Nonpharmacologic strategies fer treating atrial fibrillation

Nonpharmacologic Atrial

Strategies Fibrillation

for Treating

Berndt Lijderitz, MD, Dietrich Pfeiffer, MD, Jiirgen Tebbeniohanns,

MD, and Werner Jung, MD

Nonpharmacologic toots to treat atrial fibrillation (AF) are direct current cardioversion, radiofrequency (RF) current catheter ablation, antiarrhythmic surgery, pacing, and atrial defibrillation. In patients with sustained AF, when no cause can be found for AF or when the associated disease is mild, an attempt should be made to restore sinus rhythm. Electrical cardioversion by synchronized direct current shock can be attempted when drugs have failed and is the first choice in acutely ill patients. Virtually all patients should be anticoagulated. Tempomry pacing shoukl be available in patients with evidence of previous bmdycardia. Although efftcacy may be improved in patients pretreated with antiarrhythmic drugs, there is a considemble risk of adverse events. In AF and sinus node dysfunction, both pacing and antiarrhythmic drugs may be necessary. Pacing shouM be atrial or dual chamber, since ventricular pacing provokes AF. Failure to control the ventricular rate in AF can be treated by RF: atrioventricular (AV) node ablation, ablation of accessory pathways in preexcitation syndrome with AF, modulation of AV node, or ablation of AF. Antiarrhythmic surgery is a major procedure and may be the thempy of last resort in AF: the so-called

corridor procedure isolates the fibrillating atria from a strip of tissue connecting the sinus and AV nodes. The maze procedure attempts to abolish AF by channeling the atrial activation between a series of incisions. In patients with chronic AF, internal cardioversion should be attempted if conventional transthomcic electrical cardioversion is ineffective. Seveml studies demonstrated the feasibility and efficacy of internal atrial defibrillation in selected patients with recent onset, as well as with chronic, AF. An implantable atrial defibrillator--crs a stand-alone device or as part of a whole heart cardiovertermight be an option in the future. Nonpharmacologic tools play only a minor role in the management of paroxysmal and chronic AF. If symptoms persist despite pharmacologic therapy and other causes of persisting symptoms are excluded, consideration should be given to cardiac pacing, RF catheter treatment, or surgery. In some cases nonpharmacologic thempy of the AV node must be followed by implantation of a permanent pacemaker (due to complete AV block) and anticoagulation (due to persistence of underlying AF). (Am J Cardioll996; 77:45A-52A)

A

important than the classic causes of AF-rheumatic heart disease and thyrotoxicosis-which are declining in incidence. In about 15% of patients with chronic AF, no underlying cardiac or metabolic abnormality can be found. The arrhythmia can itself give rise to atria1 dilation. AF consists most probably of several coexisting reentrant wave fronts of activation within the atria. Atria1 activation in AF is characterized by multiple wavelets sweeping around the atria in irregular, shifting patterns; completed reentrant circuits are the exception.* The primary therapeutic goal in patients with chronic AF is the restoration and maintenance of sinus rhythm. Therapy of paroxysmal AF is based on the need to control ventricular rate during recurrences, i.e., paroxysms of AF are usually managed conservatively. The indication for attempting pharmacologic or nonpharmacologic therapy is the possibility of hemodynamic benefit by limiting the number of recurrences and controlling of the ventricular rate in case of recurrence. Nonpharmacologic tools in AF are direct current cardioversion, cardiac pacing, radiofrequency (RF) catheter

trial fibrillation (AF) is a common arrhythmia affecting 2-4% of the population > 60 years of age. It can be found in up to 10% of those >65 years old. AF may cause disabling symptoms and serious adverse effects, such as impairment of cardiac function or thromboembolic events. It is also associated with an increased risk of death. In the past the most common underlying heart disease related to chronic AF was rheumatic heart disease. Nowadays this disease occurs relatively rarely. Nevertheless, the incidence of AF is likely to increase in the future due to the aging of the population, since its prevalence increases with age. In most patients with chronic AF the arrhythmia can be attributed to organic heart disease or metabolic disorders. In Western countries ischemit and hypertensive heart disease (including sick sinus syndrome) and alcohol are numerically more

From the Department of Cardiology, University of Bonn, Bonn, Germany. Address for reprints: Berndt Ltideritz, MD, Department of Medicine and Cardiology, University of Bonn, 25, Sigmund-Freud-W., D-531 05 Bonn, Germany.

A SYMPOSIUM:

ATRIAL

FIBRILLATION

UA

TABLE I Nonpharmacologic

ablation in 50 patients with a variety of problems, including coronary artery disease, 10 (20%); mitral valve disease,4 (8%); bradycardia-tachycardia syndrome, 5 (10%); dilated cardiomyopathy, 6 (12%); diphtheria, 1 (2%); chronic obstructive lung disease, 1 (2%); and pericarditis/myocarditis, 1 (2%). The majority of patients did not show any structural heart disease (Figure 1). The duration of symptomswas 7.6 2 6.9 years at a maximal rate of 182 f 19 beats/min. Paroxysmal AF was diagnosed in 36 (72%) patients with a mean incidence of 126 + 149 attacks per year. The mean duration of AF was 12.9 f 9 hours. Permanent AF was present in 14 (28%) patients, and 4.9 + 1.1 antiarrhythmic drugs were taken until a patient was classified as drug resistant. Antibradycardia pacemakers were implanted in 10 patients (for different reasons) before RF ablation was performed. The frequency of ablation procedures was 1.1 + 5 per year, the duration of the procedure took 92 * 50 minutes, the fluoroscopy time was 20.0 f 20.4 minutes, the frequency of RF current deliveries was 9.8 2 9 per session at a cumulative energy of 20.2 f 20.9 kW (Figure 2).

Tools in Atrial Fibrillation

Direct current cordioversion/defibrillation Atrioventriculor node ablation or modification Electrical catheter LCISW

Cryogenics Surgery Alcohol or phenol oblotlon Pacemaker (pause dependent atrtol fibrillation) Antiorrhythmic surgery Corridor procedure Mare procedure Atrial (internal) defibrillation lmplontoble atrial defibrillator

TABLE II Radiofrequency Modification and Ablation of Atrioventricular Node in Atrial Fibrillation: The Bonn Experience

knder Mean age (years) Histoty Paroxysms since (years) Max. rate (bpm) Stokes-Adams attacks Afrhy(trmio Paroxysmal AF Chronic AF Additional flutter Disease Unclear Hypertensive HD Coronary HD Dilative CM Brady-tachycordio syndrome Mitral valve disease Postdiphtheric HD Chronic obstructive LD Chronic myocorditis Atriol septol defect NYHA clorsificotion Former fmotmenl A4 drugs Pacemaker RFC abktion pommeter Duration of the procedure (min) Radiation exposure (min) Number of RFC deliveries Follow-up Permanent effect Insufficient 2nd treatment

Modification

Ablation

10 tan, 41 62 f 11 (42-71)

50 32m, 1af 58 e 14 (14-81)

8.7 k 9.2 197 t 30 5 (50%)

7.6 2 6.9 la2 t 19 13 (26%)

6 (60%) 4 (40%) 2 (20%)

36 (72%) 14 (28%) 9 (18%)

3 2 2 2

21 (42%) -

(30%) (20%) (20%) (20%)

Modulation of the atrioventricular node in atrial In patients with AF with an unconfibrillation:

2.4 + 0.7

10 6 5 4 1 1 1 1 2.1

4.9 + 1.9 (2-7) 1 (10%)

4.9 -c 1.1 (O-7) 10 (20%)

123 + 41.1 24.7 f 16.9 9.9 + a.3

92 2 50 20.0 2 20.4 9.8 k 9.0

9 (90%) 1 (10%) AVN block Ill0

49 (98%) 1 (2%) Amiodorone

1 (10%)

-

AF = ntrial fibtillatiin; CM = cardiomyopafhy; HD = heart dimsee; RFC = rodlo trequency cwrm!.

(20%) (12%) (10%) (a%) (2%)

trolled ventricular rate refractory to drug therapy, RF energy has been widely used to interrupt atrioventricular (AV) conduction by ablation of the AV node.9J0However, permanent pacemaker therapy is required in most patients. For this reason, it would be desirable to modify the AV node by RF current without creating 3rd degree AV block and to slow the ventricular rate during AF. In an earlier study, modulation instead of ablation was feasible in only 35% of patients with atria1 tachyarrhythmias targeting the anterior region of the interatrial septum, close to the His bundle recording site.” A high rate of immediate and late 3rd degree AV block and a low efficacy rate were found. Recently, some studies have shown that the ventricular rate during spontaneous12-l4 or induced15J6AF was effectively reduced after targeting the posterior aspect of the right septal atrium where the slow pathway is ablated in patients with AV node reentrant tachycardia. Williamson et all2 studied 19 patients with AF and uncontrolled ventricular rates. After a total of 11 ? 5 RF energy applications to the posterior or midseptal right atrium, successful control of the ventricular response was achieved in 74% of patients. Permanent pacemaker implant due to inadvertent AV block was necessary in 4 (21%) pa-

(2%) (2%)

(2%) 2 1.0

disease; LD = lung

treatment, antiarrhythmic surgery, and internal atria1 defibrillation (Table I).2-s The selection of antiarrhythmic therapy in AP depends entirely on the clinical context and the underlying heart disease. RADIOFREQUENCY TREATMENT Ablation fibrillation: 46A

CURRENT CATHETER

of the atrioventricular

node in atrial

Our experience (Table II) includes RF

THE AMERICAN

JOURNAL

OF CARDIOLOGY

VOLUME

77

JANUARY

25, 1996

tients. The maximal ventricular rate during exercise 25 was significantly reduced from 180 ? 39 to 101 & 18 beats/min at 2 days (p
1

FIGURE 3. Radiofrequency (RF) modification of the atrioventricular (AV) node in a patient with tachyarthythmic ventricular response in atrial fibrillation before (upper tracing) and after RF current delivery (lower tmcing). I, Ill, aVF: surface electrocardiographic leads; Abl. = ablation (RFC beam); RV = right ventricle lead; HBE = His bundle electrocardiogram. Posterior approach.

A SYMPOSIUM: ATRIAL FIBRILLATION

47A

ventricular rate during induced AF after ablation of the slow pathway. l6 In a short-term follow-up electrophysiologic control study these effects persisted for 1 week. In 10 patients with uncontrolled ventricular rates during AF but without AV node reentrant tachycardia, a successful modification procedure was achieved in 9 (90%) patients (Figure 3; Table II). No inadvertent block occurred. In 1 patient AV junction ablation was performed because rate control was not achieved. In summary, RF current modulation of the AV node, by targeting the posterior aspect of the atria1 septum, is feasible in the majority of patients with AF and rapid ventricular response. Since the incidence of inadvertent AV block may be up to 20%, this procedure should be limited to those patients who are also candidates for AV node ablation and have agreed to pacemaker implantation. Ablation of atrial fibrillation: In contrast to AV node ablation or modification, which are palliative procedures (since AF persists), Haissaguerre and coworkers” recently described a case in whom successful RF catheter ablation of AF was achieved. A specially designed 14-polar catheter was used to create linear lesions in the right atrium. At first, RF energy was applied at the posterior right atrial wall sequentially through all electrodes from the anterior high atrium to the orifice of the inferior vena cava. The second RF application series targeted a line from the anterolateral mid-right atrium to the posteroseptal tricuspid ring in a nearly horizontal plane. The third position was on the anterior right atrium by connecting the anterior ends of the first 2 lesions and the posterior end of the second lesion. During the final 30 RF applications (lo-40 W), AF terminated. AF remained uninducible. The patient has been free of arrhythmias without antiarrhythmic medication for up to 3 months. The mechanism of AF in this special patient was thought to be more amenable to RF treatment, since AF was paroxysmal and showed dominant organized activity of the atria and fibrillating activity was absent in the left atrium. RF application has been used in animal models to mimic surgical procedures such as the maze procedure.‘8-2n The results with specially designed catheters with multiple electrodes to create linear lesions arc encouraging. Swartz and coworkers” also demonstrated that a catheter-based curative approach to AF is possible using 7 different long vascular sheets as introducers for a standard RF ablation catheter. These sheets fit into anatomically predesigned 48A

THE AMERICAN JOURNAL OF CARDIOLOGY

VOLUME 77

curves within the right and left atria and along the interatrial septum. The maze procedure was replicated by 8 linear lines, including a transseptal approach for the left atrium cuts and the transseptal lesion. Swartz et al presented data on 7 patients. An acute success with restoration of sinus rhythm was achieved in 6 patients, and 2 patients required a second procedure at follow-up. A hemodynamic improvement was found in 6 patients. All these encouraging preliminary results indicate the need to assesspatient safety and long-term success of RF catheter techniques in AF.

USE OF PACEMAKERS IN ATRIAL FIBRILLATION Cardiac pacing is appropriate to treat slow ventricular rates in AF, such as the use of temporary pacing in acute situations and in such specific indications as drug-induced pauses by digitalis and other antiarrhythmic drugs, in tachycardia-bradycardia syndrome, and after catheter ablation. An unusual indication is deliberate provocation of AF. In patients with symptomatic atria1 tachycardias, high rate atrial pacing can induce chronic selfsustaining AF, permitting effective rate control with digoxin and hemodynamic improvement. This unusual indication for permanent pacing may be considered as an alternative for a small wellselected group of patients in whom AV junction ablation and other antitachycardia measures are either unavailable or not feasible.3

SURGERY FOR ATRIAL FIBRILLATION For patients with AF who cannot be managed effectively with drugs, or who have had cardiogenic embolism as a result of the arrhythmia, surgery is a therapeutic alternative for restoring sinus rhythm and AV synchrony, potentially limiting the risk of stroke.22 The so-called cotidor procedure separates the fibrillating atria from a strip of tissue connecting the sinus and the AV node.23 Because both right and left atria continue to fibrillate, the hemodynamic abnormalities associated with AF are not improved. In addition, the vulnerability to the development of left atria1 thrombi is not alleviated. Originally, it was hypothesized that this small corridor of atria1 tissue would not be large enough to sustain AF. However, AF may continue and anticoagulation remains necessary. The more ambitious maze procedure for antiarrhythmic surgery attempts to abolish AF by channeling the atria1 activation between a series of incisions. The maze operation is designed to restore atria1 mechanical function and to obviate the JANUARY 25, 1996

need for anticoagulation. Both atria1 appendages are excised and the pulmonary veins arc isolated. Appropriately placed atria1 incisions not only (1) interrupt the conduction routes of the most common reentrant circuits, they also direct (2) the sinus impulse from the sinoatrial node to the AV node along a specified route and (3) preserve synchronous atrial electrical activation as a prerequisite for contraction. The entire atria1 myocardium, except for the atrial appendages and pulmonary veins, is electrically activated, thereby preserving atria1 transport function postoperativcly.24 Recently, 75 patients were reported who underwent the maze procedure. Initial success and low perioperativc mortality are remarkable.‘2 Up to November 1993, 102 maze procedures were performed at Barnes Hospital, St. Louis, MO. Encouraging results have been reported by Chang,25 who used the so-called modified atrial transsection procedure in 17 patients for surgical management of AF.

INTERNAL CARDIOVERSION ATRIAL FIBRILLATION

FOR

One of the options for conversion of AF to sinus rhythm has been cxtcrnal cardioversion using energies in the range of 50-350 J. External electrical cardioversion/defibrillation has been a remarkably effective and safe method for termination of this arrhythmia. Originally introduced by Lown and coworkcrs2” in 1962, it has been a well-accepted mode of acute therapy. However, this technique requires general anesthesia or heavy sedation. In addition, there is a potential risk of myocardial damage, ventricular tachyarrhythmias, or thromboembolism. Further, cxtcrnal cardiovcrsion/defibrillation must be undertaken in the hospital environment. Internal atria1 defibrillation has been evaluated as an alternative approach to the cxtcrnal technique for over 2 decades. Animal studies: Previous animal work has demonstrated the feasibility of low-energy transcathctcr countershock of AF. Mower et a12’ using 2 catheters, 1 in the right atrium and 1 in the superior vcna cava, have shown successful defibrillation with an energy range of 0.05-3 J in acetylcholine-induced AF in dogs. Dunbar et a12swere able to tcrminatc only 26% of AF episodes induced by talc pcricarditis in dogs. Subsequent investigations from the same group did not demonstrate increasing efficacy with sequential shocks compared with single monophasic shocks utilizing a 3-electrode lead configuration. 29 In contrast, Kumagai and coworkers”” had an efficacy rate of 47% at energies

of <0.5 J, 74% at 1 J, and 100% at <5 J in the same model. Powell et a13’reported a lower sucess rate with biphasic shocks and a right atrium-apical left thoracic patch configuration in a large sheep model. During 768 defibrillation attempts in 16 sheep, the percentage of successful cardioversion attempts increased in a dose-response manner, reaching a plateau at the avcragc energy level of 5 J. Recently, Cooper et a132studied several different lead systems using single capacitor monophasic and biphasic shocks in the same model. They found that the optimal lead systems for internal cardioversion of AF were those that had electrodes that encompass as much of the fibrillating atria1 tissue as possible and that did not create high potential gradients near the sinus or AV nodes. In this study, the right-to-left lead system using the distal coronary sinus as the left electrode and a 3/3 msec biphasic waveform resulted in low energy requirements for cardiovcrsion of AF in sheep. Studies comparing shock electrode lengths demonstrated that 6-cm electrodes located in the coronary sinus and right atrium exhibited a trend towards lower defibrillation thresholds than did 3-or 9-cm lengths in the sheep model,‘” whereas in the canine model 6-cm clcctrodes located in the right atria1 appendage and coronary sinus produced significantly lower thresholds than did 3-cm electrodes.34 Human studies: Previous studies in humans have demonstrated that high energy (200-360 J) transcatheter atria1 defibrillation is safe and effective when using standard electrophysiology catheters.35,3” A recent randomized study demonstrated that internal cardioversion using highenergy shocks (200-300 J) was more effective than external cardioversion (300-360 J) in restoring sinus rhythm and was as safe as external cardioversion in patients with chronic AF.4 However, reports on low-energy endocardial defibrillation in humans are limited. An early feasibility study of low-energy cardioversion for atria1 arrhythmias did not yield successful results in patients with AF.” More recently, preliminary studies demonstrated the feasibility of low-energy cardioversion in selected patients with recent onset as well as with chronic AF. Keane and coworker@ reported that chronic atria1 arrhythmias could be cardioverted efficiently in 15 of 16 patients with a mean atria1 defibrillation threshold of 6.7 -t 2.2 J. Johnson et a139compared in 6 patients a 6-mscc monophasic with a 3/3-msec biphasic truncated exponential waveform. The biphasic waveform required less total delivered cnergy (mean, 2.5 & 1.4 J) than the monophasic (4.7 + 3.1 J) for successful atrial defibrillation. A SYMPOSIUM: ATRIAL Fll3RltlATlON

49A

Murgatroyd and coworkers40 attempted cardioversion in 8 patients with AF of short duration, using a coronary sinus and right atria1 electrode system and a 3/3-msec biphasic waveform. In 7 of 8 patients they were successfulin terminating the atrial arrhythmia with a mean energy of 2.2 & 0.8 J. In 1 patient, shocksof up to 4.5 J were ineffective. In a recent study successfulinternal electrical defibrillation was achieved in 10 of 14 patients with a mean duration of AF of 5.7 ? 5.4 months at a mean energy of 3.7 +- 1.7J using a right to left electrode configuration.41 In contrast, Kalman and coworkers42 used either a right sided electrode configuration or a 3-lead systemfor endocardial defibrillation in 5 patients with atrial flutter and in 4 patients with AF with a mean duration of 3.75 months. Successfulcardioversion was accomplished in all 5 patients with atria1 flutter with energies of < 10 J but in only 1 patient with AF at the 10 J level. This low success rate was probably due to the less optimal lead configuration used in this study. In our department at the University of Bonn we assessed the feasibility of low-energy endocardial defibrillation in patients with atria1 tachyarrhythmias who had failed pharmacologic conversion as well as transthoracic defibrillation using energies up to 360 J. In those patients, low-energy endocardial defibrillation was attempted following sedation with diazepam and pethidine. Internal atria1 defibrillation was performed using 2 catheters, 1 located in the right atria1 appendage and 1 in the coronary sinus. A third standard catheter was positioned in the apex of the right ventricle to synchronize shocks with the R wave (Figure 4). For atria1 defibrillation measurements, we used a

FIGURE 4. left anterior oblique view (60”) of 2 coil electrodes in the right atrial appendage and in the coronary sinus. A third electrode catheter for synchronizing with the R wave is positioned in the apex of the right ventricle.

50A

THE AMERICAN JOURNAL OF CARDIOLOGY

VOLUME 77

step-up protocol, starting with 0.5 J and incrementing by 0.5 J. The mean atria1 defibrillation threshold in our patient population was 1.6 + 0.9 J (range, 0.5-3.0 J). No complications occurred; in particular, no proarrhythmic events were observed.

IMPLANTABLE

ATRIA1 DEFIBRILLATOR

The development of an implantable cardioverter-defibrillator for the management of ventricular tachyarrhythmias has stimulated investigation of a similar approach to AF. Conversion of atria1 arrhythmias has been attempted in patients with an implantable cardioverter-defibrillator using epicardial or nonthoracotomy lead configurations. Highenergy shocks applied via epicardial patches during AF were in no case successful in restoring sinus rhythm.43 In a prospective and randomized study, Saksena et a144reported on the clinical efficacy and safety of atria1 defibrillation using 3 current nonthoracotomy endocardial lead configurations: 2 rightsided vectors, RV-RA and RV-SVC leads, and 1 right to left vector, RA-left thoracic patch. Atria1 defibrillation thresholds in 21 patients with cardiac disease were lowest for the RV-SVC configuration. All patients could be cardioverted by 20 J in this configuration but only 60% were reverted by right to left shock vector.45 Recently, Poole et a146reported on the feasibility of atria1 defibrillation in 10 patients using a unipolar, single lead right ventricular pectoral defibrillation system. The atria1 defibrillation threshold was 8.3 -t 4.1 J using the active can system. AF remains a common postshock arrhythmia with implantable ventricular defibrillators and may result in inappropriate ventricular shock delivery.47a48 Adding an atria1 cardioversion system to the ventricular system would allow for better arrhythmia discrimination as well as provide more complete arrhythmia treatment coverage. Patients with symptomatic recurrences of AF despite the use of antiarrhythmic drug therapy represent potential candidates for an implantable atrial defibrillator.5,4950The number and duration of AI? episodes should be taken into account in the indications. Patients with frequent episodes must be excluded as candidates for implantation of an atrial defibrillator because of too frequent discharges, patient discomfort, and rapid battery depletion. Similarly, patients with episodes of short duration and spontaneous termination may not be good candidates. Rather it is selected patients with infrequent, symptomatic attacks of long-lasting episodes of AF despite antiarrhythmic drug therapy who may benefit from an implantable atria1 defibrillator. JANUARY 25, 1996

Automatic

diagnosis

of atrial arrhythmias:

An

atria1 defibrillator should reliably recognize AF. Atria1 electrograms are more difficult to detect than ventricular electrograms because they have lower amplitude and slower rate and therefore are easily confused with far-field ventricular events. However, recent studies have shown that AF can be recognized through atria1 electrograms using atria1 rate, amplitude, probability density function, and spectral analysis, with a sensitivity > 88% and a specificity of 1OO%.49 Pain perception: Early reports of the effect of low-energy cardioversion for the treatment of atria1 arrhythmias stressed that in some patients intolerable pain could result from discharges of < 1 J.30,37 In contrast, Murgatroyd et a140noted that shocks of < 1 J were well tolerated, but shocks above this level require sedation. Saksena et a145undertook a systematic prospective study of atria1 defibrillation using a patient pain perception questionnaire. At 1 J, 20% of patients reported pain; at 2 J this increased to 40%, with the majority experiencing pain by 3 J. These data define the limitations of internal atria1 defibrillation in a conscious patient. Finally, pain perception may have a major impact on quality of life in patients with implantable defibrillator devices.51a52 Safety: A major concern with an implantable atria1 defibrillator is the potential risk of inducing ventricular tachycardia or fibrillation when delivering a shock to convert AF. Previous animal studies have shown an incidence of induced ventricular fibrillation in the range of 2-6.5% following shocks with an energy level of 0.1-5 J.28 Powell et aP1 reported on the occurrence of ventricular fibrillation in 18 (2.4%) of 768 cardioversion attempts in 16 anesthetized sheep. In all 18 cases, the shock was poorly synchronized with the R wave. Ayers et aP3 evaluated conditions under which ventricular fibrillation might be induced during synchronized electrical conversion of AF in a sheep model. They found that synchronized transvenous atria1 defibrillation shocks delivered on beats with a short preceding ventricular cycle length (< 300 msec) were associated with a significantly increased risk of initiation of ventricular fibrillation. Induction of ventricular tachyarrhythmias has not been observed in any of the recent reports in humans if the shocks were properly synchronized to the R wave. The occurrence of bradyarrhythmias postshock has also been demonstrated. Cooper et aP2have shown that lead systems that generated high potential gradients near the sinoatrial and AV node areas resulted in more frequent episodes of postshock

conduction disturbances. In their human study Saksena and Prakash45 observed a 28% incidence of sinoatrial and AV conduction delays. These data suggest the need for ventricular pacing support during internal atria1 defibrillation. At present, the long-term use of an implantable atria1 defibrillator is a major challenge in the nonpharmacologic treatment of AF. Studies in animals as well as in humans have demonstrated the feasibility of internal atria1 defibrillation. The major issues that still have to be addressed are the pain perception and the potential risk of inducing life-threatening ventricular arrhythmias during delivery of low-energy atria1 shocks. A first step to this novel approach could be a physician-activated device. In the beginning, the implantable atria1 defibrillator should be restricted to highly selected patients with drug refractory, poorly tolerated, recurrent AF episodes. The extension of this therapy to wider subsets of patients should be dependent on the initial results with regard to clinical efficacy and safety as well as patient tolerance. Finally, cost-effectiveness studies are needed to demonstrate the benefit of this specific therapy among other therapeutic strategies available for the management of AF.

CONCLUSIONS For management of AF persisting despite pharmacologic therapy, underlying causes of the tachyarrhythmia such as congestive heart failure or thyroid disease, should be excluded. If there are no symptoms, serial echocardiographic studies should be performed. When left ventricular function is stable, further observation is adequate. If the symptoms are attributable to fibrillation, ablation or surgery should be considered. The same is true for worsening of left ventricular function in the follow-up period. Nonpharmacologic tools play only a minor role in the management of paroxysmal and chronic AF. If symptoms persist despite pharmacologic therapy and other causes of persisting symptoms are excluded, consideration should be given to cardiac pacing, RF current catheter therapy, or surgery. In some cases nonpharmacologic therapy of the AV node must be followed by implantation of a permanent pacemaker (due to complete AV block) and anticoagulation (due to persistence of underlying AF). 1. Murgatroyd FD, Camm AJ. Atrial arrhythmias.Lancet 1993;341:1317-1322. 2. L.own B, Perlroth MG, Kaidbey S. “Cardioversion” of atria1 fibrillation. N Engl J Med 1963;269:325. 3. Polk& A, Fak RH. The use of pacemakersin atria1 fibrillation. In: Falk RH, Podrid PJ, eds. Atrial Fibrillation: Mechanismsand Management. New York: Raven Press1992345-358.

A SYMPOSIUM: ATRIAL FIBRILLATION

51A

4. L&y

S, Law&

Moyal

P, Dolla

C, Brcmondy

F. A randomized atrial

comparison

Abrillation.

5. L&y

E, Kou

M, Schork

KadLsh

A, Calkins

H, Paganelli

Y, Das S, Shea IM, Gupta

of cncmal

Circularivn

S, Camm

W,

A, Shyr

and internal

F,

cardiovcrsion

of chronic

1992;X6:1415-1420.

J. An

implantable

atria1

dcIibrillator.

Cinx/afion

1993;X7:

6. Ferguwn

TB Jr, Cux JL. Surgkxl JW,

Guiraudon

GM,

symptomatic

atrial

Mechanisms

and Management.

8. Liideritz

B, Pfeitler

11th

International

Italy.

lYx43o~3oY.

9. Jackman

fibrillation.

PD.

D.

Wang

Lanara

R. Catheter in

for atrial

In: Falk

RH,

Yee

librillation.

Hen

R. Surgical

Podrid

York:

Raven

The

New

KH,

therapy

of

of atrial

DM,

Roman

Naccarelli

GV, and

Marillcva, C, Moulton JV,

using large

K,

Dyer

J,

radiofrequency

tip catheter

clcc-

J, Lonergan

1 Mohama

node

ablation

R. Itather

with

R, Kerr

radiofrequency

11. Duckeck

W, Engelstein

tachyarrhythmias:

ED,

Kuck

modulation

KH.

C.

enew.

versus

Radiofrcquency

current

therapy

ablation

nodal

conduction.

of AV

Morady

BD,

Ching

Man

K, Daoud

P. Radiofrequency

catheter

the ventricular

rate during

to control

E, Niebauer

modilication atnal

M,

Strickterger

of atrioventricular

fibrillation.

Med

GK,

Control

Fleck

of rapid

RI’.

ventricular

of the atriovcntricular Ci/&&~n

14. Della uicular

0.

response

node

Prothro

DI.,

Ijahnson

1994331:

by radiofrequency

in patients

with

TD,

catheter

medically

lbarra

atrial

I’, Carbuciahio

conduction with

C, Tondo

by ablalion

C, Riva

of the %w”

drug-refractory

atrial

S. Modulation

atticntricular

Iibrilla-

fihnllation

or

of atrioven-

node

flutter.

pathway

in

Am CoU Cbrdiol

I

lW5;25:3W6. 15. Blanck Jazayeri

Z,

Dhala

MR.

during

results.

CirrrtDfi0n M.

S, Dhala

19. KernpIer and chronic 20. Elvan ot the

catheter

effects.

P.4CE HP.

reduces

approach JM,

ME,

Corr

JW,

Klein

Schua\ler

PB, Boineau surg

BC. Surgical I?

G, Yce

durahon

Wu

atrium:

RB,

catheter

of atrial

J

T,

Adler

SW.

and monophasic

Hiroki

pulse

1‘. Arakawa

of atrial

Fallon

fibrillation

Alfcmess

WA,

CA,

K.

IDW

fIutter/IibriIIation

JT, Krishnan

m the

Smith

in sheep.

Schoenlein

in

sheep.

SC, Ruskin

I Am

Co/l

JN.

Cordial

J, Niwano

S, Abe

atrial

defibrillation Cardiol

K, Yamanouchi on chronic

S, Lacombe

sion of chronic

Bexton

DO.

Catheter

Waldo

AL.

Transve-

tibrillation.

(Abstr.)

J

flutter

Yargcr

RA,

energy

tranxatheter

Camm

AI.

of cardiac

R, Jackson

MD,

energy

of transcltheter

Cdl Cardivl 1988:12:514

RS, Spurrell

G. Sowton

cardiover-

-518.

Internal

tramvenous

arrhythmi~w.

Hean J

5r

E. lindocardial

cardiover-

PACE lY93;16:313.

and fibrillation.

EL km

K. Effect

PACE 1991;14:1571-1575.

for the treatment

Wharton

internal

JM.

Monophasic

cardioverskm

and

of atrial

biphasic

fibrillation

wave-

in humans.

1993;88:31114. FD, Sladc low-energy

humans.

JAm Cdl Cur&l

(Abstr.)

41. Alt Initial

Ii, Schmitt

with

energy

R, Coenen

Rowland

M,

atrial

E, Dcolan

Fotuhl

E. Ward

DE, Camm

atrial

librillation

P, Karch

defibrillation

I+ Oliver

cardioversion

JN.

Implantahle

tion

to sinus rhythm.

TG,

of atrial

M,

in patients

O’Nunain

in

BIasini

with

R.

chrome

P, Mehra

L,

AM.

Low

PACE 1995;

H, McGovern

BA,

Ruskin

do not convert

atrial

fibrilla-

Varanavi

S, Kolettis

and safety

of atria1

lead configurations:

A. Atrial

B. Liiderio, Futura,

Mehra

defibrillation:

eds. Textbook

T,

Mahtcw

defibrillation

a prospective

P, using

random

fcchniqucs

and clinical

of Interventivnal

results.

i%c?rophysiology.

1995, in press.

Kudenchuk

single

R,

etlicxy

endomrdial

S, P&ash

JE,

shocks

Tonkin

J Am C&1 Cardiol 1994;125A.

(Abstr.)

NY:

JM,

in humans.

SS, Garan

Krol

R. Clinical

nonthorawtomy

Power

JAm Cdl CurdiollSY3;21~3A.

(Abstr.)

S, Mongeon

Lt-

arrhythmias

cardiwxter-defibrillator

44. Saksena

G,

DM,

of short-duration

l994;23:126A.

intracardiac

JM, Jones

endwardial

Annonk,

O’Farrell

cardiovenion

PACE 19%14:17:IOfi7-1078.

fibrillation.

42. KaInxan

AKB,

C, Ammer

experience

unipolar

Wagner

GM,

T, Ardkwa

tihrillation.

/Am

AJ. Tnnsvcnous

Jonson

cardiier-

PACE 1993;16%53.

of atrial

R, Bru P. High

fibrillation.

40. Murgatroyd

atrial

GM,

X, Ayeres mod&

Y, Hiroki

P, Cniite

atrial WW.

C~&tion

W, Aycrr;

II, Gonzalez

lone atrial

cardiovcrsion

for

Internal

1994,23:125A.

35. Kumagai

forms

RE.

atria1 deIibrillation.

in two canine

cardiwersion

37. Nathan

Ideker

1993;87:167.>1686.

WF!, Janas

34. Ortiz nous

WM.

Cinulrtirn

for tramvenous

G,

fibrillation

ablation

47. Jung

in dogs.

tachyarrhythmias

R,

lead

PJ, Dolack Mongcon

right

I,

Jones

GK,

L, Bardy

GH.

AtriaI

ventricular

to pectoral

1~991;84(.suppl):lI-612.

&x/u&m

l’?14$O(supp4):1-335.

chuk

Kater

KM,

Iappas

DG,

Cox JL

node-atrioventricular

operation

treatment of atrial

of syynchronised

for atrial

CM,

of attial

fibrillation. J. NLW

capacitor

PJ, Polk

following

fibrilla-

Chang

B-C,

Cain

J 7lwruc

fibrillation.

PACE 1993;16:1540. method

discharge.

for

terminating

JAM.4

lYQl&Q

JE,

Glcva

MJ,

Raitt

defibrillation

can system

GL

GH.

with

1%$3:411

Ltideritz

C, Buono

Bardy

defibrillation

Elecbophysiol

M,

MH, using

a

J Am Co//

(abstr).

13. lncidenut

of implantahle

G, Troutmann

Dolack

ventricular

cardiobwsc

G. Sinus

Neuman

B, Maw.

shock delivery

Johnson

50. Touboul I’. Atrial P./ICE: 1995:18:616-621. 51. Ltideritz

B, Jung

implantable

G, Ganman

Incidence

transvenous

W. Deister

DM,

Kuden-

of atria1

fibrillation

systems

in man.

lead

J

417.

detibnIlator:

cardioverter

of atrial

cardioverter/detibriUa-

49. L@ S, Richard P. IF there any indication for an intracardiac the lrratment of atria1 fibrillation? J Curdiowsc E&cm&yti~l

of the

management

R, Hugl

following

tars. Circu~tion

HJ Jr, Stone

JP. The surgical

Mlrtzko

48. Jonc.s GK,

the “corridor”

D’Agostino

W,

D. A catheter-based

IIU 19!?3$%.?d.

with

R,

configurations

BA,

of atrial

evaluation

acute

lYY1:17:97~975. RB,

whyarrhythmias

Noritami

H, McGovern

L, Cervantez

R. Guiraudon

results

is it needed?

Would

A, Mameros

A, Manz

defibrillator

in ventricular

defibrillator

for

1YYkS:982-9X5.

society

M. Patient

pay

for

it’!

acceptance

tachyarrhythmias.

PACE 1993;1(,:1815-1821. 52. Lilderitz

B, Jung

ahle cardioverter

W, Deister

defibrillator

A, Maw

M.

devices:

changing

Ilina

M, Wagner

Patient

acceptance

attitudes.

Am

of implant-

Henrr J 1994,127:

117~1184. MM,

Mirowki

M, Denniston

ot acetylcholine induced Clin I&T 1972$x388.

atrial

28. Dunhar

HG,

52A

33. Tacker

In: S. Saksena,

(Abstr.). JR,

right

KG,

of atrial

cardiovcrsion

electrode

ized study.

A, Deshpandc

Tuntelder

of the

DP. Radiofrequemy

in humans.

Car&

Amarasingham

arrhythmiaeusc

fibrillation

current

J Cardiowc

1991;101:S6~%3.

25. Chang

5&555. 27. Mower

KR,

S, Lurie

defibrillation

Cardid 1995;25:11OA.

JP, SchucFtler

long-term

26. Iuwn cardiac

F, Marcus

in a dog model

ablation

J, Patten

tibrillation

fibrillation.

/Am Co/l &r&l

cmfiom~c

P, Pcquet

D, Natale

fibrillation

C, l)czcm

and

G. Silvers

Boineau

RAS,

sion of atrial

46.Poole

inducibility

to atrial

for atrial

node isolation: tion.

and

l994;17:797.

1’)95;91:2235-2244.

24. Cox JI,

of atrial

F.ble JN, Zipes

JF. Prllerjels

23. Lcitch

of slow acute

fibrillation.

E, Krum

Radiofrequemy

21. Swartz

Surgery

of atrial

Garan

conversion

45. Saksena

L, C%uang

RH.

A, Pride

22. Smith

ablation

in the

Cardio! 1990,1634Y7-501.

32. &per

DeGroot

B. Lc Metaycr

K, Silverstein

M. Ablation

Cinulnliorr Lvrative

Intluence

fibrillation:

1%;4s4:276A.

Svenson

atria

M.

atrial

Y,

transwtheter

18:1869-1875. 43. RoeIke M, Trouton

slow-pathway

I lkI5-10.52.

P, Littmann

GP,

M,

and vcntricu-

a selective

D, Manz

during

L, Fischer

J, Mughal

A, Akhtar

Am Co/l Curdi

behavior

Akhtar

I~~;w;W(suppl):I-3.x.

J. Successti~l B, Hare

response

Gcncel

1w4;s:

18. AvitaIl

nodal and after

B. Pfeiffer

on ventricular

17. Haissaguerre Ekclrophy.\io/

hefore

J, Schumacher

ablation

FL, Clemcnty

fibriuation

AJ,

IY))5;91:1@%1094.

16. Tebbcnjohanns follow-up

SS, Andemon

of atriwentricular

atrial

Circulorion

pathway

Sra J, Deshpande

Charactctixion

lar response ablation.

AA.

AC,

energy

39. Johnwn

M.

modification

refractory

tachyarrhythmias

J, Sakaguchi

of electrode

Yamanouchi

Cdl

sion of atrial

1994;90:2~~2307.

lklla

patients

Fujimura

Fetter

1984$2:377-3n4. 38. Keane D, Sulkc N, Cooke

SA

91CYl7. 13. Fcld

/Am

the dog.

DN,

cardmversion

an assessment K,

low energy

conduction

N &$J

Dunhar

synchronous

36. L&y in

PACE 1993;1(,:62%36. 12. Williamson

of atrial

Am Ham J 1994;127:9?&ItJ03.

Am Cdl

1991;83:1562-1576. J. Al&on

cardiovcrsion

1992;20:707-711.

fibrillation.

Pitha

junction

of standard

success rate of atrioventricular

T&s

in the canine:

Low

JAm CX Canibl 1991;18:175~17.58.

tion.

transvenous

31. Powell

Fibrillation:

of Arrhythmias.

of atrioventricular

comparison

management

PJ, cds. Atrial

Frontiers

KJ, Fitzgerald

Kuck

ablation

10. Yeung-Lai-Wah

atrial

D,

Luw-energy

30. Kumagai

199318:

Press 1992375.3x7.

Nonphannacolog+zal

Al,

17 patirnn:

Citcularion

(;J,

X. Friday

Bowman

current

Klein New

Congress,

WM,

Margolis

High

therapy

catheter

29. Benditt

energy

3+50. 7. Lcitch

electrode

Co//C~~dio/ 1986;7:1015-ID27.

wquencing.

176%.1772.

trodcs.

Intral-dvitary

dogJAm

N, Morady

D?l,

Tobler

tihrillation

RI-I. hmcnt for study

of vartous

of intra-atrial

models

cardiovcrskm.

53. Ayers Gritlin shock

Fetter

J, Gomick

CC. Benson

DW.

THE AMERICAN JOURNAL OF CARDIOLOGY

Benditt

Lx;.

VOLUME 77

GM,

Alfcrncsr

JC. Ventricular strength

lyo4;89:413

in a sheep 422.

JANUARY 25, i996

CA,

proarrhythmic model

effects

of tmnsvenous

DO,

Sirokman

of ventticular atrial

A, Adarm cyde

defibrillation.

length C’irculofti

JM, and