- Email: [email protected]
Radiofrequency ablation of the inferior vena cava-tricuspid valve isthmus in common atrial flutter
Radiofrequency Ablation of the Inferior Vena Cava-Tricuspid Valve Isthmus in Common Atrial Flutter Francisco G. Cosio, MD, Maria L6pez-Gil, MD, Antonio Goicolea, MD, Fernando Arribas, MD, and Jose L. Barroso, MD
Endocardial mapping has suggested that common atrial flutter (AF) is based on right atrial reentry surrounding the inferior vena cava (WC). The istk mus between the IVC and the tricuspid valve (TV) appears essential to close the circuit. To test this hypothesis, radiofrequency was applied to the II/C-N isthmus, with catheter electrodes, in 9 patients with AF. Mapping confumed a right atrial circuit surrounding the IVC in all. In 4 patients another type of AF was induced that followed the circuit in the opposite direction. Radiofrequency interrupted AF in all patients. Multiple endocardial recordings showed that interruption was due to activation block at the point of application. Radio frequency produced very brief or sustained, atrial fibrillation in 2 patients, which resulted in sinus rhythm. AF recurred in 4 patients with the same activation pattern and was interrupted again with radiofrequency in the IV07V isthmus in 3. AF was noninducible in 7 patients after 1 to 4 sessions. AF-free periods of 2 to I.9 months without drugs were observed after radiofrequency, but 2 p* tients had paroxysmal atrial fibrillation. These TV suits confirm that the IVC-TV isthmus is an ee sential part of the AF circuit. Ablation of this area may be of therapeutic value, but technical im provements are needed. Long-term efficacy of the procedure is uncertain. (Am J Cardiol1993;71:709-709)
From the Cardiology Service, Hospital Universitario de Getafe, Madrid, Spain. This study was supported by Grant PM89-0035 of Comisi6n Interministerial de Ciencia y Tecnologia, Ministerio de Education y Ciencia, Madrid, Spain. Manuscript received August 17, 1992; revised manuscript received and accepted October 5, 1992. Address for reprints: Francisco G. Cosio, MD, Jefe de Servicio de Cardiologia, Hospital Universitario de Getafe, Carretera de Toledo, 28905 Getafe, Madrid, Spain.
ommon atria1flutter (AF) is classically considered a reentrant arrhythmia basedon animal modelsL,2 and the response to programmed stimulation3 Endocardial mapping suggeststhat AF is due to a large right atrial reentry circuit? surrounding a central obstacle formed by the inferior vena cava (IVC) and adjacent areasof functional conduction block.7-9The caudal end of the circuit is closed through the myocardial isthmus between the IVC and the tricuspid valve (TV) (Figure 1). The following report describesthe effect of radiofrequency ablation of the IVC-TV isthmus, applied by way of catheter electrodeswith deflecting mechanismsand a large distal electrode. The results strongly support the proposed structure of AF circuits, and open new therapeutic perspectivesfor surgical and catheter ablation of this arrhythmia.
C
METHODS Patients: The series includes 9 consecutive patients
(Table I), 8 men and 1 woman aged 45 to 69 years (mean 59 f 8) with clinically documentedAF (Table I) for >l year. In 7 patients, AF was established for :>6 months, and in 2 it was paroxysmal. All had undergone cardioversion and had taken 22 antiarrhythmic drugs that failed to control recurrences.Associatedcardiac diseasesincluded left ventricular failure due to old myocardial infarction in 1, dilated cardiomyopathyin 1, mild chronic obstructive lung diseasein 1, previous coronary bypasssurgery in 1, systemichypertension in 1, and sick sinus syndrome in 1. Three patients had no heart disease. Mapping procedure: As described elsewhere,@ bipolar endocardial electrograms(50 to 600 Hz) were recorded sequentially from 230 points of the right atrium and coronary sinus by meansof preshapedand deflectable catheter electrodes with 0.5 to 1 cm interelectrode separation. Reference electrograms were obtained simultaneously from 3 levels of the posteroseptaland 3 of the anterolateral right atrium, with hexapolar cathebrs placed in a craniocaudal direction in these areas(Figure 1). A detlecting-tip catheter was used to record from 3 to 4 levels of the anterior, lateral, posterolateral,posterior, and septal right atrial walls, as well as the IVC-TV isthmus. Local activation times were measuredbetween the initial sharp deflections of reference and local electrograms. In double-spike electrograms,6~7 the time at the onset of each spike was recorded. The zero reference was set arbitrarily. Radiofrequency application: The patients were fasting, and were mildly sedatedwith intravenous midRADIOFREQUENCY ABLATIONOF ATRIAL FLUTfER
705
TABLE I Clinical Features, Radiofrequency Effect and Follow-Up Clinical Features Clinical Diagnosis
Age (yr) & Sex
Pt. 1 2
45M 51M
0 0
3
55M
CAD
4
56F
SH
5
61M
Mild COLD
6
63M
DC + CHF
7
64M
0
8
65M
CAD + CHF
9 Mean f SD
69M 59 +-8
sss
Rhythm Common-Px Common
Radiofrequency Application Drugs Tried F, A Q,A,PR
Flutter Changes
Ses.
Appt.
1 1 2 3 4 1
45 30 20 20 3 3
Block Block Block Block Block Block
W-TV IVC-TV WC-TV WC-TV IVC-TV WC-TV
+ + + + 0 0
1 2 1
24 15 25 18 32 15 15
Block WC-TV Block IVC-TV Block IVC-TV Fib. Disorganized Increased CL Block IVC-TV Block IVC-TV
Ind.
1 day 15 days
0 PR
+ 0 +
3 mos 7 mos 6 mos
PR 0 0
+ + 0
+ 0
18 mos 11 mos Fib.-Px
A
0 + 0 0
+ + 0
0
Q, A
Common-Px Fib.-Px Common Rare
PC, A
Common Rare Common Rare Common
A, F, P
1 2 3 1
F, A
1
27
Block IVC-TV
0
F, A, P
1
32 21 ~8
Block IVC-TV
0
F, P, A
Drug
+ +
Common Rare Common
Q, A
Recurr.
15 days 1 day 2 mos 1 mo Fib.-Px 5 mos
0 0 0 0 0
2 mos
0
Forfurther explanation see text. A = amiodarone; Appl. = number of applications; CAD = coronary artery disease; CHF = congestive heart failure; CL = cycle length; COLD = chronic obstructive lung disease; DC = dilated cardiomyopathy; F = flecainide; Fib, = atrial fibrillation; Ind. = inducible after ablation; IVC-TV = inferior vena cava-tricuspid valve Isthmus; P = propafen?ne; PC = procainamide; PR = propranolol: Px = paroxysmal; Q = quinidine; Recurr. = atrial fibrillation recurrence; Ses. = radiofrequency session number; SH = systemic hypertension; SSS = sick sinus syndrome.
azolam and fentanyl. A session including mapping and ablation lasted 3 to 5 hours. Unmodulated 500 KHz radiofrequency current (Osypka HAT 200) was applied by means of deflecting-tip catheters with large (4 mm length) distal electrodes (Mansfield or USCI). The
FIGURE1. Schematic representation of a cammon atrial tlutter circuit. The atria are represented in a left anterior view from ths atrioventricular valve rings. Shaded ms inside the valves represent the endocanlium. Openings of the venae cane, coronary sinus and pulmonary veins are shown for orientation. Hatched areas in the right atrium mark the location of functional block in the posterolateral wall, and slow conduction in the posteroseptal wall. tiws indicate the direction of activation. Large open arrow marks the infe rior vena cava-tricuspid valve isthmus. Lettering refers to recoding positions shown in Pigures 2 to 4: HA = high anterior; HPS q hi@ posteroseptal; LA = low anterior; LPS = low posteroseptal; MA = midanterior; IMPS q midposteroseptal. 706
THE AMERICANJOURNALOF CARDIOLOGY VOLUME71
catheter was advancedfrom the IVC to the right ventricle, in contact with the basal diaphragmatic wall; then it was slowly withdrawn until bipolar recordings from the distal electrodes showed atrial electrograms. Radiofreauencv (25 to 30 W X 30 to 60 seconds)was __applied ai mu&le points, every 3 to 4 mm, along the withdrawal line toward the IVC, until AF was interrupted or recordings showed absenceof atrial electrograms. When AF persisted,the procedure was repeated 3 to 4 times along the sameline, looking for areaswith sharp atrial electrograms.If still ineffective, the catheter was advancedfrom the superior vena cava in an attempt to obtain better contact. In the event of impedancerise, the electrode was withdrawn and cleaned of adherent clot before new applications. When AF could not be interrupted, the session was ended after an arbitrary number of applications, after the target area had been covered, and AF was interrupted by stimulation. When radiofrequency interrupted AF, induction was again attempted. Transthoracic bidimensional echocardiograms were recorded before and after each session. Ethical aspects: The project was approved by the Hospital Human ResearchCommittee. Informed consent was obtained from patients before each procedure. RESULTS Electrophysiologic study: Clinically documented AF was reinduced by programmed stimulation (1 to 3 atrial extrastimuli over paced cycle length of 600 to 250 ms) in all but patient 8, in whom clinical flutter was not interrupted by stimulation before the ablation procedure. In patient 1, isoproterenol infusion was necessary to induce sustainedAF. In 4 patients an atypical (“rare”) es with positive deflections in lead II was also
MARCH15,1993
induced and mapped (Table I). As in our previous experience,H AF activation surrounded the IVC; the direction was caudocranial in the septum and craniocaudal in the anterior wall (Figures 1 and 2). Double-spike electrograms were recorded in the posterolateral wall, and fragmented or double electrograms suggestive of conduction delay were recorded from the low posteroseptal area. Rare flutter in 4 patients followed the samecircuit in the opposite direction9J0 (Figure 2). In all common and rare flutter, the IVC-TV isthmus electrogram bridged the gap between the low anterior and low posteroseptal electrograms.
Slight cycle oscillation (;t 20 ms) was observed in 3 instances (patients 1 to 3) preceding interruption. Sudden stable cycle prolongation by 40 ms was observedn-r an unsuccessful session in 1 patient (no. 6). Radiofrequency produced a brief run of a faster, disorganized rhythm, leading to AF interruption in 1 patient (no. 6, Figure 4) and self limited sustainedfibrillation in auother (no. 5). In 7 patients, AF became noninduciblc after 1 to 4 radiofrequency sessions. Moderate to severepain was%feltwhen radiofrequency was applied close to the lVC, but it could be well controlled with sedation. No thrombi or pericardial efImmediate effect of radiofrequemy (Table I): fusions were detectedby echocardiographyafter the proThere was a total of 15 radiofrequency sessions(1 to 4 cedurc. sessions per patient). The number of applications per Follow-up (TaMe I): AF recurred in 3 patients 1 to sessionwas 3 to 32 (mean 21 f 8). Al! was interrupted 15 days after the lirst treatment. Patients 2 and 6 uusuddenly in all patients, at least on 1 occasion, during derwent new treatment and became finally free of reor secondsafter an application (Figure 3, Table I). In 3 currence for 2 and 7 months without antiarrhythmic instances the successful application was delivered with drugs. In 2 patients (nos. 1 and 3), AF recurred 18 and a superior vena cava approach. Electrogram sequences 6 months after ablation without antiarrhythmic drugs. during interruption showed block of activation at the Patient 1 was treated again, and has been free of AF for IVC-TV isthmus, where radiofrequency was applied. 11 months without drugs.
1 RGURE 2. Activation sequence in cam mon (IeRJ and %tW (rigm atrial flutter (AF) in patient 4. The tracings are lead II and d&t atrial electrograms Rum the high anterior (HA), midanterior (MA) and low anterior (LA) wall, inferior vena cavz+trlcuspid valve (IGT) istlsnuq and tow posteroseptal (LPS), nddposterosep tal (MP) and high posterior (HP) wall. Electvogram sequence is inverted in rare AF. Note furgmented electregrame at LPS wall in common AF and at IGT in rare AF. Activation of IGT Midges LA and LPS in both cases. Values are expressed in milliseconds
FIGURE 3. Cormnon atrial flutter intewup tion during radiofrequency application in patient 4. Tracings are lead II and electro&ams fmm hi@ anterior (HA), midanterior (MA), low anterior (LA), low pesteroseptal (LPS), midposterior (MP), and high posterior (HP) right atrium. The inferior vena cav&tvicuspid valve (M-T) ekctmgram shows radiofrequency art6 fact. Initial electrogram sequence, interrupted et the inferior vena cava-tricuspid valve, after activation of the low anterior and low postereseptal right atrium does net eccur. Values are expressed in miC liseconds.
,,,, MP , I .HP
RADIOFREQUENCY ABLATION OF ATRIAL FLUTTER 707
Recurrent AF maps in all cases showed the same circuit and cycle length (3~20 ms) as the first study. Recurrent AF was also interrupted by radiofrequency applications at the IVC-TV isthmus. The electrograms recorded from the isthmus in recurrent AF were low voltage and fragmented, except at the point where radiofrequency was successful, where they tended to be sharper. In patient 6, ablation of recurrent AF was unsuccessful in 1 session because of catheter instability, but AF could be interrupted at another session, stabilizing the catheter over the isthmus by inducing short periods of apnea, under general anesthesia. Patients 5 and 7 have had paroxysmal atria1 fibrillation after ablation. In patient 5, iibrillation had been documented previously, but only rarely, and AF was by far the predominant arrhythmia. AF did not occur after radiofrequency in these 2 patients. DISCUSSION
These observations contirm the critical role of the IVC-TV isthmus in the AF circuit, because local ablation at this point interrupted AF reproducibly by causing activation block between the anterior and posteroseptal arms of the circuit.6,7 Modification of conduction through the isthmus led to cycle oscillation and disorganization into very brief or sustained atrial fibrillation in 2 patients. Radiofrequency produces such localized lesions,11-13 that the observed effects are most likely due to local modification or ablation of the IVC-TV isthmus. Great care was applied to deliver radiofrequency precisely at the same “line” in all instances, guiding the catheter position by angling fluoroscopy, and avoiding the posteroseptal wall. AF was inducible or recurred in several cases after ablation, suggesting that interruption could result from transient isthmus modification. Patient 6 is interesting in this respect, because Al! was not inducible after the lirst
I
I.
-1000-----I k
ablation, and yet it recurred within 1.5 days. Recurrent AF showed the same activation pattern and cycle length, and could be interrupted by radiofrequency at the IVCTV isthmus in all cases, ruling out that it was due to the appearance of a different reentry circuit. In this early experience we have met with sign&ant technical dihiculties to achieve complete ablation of the isthmus with an available catheter. Catheter stability is compromised by respiration, ventricular systole and atrial contraction, which is quite evident in AF. Unstable catheter-tissue contact may explain the need for a large number of applications to obtain results, and the difficulty to produce complete interruption of the isthmus. There is a clear need for a specially designed catheter to improve the results of the procedure. Surgical cryoablation of the IVC-TV area has been reported to interrupt AE l4 However, the procedure was not limited to the isthmus, but extended toward the posteroseptal wall and Koch’s triangle. Other investigators have used direct-current shocks aimed at areas of fragmented electrograms around Koch’s triangle, with favorable results15J6 but with some risk of atrioventricular block. Direct-current shocks will interrupt AF through cardioversion, and may produce necrosis far from the catheter delivering the shock17; therefore, their effects on AF are difficult to interpret from the pathogenic point of view. On the bases of mapping data, we directed ablation to a well-detined anatomic isthmus, which appeared as a necessary link in the circuit, regardless of the type of electrograms recorded locally. This approach was described by Rosenblueth and Garcia Ramos2 as a way to make flutter noninducible in their dog model, where activation encircled both venae cavae. The lack of acute complications with our approach reflects the advantages of the isthmus as an ablation target because of its easy location and its distance to the interatrial and atrioven-
h
h
FlGURE 4. Atrial flutter disorganization during radiotkquency (RF) delivery at the inferior vena cava-tncuspui valve (RIG-TV) isthmus in patient 5, leading to interruption. The tracings are lead II and electrograms from high antedor (HA), midanterior (MA), low anterior (LA), low posteroseptal (LPS), midposteroseutal (MPS) and high posteroseptal (HPS) right atrium. Radioiiequency artifact is recorded on the inferior vena cavzttricuspid valve isthmus electrogram. Note cycle oscillation, related to changes in low anterio+l.PS electrogram interval. Anterior wall sequence is maintained, except in the last active tion where it is inverted. The relative sequence of LPS and midposteroseptal electrogram changes, probably reflecting corn plex local conduction patterns around Koch’s triangle. Values are expressed in milliseconds.
708
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 71
MARCH 15,1993
tricular conduction structures. Specifically designed catheters might improve the effectiveness of the procedure, and significantly shorten the time needed for ablation. The conlirmation of the role of the IVC-TV isthmus, as an essential part of the AF circuit, opens new perspectives for surgical or catheter ablation therapy. A larger experience is nevertheless necessary to determine precisely the safety margin of the technique. Long-term effects especially should be followed with caution. In this small group of patients AF appears to be prevented in some cases, but delayed recurrences still occur. On the other hand, development of atria1 fibrillation in 2 patients is troubling, although not totally unexpected. AF may be just one manifestation of basic anatomic or electrophysiologic disturbances, or both, of the atria,18,19 which may lead to other types of reentry such as atrial fibrillation. Criteria for optimal patient selection should be developed in the future. Acknowledgment: We are thankful to Pilar Adoue, RN, Isabel delas Fuentes, RN, and Pilar Gomez, RN, for their invaluable cooperation, and to Ana Femandez for secretarial assistance. Jose Domfnguez Pall& performed the photographic work.
1. Lewis T, Feil S, Stroud WD. Observations upon flutter and fibrillation. II. The nahue of auricular flutter. Hearr 1920;7:191-233. 2. Rosenbluetb A, Garcia Ramos J. Estudios sobre el flutter y la fib&&%. II. La influencia de 10s obst&xlos atificiales en el flutter auricular experimental. Arch lnst Cardiol Ma 1941;17: I-19. 3. Waldo AL, McLean WAH, Karp RB, Kouchoukos NT, James TN. Entrainment and interruption of ahial flutter with atrial pacing. Studies in man following open
heart surgery. Circulation 1977;56:737-745. 4. Puech P, Latour H, Grolleau R. Le flutter et ses limites. Arch Ma! Coeur 1970;63:11f%144. 5. Chauvin M, Brechenmacher C, Voegtlin JR. Application de la cartographic endocavitaire a l’&de du flutter auriculaire. Arch Mal Coeur 1983;76:1020-1030. 6. Cosio FG, Anibas F, Palacios .I, Tasc6n J, tipez-Gil M. Fragmented electmgrams and continuous electrical activity in atrial flutter. Am .I Cardiol 1986;57: 1309-1314. 7. Cosio FG, Arribas F, Barber0 JM, Kalhneyer C, Goicolea A. Validation of double spike electrograms as markers of conduction delay or block in atria1 flutter. Am J Cal-dial 1988;61:775-780. 8. Cosio FG, Goicolea A, L@ez-Gil M, Anibas F, Barroso JL. Atria1 endocaniial mapping in the rare form of atria1 flutter. Am J Cardiol 1990;66:715-720. 9. Olshansky B, Okumura K, Henthom RW, Waldo AL. Characterization of dlouble potentials in human atrial flutter: studies during transient entrainment. J Am Coil Cardiol 1990; 15:833-841. 10. Beckman K, Huang-Ta-Liig, Fmfchek J, Wyndham CRC. Classic and concealed entrainment of typical and atypical atrial flutter. PACE 1986;9:82&835. 11. Wittkampf FHN, Hauer RNW, Robles de Medina EO. Control of radiofrequency lesion size. by power regulation. Circulation 1989;80:962-968. 12. Hindricks G, Haverkamp W, Giilker H, Rissel U, Budde T, Richter KD, Borggrefe M, Breithardt G. Radiofrequency coagulation of ventricular myocardiurn: improved prediction of lesion size by monitoring catheter tip temperature. .Eur Heart J 1989;10:972-984. 13. Haines DE, Watson DD. Tissue heating during radiofrequency catheter ablation: a thermodynamic model and observations in isolated perfused and superfused canine right venbicular free wall. PACE 1989;12:962-976. 14. Klein GJ, Guimudon GM, Sharma AD, Milstein S. Demonstration of macroreentry and feasibility of operative therapy in the common type of atrial flutter. Am J Cardiol 1986;57:587-591. 15. Chauvin C, Brechenmacher C. A clinical study of que application of endocwdial fulguration in the treatment of recurrent atrial flutter. PACE 1989;12:21%2:24. 16. Saoudi N, Atallah G, Kirkorian G, Touboul P. Catheter ablation of the atria1 myocardium in human type I &al flutter. Circulation 1990; 8 1:762-77 1, 17. Lemery R, Leung TK, LavallCe E, Giiard A, Talajic M, Roy D, Montpetit M. In vitro and in viva effects within the coronary sinus of nonarcing and arcing shocks using a new system of low-energy DC ablation. Circulation 1991;83:279-293, 18. Cosio FG, Palacios J, Vidal JM, Co&a EG, G6mez-S&hex MA, Tamargo L. Electrophysiologic studies in atrial fibrillation. Slow conduction of premature impulses: a possible manifestation of the background for reentry. Am J Cardiol 1983;51:122-130. 19. Buxton AE, Waxman HL, Marchlinski FE, Josephson ME. Atrial conduction: effects of extrastimuli with and without atria1 dysrhythmias. Am J Cardiol 1984, 54:155-761.
RADIOFREQUENCY
ABLATION
OF ATRIAL FLUTTER
709
Endocardial mapping has suggested that common atrial flutter (AF) is based on right atrial reentry surrounding the inferior vena cava (WC). The istk mus between the IVC and the tricuspid valve (TV) appears essential to close the circuit. To test this hypothesis, radiofrequency was applied to the II/C-N isthmus, with catheter electrodes, in 9 patients with AF. Mapping confumed a right atrial circuit surrounding the IVC in all. In 4 patients another type of AF was induced that followed the circuit in the opposite direction. Radiofrequency interrupted AF in all patients. Multiple endocardial recordings showed that interruption was due to activation block at the point of application. Radio frequency produced very brief or sustained, atrial fibrillation in 2 patients, which resulted in sinus rhythm. AF recurred in 4 patients with the same activation pattern and was interrupted again with radiofrequency in the IV07V isthmus in 3. AF was noninducible in 7 patients after 1 to 4 sessions. AF-free periods of 2 to I.9 months without drugs were observed after radiofrequency, but 2 p* tients had paroxysmal atrial fibrillation. These TV suits confirm that the IVC-TV isthmus is an ee sential part of the AF circuit. Ablation of this area may be of therapeutic value, but technical im provements are needed. Long-term efficacy of the procedure is uncertain. (Am J Cardiol1993;71:709-709)
From the Cardiology Service, Hospital Universitario de Getafe, Madrid, Spain. This study was supported by Grant PM89-0035 of Comisi6n Interministerial de Ciencia y Tecnologia, Ministerio de Education y Ciencia, Madrid, Spain. Manuscript received August 17, 1992; revised manuscript received and accepted October 5, 1992. Address for reprints: Francisco G. Cosio, MD, Jefe de Servicio de Cardiologia, Hospital Universitario de Getafe, Carretera de Toledo, 28905 Getafe, Madrid, Spain.
ommon atria1flutter (AF) is classically considered a reentrant arrhythmia basedon animal modelsL,2 and the response to programmed stimulation3 Endocardial mapping suggeststhat AF is due to a large right atrial reentry circuit? surrounding a central obstacle formed by the inferior vena cava (IVC) and adjacent areasof functional conduction block.7-9The caudal end of the circuit is closed through the myocardial isthmus between the IVC and the tricuspid valve (TV) (Figure 1). The following report describesthe effect of radiofrequency ablation of the IVC-TV isthmus, applied by way of catheter electrodeswith deflecting mechanismsand a large distal electrode. The results strongly support the proposed structure of AF circuits, and open new therapeutic perspectivesfor surgical and catheter ablation of this arrhythmia.
C
METHODS Patients: The series includes 9 consecutive patients
(Table I), 8 men and 1 woman aged 45 to 69 years (mean 59 f 8) with clinically documentedAF (Table I) for >l year. In 7 patients, AF was established for :>6 months, and in 2 it was paroxysmal. All had undergone cardioversion and had taken 22 antiarrhythmic drugs that failed to control recurrences.Associatedcardiac diseasesincluded left ventricular failure due to old myocardial infarction in 1, dilated cardiomyopathyin 1, mild chronic obstructive lung diseasein 1, previous coronary bypasssurgery in 1, systemichypertension in 1, and sick sinus syndrome in 1. Three patients had no heart disease. Mapping procedure: As described elsewhere,@ bipolar endocardial electrograms(50 to 600 Hz) were recorded sequentially from 230 points of the right atrium and coronary sinus by meansof preshapedand deflectable catheter electrodes with 0.5 to 1 cm interelectrode separation. Reference electrograms were obtained simultaneously from 3 levels of the posteroseptaland 3 of the anterolateral right atrium, with hexapolar cathebrs placed in a craniocaudal direction in these areas(Figure 1). A detlecting-tip catheter was used to record from 3 to 4 levels of the anterior, lateral, posterolateral,posterior, and septal right atrial walls, as well as the IVC-TV isthmus. Local activation times were measuredbetween the initial sharp deflections of reference and local electrograms. In double-spike electrograms,6~7 the time at the onset of each spike was recorded. The zero reference was set arbitrarily. Radiofrequency application: The patients were fasting, and were mildly sedatedwith intravenous midRADIOFREQUENCY ABLATIONOF ATRIAL FLUTfER
705
TABLE I Clinical Features, Radiofrequency Effect and Follow-Up Clinical Features Clinical Diagnosis
Age (yr) & Sex
Pt. 1 2
45M 51M
0 0
3
55M
CAD
4
56F
SH
5
61M
Mild COLD
6
63M
DC + CHF
7
64M
0
8
65M
CAD + CHF
9 Mean f SD
69M 59 +-8
sss
Rhythm Common-Px Common
Radiofrequency Application Drugs Tried F, A Q,A,PR
Flutter Changes
Ses.
Appt.
1 1 2 3 4 1
45 30 20 20 3 3
Block Block Block Block Block Block
W-TV IVC-TV WC-TV WC-TV IVC-TV WC-TV
+ + + + 0 0
1 2 1
24 15 25 18 32 15 15
Block WC-TV Block IVC-TV Block IVC-TV Fib. Disorganized Increased CL Block IVC-TV Block IVC-TV
Ind.
1 day 15 days
0 PR
+ 0 +
3 mos 7 mos 6 mos
PR 0 0
+ + 0
+ 0
18 mos 11 mos Fib.-Px
A
0 + 0 0
+ + 0
0
Q, A
Common-Px Fib.-Px Common Rare
PC, A
Common Rare Common Rare Common
A, F, P
1 2 3 1
F, A
1
27
Block IVC-TV
0
F, A, P
1
32 21 ~8
Block IVC-TV
0
F, P, A
Drug
+ +
Common Rare Common
Q, A
Recurr.
15 days 1 day 2 mos 1 mo Fib.-Px 5 mos
0 0 0 0 0
2 mos
0
Forfurther explanation see text. A = amiodarone; Appl. = number of applications; CAD = coronary artery disease; CHF = congestive heart failure; CL = cycle length; COLD = chronic obstructive lung disease; DC = dilated cardiomyopathy; F = flecainide; Fib, = atrial fibrillation; Ind. = inducible after ablation; IVC-TV = inferior vena cava-tricuspid valve Isthmus; P = propafen?ne; PC = procainamide; PR = propranolol: Px = paroxysmal; Q = quinidine; Recurr. = atrial fibrillation recurrence; Ses. = radiofrequency session number; SH = systemic hypertension; SSS = sick sinus syndrome.
azolam and fentanyl. A session including mapping and ablation lasted 3 to 5 hours. Unmodulated 500 KHz radiofrequency current (Osypka HAT 200) was applied by means of deflecting-tip catheters with large (4 mm length) distal electrodes (Mansfield or USCI). The
FIGURE1. Schematic representation of a cammon atrial tlutter circuit. The atria are represented in a left anterior view from ths atrioventricular valve rings. Shaded ms inside the valves represent the endocanlium. Openings of the venae cane, coronary sinus and pulmonary veins are shown for orientation. Hatched areas in the right atrium mark the location of functional block in the posterolateral wall, and slow conduction in the posteroseptal wall. tiws indicate the direction of activation. Large open arrow marks the infe rior vena cava-tricuspid valve isthmus. Lettering refers to recoding positions shown in Pigures 2 to 4: HA = high anterior; HPS q hi@ posteroseptal; LA = low anterior; LPS = low posteroseptal; MA = midanterior; IMPS q midposteroseptal. 706
THE AMERICANJOURNALOF CARDIOLOGY VOLUME71
catheter was advancedfrom the IVC to the right ventricle, in contact with the basal diaphragmatic wall; then it was slowly withdrawn until bipolar recordings from the distal electrodes showed atrial electrograms. Radiofreauencv (25 to 30 W X 30 to 60 seconds)was __applied ai mu&le points, every 3 to 4 mm, along the withdrawal line toward the IVC, until AF was interrupted or recordings showed absenceof atrial electrograms. When AF persisted,the procedure was repeated 3 to 4 times along the sameline, looking for areaswith sharp atrial electrograms.If still ineffective, the catheter was advancedfrom the superior vena cava in an attempt to obtain better contact. In the event of impedancerise, the electrode was withdrawn and cleaned of adherent clot before new applications. When AF could not be interrupted, the session was ended after an arbitrary number of applications, after the target area had been covered, and AF was interrupted by stimulation. When radiofrequency interrupted AF, induction was again attempted. Transthoracic bidimensional echocardiograms were recorded before and after each session. Ethical aspects: The project was approved by the Hospital Human ResearchCommittee. Informed consent was obtained from patients before each procedure. RESULTS Electrophysiologic study: Clinically documented AF was reinduced by programmed stimulation (1 to 3 atrial extrastimuli over paced cycle length of 600 to 250 ms) in all but patient 8, in whom clinical flutter was not interrupted by stimulation before the ablation procedure. In patient 1, isoproterenol infusion was necessary to induce sustainedAF. In 4 patients an atypical (“rare”) es with positive deflections in lead II was also
MARCH15,1993
induced and mapped (Table I). As in our previous experience,H AF activation surrounded the IVC; the direction was caudocranial in the septum and craniocaudal in the anterior wall (Figures 1 and 2). Double-spike electrograms were recorded in the posterolateral wall, and fragmented or double electrograms suggestive of conduction delay were recorded from the low posteroseptal area. Rare flutter in 4 patients followed the samecircuit in the opposite direction9J0 (Figure 2). In all common and rare flutter, the IVC-TV isthmus electrogram bridged the gap between the low anterior and low posteroseptal electrograms.
Slight cycle oscillation (;t 20 ms) was observed in 3 instances (patients 1 to 3) preceding interruption. Sudden stable cycle prolongation by 40 ms was observedn-r an unsuccessful session in 1 patient (no. 6). Radiofrequency produced a brief run of a faster, disorganized rhythm, leading to AF interruption in 1 patient (no. 6, Figure 4) and self limited sustainedfibrillation in auother (no. 5). In 7 patients, AF became noninduciblc after 1 to 4 radiofrequency sessions. Moderate to severepain was%feltwhen radiofrequency was applied close to the lVC, but it could be well controlled with sedation. No thrombi or pericardial efImmediate effect of radiofrequemy (Table I): fusions were detectedby echocardiographyafter the proThere was a total of 15 radiofrequency sessions(1 to 4 cedurc. sessions per patient). The number of applications per Follow-up (TaMe I): AF recurred in 3 patients 1 to sessionwas 3 to 32 (mean 21 f 8). Al! was interrupted 15 days after the lirst treatment. Patients 2 and 6 uusuddenly in all patients, at least on 1 occasion, during derwent new treatment and became finally free of reor secondsafter an application (Figure 3, Table I). In 3 currence for 2 and 7 months without antiarrhythmic instances the successful application was delivered with drugs. In 2 patients (nos. 1 and 3), AF recurred 18 and a superior vena cava approach. Electrogram sequences 6 months after ablation without antiarrhythmic drugs. during interruption showed block of activation at the Patient 1 was treated again, and has been free of AF for IVC-TV isthmus, where radiofrequency was applied. 11 months without drugs.
1 RGURE 2. Activation sequence in cam mon (IeRJ and %tW (rigm atrial flutter (AF) in patient 4. The tracings are lead II and d&t atrial electrograms Rum the high anterior (HA), midanterior (MA) and low anterior (LA) wall, inferior vena cavz+trlcuspid valve (IGT) istlsnuq and tow posteroseptal (LPS), nddposterosep tal (MP) and high posterior (HP) wall. Electvogram sequence is inverted in rare AF. Note furgmented electregrame at LPS wall in common AF and at IGT in rare AF. Activation of IGT Midges LA and LPS in both cases. Values are expressed in milliseconds
FIGURE 3. Cormnon atrial flutter intewup tion during radiofrequency application in patient 4. Tracings are lead II and electro&ams fmm hi@ anterior (HA), midanterior (MA), low anterior (LA), low pesteroseptal (LPS), midposterior (MP), and high posterior (HP) right atrium. The inferior vena cav&tvicuspid valve (M-T) ekctmgram shows radiofrequency art6 fact. Initial electrogram sequence, interrupted et the inferior vena cava-tricuspid valve, after activation of the low anterior and low postereseptal right atrium does net eccur. Values are expressed in miC liseconds.
,,,, MP , I .HP
RADIOFREQUENCY ABLATION OF ATRIAL FLUTTER 707
Recurrent AF maps in all cases showed the same circuit and cycle length (3~20 ms) as the first study. Recurrent AF was also interrupted by radiofrequency applications at the IVC-TV isthmus. The electrograms recorded from the isthmus in recurrent AF were low voltage and fragmented, except at the point where radiofrequency was successful, where they tended to be sharper. In patient 6, ablation of recurrent AF was unsuccessful in 1 session because of catheter instability, but AF could be interrupted at another session, stabilizing the catheter over the isthmus by inducing short periods of apnea, under general anesthesia. Patients 5 and 7 have had paroxysmal atria1 fibrillation after ablation. In patient 5, iibrillation had been documented previously, but only rarely, and AF was by far the predominant arrhythmia. AF did not occur after radiofrequency in these 2 patients. DISCUSSION
These observations contirm the critical role of the IVC-TV isthmus in the AF circuit, because local ablation at this point interrupted AF reproducibly by causing activation block between the anterior and posteroseptal arms of the circuit.6,7 Modification of conduction through the isthmus led to cycle oscillation and disorganization into very brief or sustained atrial fibrillation in 2 patients. Radiofrequency produces such localized lesions,11-13 that the observed effects are most likely due to local modification or ablation of the IVC-TV isthmus. Great care was applied to deliver radiofrequency precisely at the same “line” in all instances, guiding the catheter position by angling fluoroscopy, and avoiding the posteroseptal wall. AF was inducible or recurred in several cases after ablation, suggesting that interruption could result from transient isthmus modification. Patient 6 is interesting in this respect, because Al! was not inducible after the lirst
I
I.
-1000-----I k
ablation, and yet it recurred within 1.5 days. Recurrent AF showed the same activation pattern and cycle length, and could be interrupted by radiofrequency at the IVCTV isthmus in all cases, ruling out that it was due to the appearance of a different reentry circuit. In this early experience we have met with sign&ant technical dihiculties to achieve complete ablation of the isthmus with an available catheter. Catheter stability is compromised by respiration, ventricular systole and atrial contraction, which is quite evident in AF. Unstable catheter-tissue contact may explain the need for a large number of applications to obtain results, and the difficulty to produce complete interruption of the isthmus. There is a clear need for a specially designed catheter to improve the results of the procedure. Surgical cryoablation of the IVC-TV area has been reported to interrupt AE l4 However, the procedure was not limited to the isthmus, but extended toward the posteroseptal wall and Koch’s triangle. Other investigators have used direct-current shocks aimed at areas of fragmented electrograms around Koch’s triangle, with favorable results15J6 but with some risk of atrioventricular block. Direct-current shocks will interrupt AF through cardioversion, and may produce necrosis far from the catheter delivering the shock17; therefore, their effects on AF are difficult to interpret from the pathogenic point of view. On the bases of mapping data, we directed ablation to a well-detined anatomic isthmus, which appeared as a necessary link in the circuit, regardless of the type of electrograms recorded locally. This approach was described by Rosenblueth and Garcia Ramos2 as a way to make flutter noninducible in their dog model, where activation encircled both venae cavae. The lack of acute complications with our approach reflects the advantages of the isthmus as an ablation target because of its easy location and its distance to the interatrial and atrioven-
h
h
FlGURE 4. Atrial flutter disorganization during radiotkquency (RF) delivery at the inferior vena cava-tncuspui valve (RIG-TV) isthmus in patient 5, leading to interruption. The tracings are lead II and electrograms from high antedor (HA), midanterior (MA), low anterior (LA), low posteroseptal (LPS), midposteroseutal (MPS) and high posteroseptal (HPS) right atrium. Radioiiequency artifact is recorded on the inferior vena cavzttricuspid valve isthmus electrogram. Note cycle oscillation, related to changes in low anterio+l.PS electrogram interval. Anterior wall sequence is maintained, except in the last active tion where it is inverted. The relative sequence of LPS and midposteroseptal electrogram changes, probably reflecting corn plex local conduction patterns around Koch’s triangle. Values are expressed in milliseconds.
708
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 71
MARCH 15,1993
tricular conduction structures. Specifically designed catheters might improve the effectiveness of the procedure, and significantly shorten the time needed for ablation. The conlirmation of the role of the IVC-TV isthmus, as an essential part of the AF circuit, opens new perspectives for surgical or catheter ablation therapy. A larger experience is nevertheless necessary to determine precisely the safety margin of the technique. Long-term effects especially should be followed with caution. In this small group of patients AF appears to be prevented in some cases, but delayed recurrences still occur. On the other hand, development of atria1 fibrillation in 2 patients is troubling, although not totally unexpected. AF may be just one manifestation of basic anatomic or electrophysiologic disturbances, or both, of the atria,18,19 which may lead to other types of reentry such as atrial fibrillation. Criteria for optimal patient selection should be developed in the future. Acknowledgment: We are thankful to Pilar Adoue, RN, Isabel delas Fuentes, RN, and Pilar Gomez, RN, for their invaluable cooperation, and to Ana Femandez for secretarial assistance. Jose Domfnguez Pall& performed the photographic work.
1. Lewis T, Feil S, Stroud WD. Observations upon flutter and fibrillation. II. The nahue of auricular flutter. Hearr 1920;7:191-233. 2. Rosenbluetb A, Garcia Ramos J. Estudios sobre el flutter y la fib&&%. II. La influencia de 10s obst&xlos atificiales en el flutter auricular experimental. Arch lnst Cardiol Ma 1941;17: I-19. 3. Waldo AL, McLean WAH, Karp RB, Kouchoukos NT, James TN. Entrainment and interruption of ahial flutter with atrial pacing. Studies in man following open
heart surgery. Circulation 1977;56:737-745. 4. Puech P, Latour H, Grolleau R. Le flutter et ses limites. Arch Ma! Coeur 1970;63:11f%144. 5. Chauvin M, Brechenmacher C, Voegtlin JR. Application de la cartographic endocavitaire a l’&de du flutter auriculaire. Arch Mal Coeur 1983;76:1020-1030. 6. Cosio FG, Anibas F, Palacios .I, Tasc6n J, tipez-Gil M. Fragmented electmgrams and continuous electrical activity in atrial flutter. Am .I Cardiol 1986;57: 1309-1314. 7. Cosio FG, Arribas F, Barber0 JM, Kalhneyer C, Goicolea A. Validation of double spike electrograms as markers of conduction delay or block in atria1 flutter. Am J Cal-dial 1988;61:775-780. 8. Cosio FG, Goicolea A, L@ez-Gil M, Anibas F, Barroso JL. Atria1 endocaniial mapping in the rare form of atria1 flutter. Am J Cardiol 1990;66:715-720. 9. Olshansky B, Okumura K, Henthom RW, Waldo AL. Characterization of dlouble potentials in human atrial flutter: studies during transient entrainment. J Am Coil Cardiol 1990; 15:833-841. 10. Beckman K, Huang-Ta-Liig, Fmfchek J, Wyndham CRC. Classic and concealed entrainment of typical and atypical atrial flutter. PACE 1986;9:82&835. 11. Wittkampf FHN, Hauer RNW, Robles de Medina EO. Control of radiofrequency lesion size. by power regulation. Circulation 1989;80:962-968. 12. Hindricks G, Haverkamp W, Giilker H, Rissel U, Budde T, Richter KD, Borggrefe M, Breithardt G. Radiofrequency coagulation of ventricular myocardiurn: improved prediction of lesion size by monitoring catheter tip temperature. .Eur Heart J 1989;10:972-984. 13. Haines DE, Watson DD. Tissue heating during radiofrequency catheter ablation: a thermodynamic model and observations in isolated perfused and superfused canine right venbicular free wall. PACE 1989;12:962-976. 14. Klein GJ, Guimudon GM, Sharma AD, Milstein S. Demonstration of macroreentry and feasibility of operative therapy in the common type of atrial flutter. Am J Cardiol 1986;57:587-591. 15. Chauvin C, Brechenmacher C. A clinical study of que application of endocwdial fulguration in the treatment of recurrent atrial flutter. PACE 1989;12:21%2:24. 16. Saoudi N, Atallah G, Kirkorian G, Touboul P. Catheter ablation of the atria1 myocardium in human type I &al flutter. Circulation 1990; 8 1:762-77 1, 17. Lemery R, Leung TK, LavallCe E, Giiard A, Talajic M, Roy D, Montpetit M. In vitro and in viva effects within the coronary sinus of nonarcing and arcing shocks using a new system of low-energy DC ablation. Circulation 1991;83:279-293, 18. Cosio FG, Palacios J, Vidal JM, Co&a EG, G6mez-S&hex MA, Tamargo L. Electrophysiologic studies in atrial fibrillation. Slow conduction of premature impulses: a possible manifestation of the background for reentry. Am J Cardiol 1983;51:122-130. 19. Buxton AE, Waxman HL, Marchlinski FE, Josephson ME. Atrial conduction: effects of extrastimuli with and without atria1 dysrhythmias. Am J Cardiol 1984, 54:155-761.
RADIOFREQUENCY
ABLATION
OF ATRIAL FLUTTER
709