Eighty-seven patients affected by AFL lasting between 2 hand 30 days were randomized in two groups: Group 1di.v. ibutilide treatment, upto 2 mg, and Group 2dATP, with ‘‘burst’’ and ‘‘ramp’’ pacing protocols. Sinusrhythm was restored in 36/45 (80%) patients in Group 1 vs. 18/42 (43%) in Group 2(P!0:0005). In Group 1, mean AFL duration was 11.4 G 7.7 days in respondersvs. 12.1 G 7.6 in non-responders (P ¼ ns), while in Group 2 it was 2.7 G 1.4 vs.14.2 G 5.4 days (responders vs. non-responders, respectively, P!0:0001); 30/36(83%) responders in Group 1 had AFL O48 h vs. 10/18 (56%) responders in Group 2(P!0:05). Non-sustained polymorphic ventricular tachycardia occurred in 2 patientsin Group 1 vs. none in Group 2 (P ¼ ns). It did not require any specific treatmentexcept the interruption of ibutilide infusion.Conclusion Both ibutilide and ATP proved to be safe and effective for recent onsettype I AFL termination, but ibutilide was more effective when the arrhythmia hadlasted longer than 48 h.ª 2004 The European Society of Cardiology. Published by Elsevier Ltd. All rightsreserved.

KEYWORDSatrial flutter;ibutilide;transoesophageal

atrial pacing

) Corresponding author. Tel.: D39-06-5815205; fax: D39-06-5815205.E-mail address: gipulig@tin.it (G. Pulignano).

1099-5129/$30 ª 2004 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.eupc.2004.03.007

302 A. Mazza et al.

by guest on June 1, 2013http://europace.oxfordjournals.org/

Dow

nloaded from

Introduction

Overdrive atrial pacing by transoesophageal ap-proach (ATP) is a safe and effective treatment forrestoring sinus rhythm in patients with type I atrialflutter (AFL), with a success rate of 55% reported inthe literature, especially in the case of arrhythmiasof recent onset [1e5]. However, the efficacy of ATPprogressively decreases as time goes by, up to thepoint of becoming insignificant for AFL the durationof which is over 48 h [6].

Ibutilide, a new antiarrhythmic class III drug[7e9], proved to be effective in 53%e76% of cases[10e12] when administered at full dose, also inpatients with a non-recent onset, despite its usebeing burdened by an incidence of polymorphicventricular tachycardia, requiring electrical defi-brillation in 17% of cases [11].

The purpose of our study was to compare theefficacy and safety of ibutilide vs. ATP in a groupof patients consecutively presenting with AFL, upto 30 days after the onset.

Methods

The study included all the patients we saw betweenJanuary 2000 and March 2002, suffering from ‘‘typ-ical’’ or ‘‘reverse typical’’ AFL (sawtooth patternof F waves in inferior leads or mainly positive andsymmetrical F waves in the same leads, respec-tively, and an absence of an isoelectric line in both),of 2 h to 30 days duration, and no exclusion cri-teria. Exclusion criteria were: atypical AFLmorpho-logies in a 12-lead ECG, haemodynamic instability(systolic blood pressure !90 mmHg, NYHA classIIIeIV), age!18 years, pregnancy, hyperthyroidism,ventricular tachycardia suggested by the clinical in-terview or QTc interval O440 ms in a 12-lead ECG,acute myocardial infarction or heart surgery!30 days, treatment in progress with class I orclass III drugs, average ventricular rate (AVR) !60/min, left ventricular ejection fraction !40%, bodyweight !60 kg.

After performing a transthoracic M-mode andbidimensional echocardiograms with echo-Doppler,and a full haematochemical assessment, thepatients admitted to the study with AFL O48 h,duly treated with anticoagulants, either by intra-venous heparin according to the accelerated pro-cedure, or already under treatment with oralwarfarin, underwent transoesophageal echocardio-gaphy (HewlettePackard mod. SONOS 5500), inorder to exclude the presence of thrombi in the left

atrial appendage and/or left atrium, before beingrandomized.

Patients were randomized in two groups: Group1dtreatment by intravenous ibutilide and GR 2dtreatment by ATP.

In Group 1, ibutilide was given by intravenousinfusion of 1 mg in 10 min, followed by a furtheradministration of 1 mg, after an interval of 10 minif the first dose had not been successful. The pa-tients were monitored by means of a single ECGlead, from the beginning of the ibutilide admin-istration until 4 h later, with periodic assess-ments of the QT and QTc intervals by 12-lead ECG(HewlettePackard, mod. M1700A), at a paperspeed of 50 mm/s, at time 0, 10, 20, 30, 60, 90,120 min and at 4 h from the beginning of the treat-ment. AFL cycle lengths and QT intervals weremeasured to the nearest 10 ms. The criteria forsuspending the ibutilide administration were: ter-mination of the patient’s atrial flutter, systolicblood pressure !90 mmHg, advanced AV block(AVR! 60/min), QRS duration R50% comparedwith the baseline ECG, QTc O600 ms, prematureventricular beats in triplets. The effectiveness ofthe treatment was assessed 90 min after the initia-tion of the first dose of ibutilide.

In Group 2, ATP (MEDTRONIC, Minneapolis, MN,USA, stimulator mod. 5328) was performed settingthe output of the stimulator slightly above the pac-ing threshold. Atrial capture during atrial pacingwas considered to have been achieved when themorphology of the atrial waves changed on thesurface ECG or when ventricular response duringpacing changed. The stimulation protocol consistedof ‘‘burst’’ pacing, 8 s duration, starting with a rateof 130% of the AFL rate, then decreased in steps of10 ms until termination of the AFL was achieved, oruntil a 100 ms pacing cycle length was reached. IfAFL persisted, we proceeded to perform ‘‘ramp’’pacing, with cycles from 250 to 80 ms for 10 s dura-tion, up to five times, either obtaining terminationof AFL (sinus rhythm or atrial fibrillation), or endingthe procedure.

The non-responder patients of both groupsunderwent transthoracic electrical cardioversionby biphasic synchronous DC shock, with a defibrilla-tion procedure of up to three shocks at 70, 120and 150 J.

Statistical analysis

The results are expressed as meansG the standarddeviations. In Group 1, two-tailed Student’s t testwas used to compare continuous variables, derivedfrom 12-lead ECG tracings recorded at baseline,

Termination of type 1 atrial flutter 303

by guest on June 1, 2013http://europace.oxfordjournals.org/

Dow

nloaded from

and from those which showed the maximumlengthening of the AFL cycle or of the QTc interval,independent of the recording time. The chi-squaretest was used to compare categorical variables.Fisher’s exact test was used when appropriate.P values !0.05 were considered significant.

Results

Eighty-seven patients (45 male, 42 female, meanage 57G 10 years) entered the study. They wereall symptomatic with palpitations and sometimeswith exertional dyspea, so that the onset of the ar-rhythmia could be quite well defined in all cases. A‘‘typical’’ morphology of AFL in the 12-lead ECG,with the characteristic sawtooth pattern of flutterwaves in inferior leads and absence of an isoelec-tric line were present in 65 (75%) patients, whilea ‘‘reverse typical’’ morphology with mainly posi-tive and symmetrical flutter waves in the sameleads was observed in 22 (25%) patients. The at-rio-ventricular conduction ratio varied from 2:1to 4:1. Organic heart disease was present in 69(79%): hypertensive in 39 (45%), valvular in 15(17%), ischaemic in 8 (9%), dilated cardiomyopathyin 4 (5%), congenital in 3 (3%), while AFL was iso-lated in 18 (21%). Forty-five patients were random-ized in Group 2, and 42 to Group 2. There were nosignificant statistical differences concerning theclinical and instrumental variables between thetwo groups (Table I). In Group 1 we found a length-ening of the AFL cycle length, from 227G 18to 245G 21 ms (P!0:0001), and of the QTc inter-

Table 1 Clinical and instrumental characteristics ofthe study population

Group 1(n ¼ 45)

Group 2(n ¼ 42)

P

Age ( years) 57G 11 57G 9 NSGender (M/F) 23/22 22/20 NSLeft atrium (mm) 38G 7 39G 5 NSEjection fraction (%) 45G 4 46G 5 NSHeart disease 35 (78%) 34 (81%) NSAtrial fluttercycle length (ms)

227G 18 225G 20 NS

Typical fluttermorphology

33 (73%) 32 (76%) NS

Reverse typicalflutter morphology

12 (27%) 10 (24%) NS

Atrial flutterduration (days)

11.5G 7.7 9.3G 7 NS

QTc interval (ms) 404G 19 402G 22 NS

val, from 404G 19 to 450G 20 ms (P!0:0001)(Fig. 1). In Group 1 we obtained the restoration ofsinus rhythm in 80% of cases (36/45)d71% (32/45)during or immediately after the first infusion ofibutilide in 9% (4/45) during or immediatelyafter the second dosedvs. 43% (18/42) in Group 2(P!0:0005) (Fig. 2). In this group, sinus rhythm res-toration occurred directly in 12% (5/42) of cases,while in 31% (13/42) after a short phase of atrial fi-brillation. In Group 2, atrial capture was obtainedin all patients with a mean output of the stimulatorof 20.5G 2.4 mA (range 16e24 mA) and a fixedduration of 10 ms. None of the patients in thisgroup was symptomatic due to the pacing and nosedation was required. Mean time to conversionin responders of Group 1 was 15G 5 min (range7e32 min). In Group 1, the average AFL durationamong the responders was 11.4G 7.7 vs. 12.1G7.6 days in non-responders (P ¼ ns), while in Group2 it was 2.7G 1.4 for responders vs. 14.2G 5.4 fornon-responders (P ¼ 0:0001) (Fig. 3).

When considering the whole study population(i.e., responders and non-responders), 8 patientshad AFL !48 h both in Group 1 and in Group 2;among them, sinus rhythm was restored in 6/8(75%) in Group 1 vs. 8/8 (100%) in Group 2(P ¼ ns). The remaining 37 patients in Group 1and 34 in Group 2 had AFL O48 h; among them,sinus rhythm restoration was achieved in 30/37(81%) in Group 1 vs. 10/34 (29%) in Group 2(P!0:006) (Fig. 4).

In Group 1 the average lengthening of the QTcinterval was 46G 6 ms, without any statisticallysignificant difference between responders and non-responders (respectively, 46G 7 vs. 47G 4 ms,P ¼ ns).

In two patients of Group 1, non-sustained poly-morphic ventricular tachycardia occurred (at thebeginning of the second infusion in one and duringthe second infusion in the other); in both cases,ibutilide failed to restore sinus rhythm, but theventricular arrhythmias receded once the drughad been suspended, without requiring any specifictreatment. When polymorphic ventricular tachy-cardia occurred QTc intervals were, respectively,450 and 480 ms, with a prolongation of 50 ms withrespect to baseline and with an atrio-ventricularconduction ratio varying from 2:1 to 4:1. Both ofthem had arterial hypertension and the brady-tachycardia syndrome was ruled out by 24-h Holtermonitoring performed after the electrical cardio-version had restored the sinus rhythm. No adverseevent occurred among Group 2 patients, nor amongthose of Group 1 during the 4-h monitoring(P ¼ ns). At the end of the study, the non-respond-ers of both groups underwent external electrical

304 A. Mazza et al.

Dow

100

140

180

220

260

300

Baseline Ibutilide

ms P<0,0001

AFL cycle length

245±21227±18

300

340

380

420

460

500

Baseline Ibutilide

ms P<0,0001

450±20

404±19

QTc interval

Figure 1 Atrial flutter cycle length and QTc interval at baseline and after ibutilide in Group 1 patients (meanvaluesG SD).

by guest on June 1, 2013http://europace.oxfordjournals.org/

nloaded from

cardioversion, which proved effective in restoringthe sinus rhythm in all cases.

Discussion

The success in restoring sinus rhythm by ATP in pa-tients suffering from AFL has been reported exten-sively in the literature, reaching about 55% inrecent onset AFL [3,4], particularly if within 24 h.However, the success percentage of this tech-nique progressively declines with time, until it be-comes insignificant in AFL of over 48 h duration [6].This limit seems to be due to stabilization of thereentrant circuit, which makes penetration of thestimulated wavefront into the circuit more andmore difficult, so preventing pacing from extin-guishing the excitable gap which characterizesAFL. Therefore, this is consistent with our study’ssuccess in restoring, by means of ATP, sinus rhythm

0

10

20

30

40

50

60

70

80

90

100

Gr 1Gr 2

P<0,0005

Figure 2 Rate of sinus rhythm restoration in patients ofGroup 1 vs. Group 2.

in 43% of cases among a group of patients whoseAFL had an average duration of 9 days. Further-more, the average duration of the arrhythmiaamong the responders of that group was 2.7 days,significantly shorter than that recorded amongnon-responders. We, notably, excluded from thestudy patients with atypical morphologies of AFL,just because they are less likely to be interruptedby ATP, irrespective of AFL duration.

Conversely, ibutilide, a new class III antiarrhyth-mic drug, whose electrophysiological effect is thatof increasing the depolarizing slow sodium current,and of blocking the repolarizing outward potas-sium current during phase 3, thus prolonging theduration of the monophasic action potential andof the refractory period [13], proved to be very ef-fective in terminating AFL of longer duration, whengiven in full dose, obtaining positive results from53% to 76% cases [10e12]; the prolongation of therefractory period due to the drug’s administrationmay cause the whole excitable gap of the reentrant

0

5

10

15

20

25

responders non resp responders non resp

2.7+1.4

11.5+7.714.2+5.4

12.1+7.6

p<0,001 p= ns

Group 1 Group 2

AF

L d

urat

ion

(day

s)

Figure 3 Atrial flutter duration (meanG SD) in patientsresponders vs. non-responders in both groups.

Termination of type 1 atrial flutter 305

by guest on June 1, 2013http://europace.oxfordjournals.org/

Dow

nloaded from

circuit to be affected by refractoriness, thus termi-nating the AFL due to the reentrant wavefront’sinability to proceed [14]. Furthermore, the effec-tiveness of ibutilide in terminating AFL appearedto be definitely greater than the results yieldedby other class I and class III antiarrhythmic drugs,as proven by a number of direct comparisons[12,15]. This may be due to the lack of a reverseuse-dependent effect on refractory periods, whichis an effect that limits the efficacy of other antiar-rhythmic drugs [16,17].

In our study, the termination of AFL with therestoration of sinus rhythm was achieved in80% patients treated with ibutilide, a percentagesignificantly greater than the positive results ob-tained by ATP. In agreement with what has beenpreviously mentioned, this difference may be dueto the AFL’s time of onset, which in our studyreached up to one month. While the effectivenessof ibutilide remained high for any duration of AFLup to 30 days, with 83% responders in Group 1 suf-fering from AFL O48 h, the average duration ofAFL in ATP responders was 2.7 days, with 44% res-ponders suffering from AFL!48 h, and the remain-ing cases of AFL whose duration was only slightlylonger.

The success percentage obtained in our popula-tion exceeds the one reported by Stambler andcolleagues [11], who showed an efficacy of 63%in a group of patients with AFL the duration ofwhich ranged between 3 h and 45 days: in this case,it is possible that the study’s different inclusioncriteria, and the population’s different clinical fea-tures, such as the percentage of patients sufferingfrom valvular heart disease, with left atrial dilata-tion or reduced ejection fraction, may explain ourpopulation’s achievement of a better success per-centage. Besides this, Vos and colleagues [15]

0

10

20

30

40

50

60

70

80

90

100

Gr 1Gr 2

P=nsP<0,006

AFL < 48 hours AFL > 48 hours

Success rate according to time of onset of AFL

Figure 4 Comparison of patient responders of Group 1vs. Group 2, according to a cut-off point of 48 h.

reported an efficacy of ibutilide of 70% in terminat-ing AFL whose onset occurred within 45 days, a per-centage that rises to 76% in the study by Volgmanand colleagues [12], which concerned 17 patientssuffering from AFL with an average duration of 22days, treated with ibutilide up to 2 mg.

In concordance with the electrophysiologicalproperties of this drug, in our study, ibutilideshowed a slight effect on conduction velocity: infact, among the treated patients, the AFL’s cyclewas lengthened by an average of 18 ms, consider-ably less than the average increase showed by theQTc interval. This testifies the drug’s stronger elec-trophysiological effect on refractory periods thanon conduction velocity, as already fully proven byprevious studies, both experimental and performedon man [18e20].

In our study, the average lengthening of the QTcinterval in patients treated with ibutilide was46 ms, without any significant difference betweenresponder and non-responder patients. It is knownfrom the literature that the extent of the QTc inter-val lengthening after ibutilide does not anticipatethe drug’s effectiveness on AFL nor on atrial fibril-lation [10,11]. Despite the lengthening of the QTcinterval reflecting the drug’s effect on the ventric-ular refractory period, Buchanan and colleaguesshowed in experimental models that the increaseof the atrial refractoriness represents an importantaspect of the antiarrhythmic effectiveness of ibuti-lide [21,22]. We suggest that the width of the excit-able gap, variable from case to case, is able toexplain the lack of relation between the extent ofthe QTc interval increase and the termination ofAFL: slight gaps may be filled by rather modestQTc increases, while excitable gaps of considerablewidth may not be extinguished even by a moremarked lengthening of the atrial refractory period.

In our study, ibutilide failed to restore sinusrhythm in 9 patients; since our study was focusedon the direct comparison between two treatmentmodalities of AFL, we did not plan a cross-over de-sign, so that none of the non-responder patients inGroup 1 underwent ATP. In this respect, the useful-ness of ibutilide for enhancing pacing terminationof AFL has already been demonstrated by Stamblerand colleagues [20], who showed that ibutilide andprocainamide facilitate pacing induced termina-tion of AFL compared with placebo (88% vs. 18%),and has been recently described by Cheng and col-leagues [23], though in a small number of patients;therefore, this strategy might represent an effica-cious step-up protocol for sinus rhythm restorationin patients affected by AFL, consisting of ibutilideas a first-line treatment, followed by ATP if ineffec-tive. It is conceivable that this protocol would

306 A. Mazza et al.

by guest on June 1, 2013http://europace.oxfordjournals.org/

Dow

nloaded from

leave only a small proportion of patients as candi-dates for electrical cardioversion.

We did not encounter any case of polymorphicventricular tachycardia that required electricaldefibrillation, but the administration of ibutilidewas suspended in two cases due to the onset ofnon-sustained polymorphic ventricular tachycar-dia. However, it should be underlined that theproarrhythmic events reported in the literaturehave always occurred immediately or a few mi-nutes after the treatment had begun, in patientswith a base QTc interval O440 ms, or with a com-promised haemodynamic condition (advancedfunctional class, reduced ejection fraction) [10],which in our study represented an exclusion criteri-on. Therefore, an accurate selection of patientschosen for ibutilide treatment can minimize thedrug’s risk of eliciting this arrhythmia.

In conclusion, both ibutilide and ATP proved tobe effective and safe in terminating AFL with a veryrecent onset, but the use of ibutilide appears to bethe best choice in AFL of over 48 h duration.

References

[1] Chung DC, Kerr CR, Cooper J. Termination of spontaneousatrial flutter by transesophageal pacing. Pacing ClinElectrophysiol 1987;10:1147e53.

[2] Di Marco JP. Transesophageal pacing in atrial flutter. Chest1987;92:4e5.

[3] Disertori M, Inama G, Vergara G, Guarnerio M, Furlanello F.Impiego della stimolazione atriale transesofagea nel trat-tamento del flutter atriale. G Ital Cardiol 1984;14:153e7.

[4] Guarnerio M, Furlanello F, Del Greco M, Vergara G, Inama G,Disertori M. Transesophageal atrial pacing: a first-choicetechnique in atrial flutter therapy. Am Heart J 1989;117:1241e52.

[5] Volkmann H, Dannberg G, Heinke M, Kuhnert H. Termina-tion of tachycardias by transesophageal electrical pacing.Pacing Clin Electrophysiol 1992;15:1962e6.

[6] Disertori M. La stimolazione atriale transesofagea neltrattamento del flutter atriale: utilita e limiti. G Ital Cardiol1993;23:281e3.

[7] Lee KS, Gibson JK. Unique ionic mechanism of action ofibutilide on freshly isolated heart cells. Circulation 1995;92:2755e7.

[8] Lee KS, Tsai TD, Lee EW. Membrane activity of class IIIantiarrhythmic compounds: a comparison between

ibutilide, D-sotalol, E-4031, sematilide and dofetilide. EurJ Pharmacol 1993;234:43e53.

[9] Murray KT. Ibutilide. Circulation 1998;97:493e7.[10] Ellenbogen KA, Stambler BS, Wood MA, et al. Efficacy of

intravenous ibutilide for rapid termination of atrialfibrillation and atrial flutter: a doseeresponse study. J AmColl Cardiol 1996;28:130e6.

[11] Stambler BS, Wood MA, Ellenbogen KA, Perry KT,Wakefield LK, VanderLugt JT. Efficacy and safety of repeatedintravenous doses of ibutilide for rapid conversion of atrialflutter or fibrillation. Circulation 1996;94:1613e21.

[12] Volgman AS, Stambler BS, Kappagoda C, et al. Comparisonof intravenous ibutilide versus procainamide for the rapidtermination of atrial fibrillation or flutter. Pacing ClinElectrophysiol 1996;19:608.

[13] Naccarelli GV, Lee KS, Gibson JK, VanderLugt J. Electro-physiology and pharmacology of ibutilide. Am J Cardiol1996;78(Suppl 8A):12e6.

[14] Inoue H, Yamashita T, Nozaki A, Sugimoto T. Effects ofantiarrhythmic drugs on canine atrial flutter due to reentry:role of prolongation of refractory period and depression ofconduction to excitable gap. J Am Coll Cardiol 1991;18:1098e104.

[15] Vos MA, Golitsyn SR, Stangl K, et al. Superiority of ibu-tilide (a new class III agent) over DL-sotalol in convert-ing atrial flutter and atrial fibrillation. Heart 1998;79:568e75.

[16] Hondeghem LM, Snyders DJ. Class III antiarrhythmic agentshave a lot of potential but a long way to go. Reducedeffectiveness and dangers of reverse use dependence.Circulation 1990;81:686e90.

[17] Singh BN, Ahmed R. Class III antiarrhythmic drugs. Curr OpinCardiol 1994;9:12e22.

[18] Cimini MG, Brunden MN, Gibson JK. Effects of ibutilidefumarate, a novel antiarrhythmic agent, and its enan-tiomers on isolated rabbit myocardium. Eur J Pharmacol1992;222:93e8.

[19] Buchanan LV, Kabell G, Gibson JK. Acute intravenousconversion of canine atrial flutter: comparison of anti-arrhythmic agents. J Cardiovasc Pharmacol 1995;25:539e44.

[20] Stambler BS, Wood MA, Ellenbogen KA. Comparativeefficacy of intravenous ibutilide versus procainamide forenhancing termination of atrial flutter by atrial overdrivepacing. Am J Cardiol 1996;77:960e6.

[21] Buchanan LV, Lemay RJ, Walters RR, et al. Antiarrhythmicand electrophysiologic effects of intravenous ibutilide andsotalol in the canine sterile pericarditis model. J CardiovascElectrophys 1996;7:113e9.

[22] Buchanan LV, Turcotte UM, Kabell GG, Gibson JK. Antiar-rhythmic and electrophysiologic effects of ibutilide ina chronic canine model of atrial flutter. J CardiovascPharmacol 1993;22:10e4.

[23] Cheng J, Glatter K, Yang Y, et al. Electrophysiologicalresponse of the right atrium to ibutilide during typical atrialflutter. Circulation 2002;106:814e9.