JACC Vol. 13, No. 2 February 1989:491-6

491

Catheter Ablation of Accessory Pathways Using Radiofrequency Energy in the Canine Coronary Sinus

JONATHAN LANGBERG, MD, JERRY C. GRIFFIN, MD, FACC, JOHN M. HERRE, MD, FACC, MICHAEL C. CHIN, BS, MAURICE LEV, MD, FACC,* SAROJA BHARATI, MD, FACC, MELVIN M. SCHEINMAN, MD, FACC San Francisco, California and Browns Mills, New Jersey

Ablation of a left-sided accessory pathway with high energy direct-current shocks delivered by an electrode catheter in the coronary sinus is associated with the risk of coronary sinus rupture. The safety and effectiveness of closed chest catheter desiccation in the coronary sinus with use of radiofrequency energy was studied. Radiofrequency energy (174 -C 74 J) was applied between the distal electrode of a standard electrode catheter placed 3 to 6 cm inside the coronary sinus and a large posterior chest wall patch in 16 dogs.

No arrhythmias or hemodynamic changes were ob- served. Three dogs were killed approximately 1 h after ablation and 13 after 2 to 4 weeks. Lesions in the atrioven- tricular (AV) sulcus were observed in 14 of 16 dogs. Lesions

were 11.6 -C 6 mm in length, 4.3 f 2.3 mm in width and 2.8 2 1.4 mm in depth. Microscopic examination showed well circumscribed areas of necrosis and fibrosis in the fat of the AV sulcus. The media and intima of the circumflex coro- nary artery were not involved nor was the endocardium or mitral apparatus damaged in any dog. Coronary sinus thrombus was present in 3 of 16 dogs.

Large amounts of radiofrequency energy can be safely applied to the coronary sinus. The size and location of the lesions produced suggest that this technique may be useful for the interruption of left-sided accessory AV connections in humans.

(J Am Co11 Cardiol1989;13:491-6)

More than 50% of patients with the Wolff-Parkinson-White syndrome have left-sided accessory pathways (I). Surgical division of one or more pathways is indicated for patients with life-threatening arrhythmias, for those whose arrhyth- mias are refractory to medical therapy and for patients intolerant to antiarrhythmic drugs. Patients may prefer sur- gery to a lifetime of drug treatment. There has been consid- erable interest in development of a closed chest technique for interruption of left free wall accessory pathways.

The use of direct-current shock from a conventional defibrillator delivered through an electrode catheter has become an established technique for induction of atrioven- tricular (AV) node block (2). Successful ablation of postero-

From the Department of Medicine and the Cardiovascular Research Institute, University of California, San Francisco. California and the *Deborah Heart and Lung Center, Browns Mills, New Jersey. This research was supported by Grants HL07934 and HL30558-05 from the National Heart. Lung, and Blood Institute, National Institutes of Health. Bethesda, Maryland.

Manuscript received January 25, 1988: revised manuscript received Au- gust 17, 1988, accepted September 7, 1988.

-for Jonathan Langberg. MD, Room 312, Moffitt Hospi- tal, University of California, San Francisco, California 94143-0214.

01989 by the American College of Cardiology

septal accessory pathways has also been achieved by deliv- ery of shocks to the coronary sinus OS (3). Fisher et al. (4) attempted to ablate left-sided accessory pathways by deliv- ering multiple shocks within the coronary sinus. Although pre-excitation was abolished initially, accessory pathway conduction eventually recurred in all patients. In addition, coronary sinus rupture and tamponade occurred in one patient. Ward and Camm (5), using six direct-current shocks in the coronary sinus, reported a successful ablation of a left posterior pathway, but noted a peak serum creatine kinase level >I ,000 IU.

Radiofrequency current produces desiccation of tissue through resistive heating and has not been associated with generation of shock waves. Catheter ablation with use of radiofrequency energy has been shown to produce discrete areas of coagulation necrosis when applied to the ventricular endocardium (6) or His bundle regions (7). Successful abla- tion of a right-sided accessory pathway with radiofrequency has also been reported recently (8). We investigated the short- and long-term effects of experimental radiofrequency catheter desiccation within the coronary sinus. Histopatho- logic effects were assessed regarding the safety and possible

0735.1097/89/$3.50

492 LANGBERG ET AL. JACC Vol. 13. No. 2 CORONARY SINUS ABLATION USING RADIOFREQUENCY February 1989:491-6

cs CATHET

Figure 1. Apparatus used to apply radiofrequency (RF) energy to the coronary sinus (CS). The controller allowed precise timing of energy delivery and produced 0 to I V direct current signals corresponding to delivered voltage (V), current (I), power (P) and impedance (Z).

efficacy of this approach of ablation for left-sided accessory pathways near the coronary sinus.

Methods Experimental protocol. Sixteen adult mongrel dogs of

either gender weighing between 20 and 30 kg were used. Dogs were anesthetized with fentanyl, 0.04 mg, and drope- ridol, 2 mg intramuscularly followed by sodium pentobar- bital, IO to 30 mg/kg body weight intravenously. The dogs were intubated and ventilated with a Harvard volume ven- tilator. Blood pressure and electrocardiographic (ECG) leads I, II and III were monitored continuously. Hemostatic sheaths were placed in the right internal jugular vein and right femoral arteries through cutdown procedures. A 7F lumen electrode catheter (USCI, Inc.) was introduced into the internal jugular vein and advanced 3 to 6 cm into the coronary sinus. A hand injection of 5 to IO ml of contrast agent was used to confirm the location of the catheter.

Radiofrequency ablation. Lesions were produced by ap- plying radiofrequency energy between the distal electrode of the coronary sinus catheter and a large posterior chest wall patch. The output of the radiofrequency power source was routed through a microprocessor-based controller. This al- lowed timing of onset and offset of energy delivery as well as telemetry of applied voltage, current, power and impedance (Fig. I). Unmodulated radiofrequency energy of 500 to 750 kHz was used. For the first five dogs, a prototype radiofre- quency source (Oscar Hat 100) was used. This device automatically discontinued energy application if a rise in impedance occurred. To allow for a greater range of deliv- ered voltage and duration of radiofrequency energy, the subsequent 10 lesions were made with a source that had a continuous output and a higher peak voltage (Microvasive Bicap 4005).

RF LESION

Figure 2. Dimensions used to estimate radiofrequency (RF) lesion size in the AV sulcus. Length (L) and width (W) were measured from the epicardial surface and depth (D) through the middle of the lesion. LA = left atrium; LV = left ventricle; MV = mitral valve.

Analysis of results. Electrocardiogram and blood pres- sure were monitored for approximately 1 h after the ablation. Three dogs were then killed, and 13 were allowed to recover and were restudied after 3 to 5 weeks. Anesthesia was induced as before. Retrograde angiography of the coronary sinus and subselective left circumflex angiography with a 4F Judkins left coronary catheter were performed in 6 of these 13 dogs. The dogs were then killed and their heart fixed in 10% formalin solution. The area of the atrioventricular (AV) sulcus involved with the lesion was readily apparent on inspection of the epicardial surface in all but two hearts. The epicardial dimensions of the lesion were measured (Fig. 2). Blocks encompassing the lesion were sectioned serially and every 40th section was retained and stained alternately with hematoxylin-eosin and Weigert-van Gieson stains. In the two dogs whose lesions were not visible, the entire extent of the AV groove over the left ventricle was sectioned.

Results Energy delivery variables for each dog are shown in

Table 1. Applied power was 23.3 + 7.6 W in the 16 dogs. Total delivered energy was 174 ? 74 J (range 68 to 330). An abrupt rise in impedance, resulting in a fall in current and power output, was observed in I3 of the 16 dogs after 3.4 + I .2 s. Tissue impedance rose from 139 f 38 to 480 ? 67 ohms in these dogs.

All 16 dogs survived the ablation. Although the ECG was not visible during the application of energy, no arrhythmias or ST-T abnormalities were observed in the immediate postablation period or during the subsequent I to 2 h (Fig. 3). Blood pressure remained stable during and after radiofre- quency delivery (Fig. 4).

Gross morphologic findings. A macroscopic lesion was produced in 14 of the I6 dogs (Fig. 5). The mean lesion dimensions were II .6 X 4.3 X 2.8 mm (Table I). There was

JACC Vol. 13. No. 2 LANGBERG ET AL. 493 February 1989:491-6 CORONARY SINUS ABLATION USING RADIOFREQUENCY

Table 1. Radiofrequency Energy Delivery Variables and Resultant Coronary Sinus Lesion Characteristics in 16 Dogs

Radiofrequency CS Lesion Dimensions

Dog Power Energy Impedance Length Width Depth CS No. (W) (J) (ohms) (mm) (mm) (mm) LA LV Thrombus

1 15 I50 134 10 6 4 _ _ 2 28 68 80 4 3 2 + _

3 17 161 103 9 5 5 _ t t 4 35 177 200 10 4 5 _ t

5 27 192 107 13 6 3 6 17 I48 125 21, 6 3 _ _ 7 14 140 142 16 5 3 + _

8 7 101 122 15 6 2 + t 9 20 100 219 11 4 3 + _ _

IO 26 96 I25 14 3 4 _ _

I1 33 330 I75 14 9 3 t + 12 27 270 I04 15 5 2 + _ _ 13 30 244 I52 0 0 0 _ _ 14 28 126 136 0 0 0 _ _ 15 25 247 161 17 3 3 _ _

I6 24 238 I44 17 4 3 + _ t Mean 23.3 174 139 11.6 4.3 2.8 SD ~7.6 274 236 k6.0 22.3 21.4

CS = coronary sinus: LA = left atria1 involvement; LV = left ventricular involvement: - = absent; t = present

no correlation between delivered energy and resultant lesion size. There was no difference in the size of the lesions of Dogs 7,8 and 9 studied 1 h after ablation and those in the 13 dogs studied after 2 to 4 weeks. Seven of 13 lesions had superficial (< 1 mm) involvement of the epicardial surface to the left atrium and 2 involved the left ventricular epicardium. The endocardium and mitral apparatus were not involved in any dog and no pericardial reaction was observed. Two of the dogs killed 1 month after ablation had partial thrombotic occlusion and one had total occlusion of the coronary sinus at the level of the lesion. No thrombus was seen at or near the OS of the coronary sinus. Six dogs had coronary angiog- raphy immediately after ablation and again just before they

Figure 3. Surface electrocardiogram (ECG) and intracardiac elec- trograms immediately before (A) and after (B) the application of radiofrequency to the coronary sinus. Note the absence of changes in the ECG or in the coronary sinus (CS) electrograms. CS l-2 = coronary sinus bipolar electrogram recorded from poles 1 and 2: CS 3-4 = coronary sinus bipolar electrogram recorded from poles 3 and 4.

were killed. No coronary spasm or fixed lesions were seen (Fig. 6A). Various degrees of coronary sinus narrowing were seen at the level of the lesion in all cases (Fig. 6B).

Microscopic Findings.

Lesions examined 1 h after ablation (n = 3). Coagulation necrosis and hemorrhage of the fat in the AV groove were seen in all cases. The necrosis involved the full thickness of the coronary sinus and there was localized disruption of the elastic lamina (Fig. 7). The endocardial surface of the left

Figure 4. Monitoring and telemetry during application of radiofre- quency energy to the coronary sinus. Blood pressure (BP) remains stable during the ablation. Although the electrocardiogram is not visible, the arterial pressure tracing suggests that no arrhythmias occurred while current was Rowing. Curves in the middle of the tracing correspond to applied voltage, current, power and imped- ance. The arrow marks the occurrence of an abrupt rise in imped- ance, presumably due to desiccation of tissue around the distal electrode. A = amperes; R = ohms; W = watts; V = volts.

; i i i i i 1 i i i i \I I I I I I I 11, ,,,

\ _-

494 LANGBERG ET AL. JACC Vol. 13, No. 2 CORONARY SINUS ABLATION USING RADIOFREQUENCY February 1989:491-6

Figure 5. Dog 4. Gross appearance of the lesion produced in the atrioventricular sulcus 3 weeks after application of radiofrequency to the coronary sinus. Arrows denote the extent of injury. C = circumflex coronary artery; other abbreviations as in Figure 2.

atrium and ventricle were spared, as was the mitral appara- tus. One lesion extended to and involved the adventitia of the circumflex coronary artery but no changes were seen in the media or intima.

Lesions examined 3 to 5 weeks after ablation (n = 13). Eleven of the 13 dogs had pathologic changes in the AV groove. In all of these, there was damage to the coronary sinus, with intimal proliferation and degeneration of the elastic lamina. As described, three dogs had chronic throm- bus in the lumen of the coronary sinus (Fig. 8). The injury surrounding the coronary sinus was characterized by dense fibrosis and chronic inflammatory cells in the fat of the AV groove. There was cartilage formation in two of the lesions. The demarcation between fibrotic and healthy tissue was very sharp, particularly in the areas of the lesion extending onto the epicardial surface of the atrium and ventricle (Fig. 9). As with the dogs examined 1 h after ablation, there were no changes in the endocardium, mitral apparatus or the media and intima of the circumflex coronary artery.

Discussion There is considerable interest in developing a nonsurgical

alternative to a lifetime of drug therapy for patients with left lateral accessory pathways in the Wolff-Parkinson-White syndrome. Direct current shock catheter ablation iq the coronary sinus has been abandoned because of untoward

Figure 6. Angiograms. A, Dog 11. Single frame from a subselective angiogram of the circumflex coronary artery 2 weeks after delivery of 244 J to the coronary sinus. Note the normal appearance of the vessel in the region of the lesion (arrow). B, Dog 12. Retrograde angiogram of the coronary sinus 2.5 weeks after radiofrequency energy application. The narrowing in the area of the lesion is evident (arrow).

side effects including coronary sinus rupture and the possi- bility of intimal hyperplasia of the circumflex coronary artery (9). In this study, large amounts of radiofrequency energy were applied to the coronary sinus without perfora- tion, or left circumflex coronary artery, endocardial or mitral valve involvement. The lesions produced were well circum- scribed and confined to the AV sulcus. Preliminary results

JACC Vol. 13, No. 2 LANGBERG ET AL. 495 February 1989:491-6 CORONARY SINUS ABLATION USING RADIOFREQUENCY

Figure 7. Low power micrograph of a lesion examined 1 h after ablation. The arrows delineate the extent of fibrosis and hemor- rhage. Note the intimal disruption in the coronary sinus KS). Weigert-van Gieson stain; original magnification x 10, reduced by 25%. LCx = left circumflex coronary artery; other abbreviations as in Figure 2.

from other laboratories are consistent with our findings. Huang et al. (10) applied radiofrequency energy by way of an electrode catheter in the coronary sinus of 10 dogs. The lesions that resulted were similar in extent to those in our study. As in our study, there were no untoward side effects of the procedure. Jackman et al. (11) described a technique for percutaneous ablation of left free wall accessory path- ways involving application of radiofrequency energy be- tween the distal electrode of a catheter within the coronary sinus and a second electrode catheter positioned underneath the mitral apparatus. This technique has also been shown to be safe in animals and appears to produce well circum- scribed lesions in the AV sulcus and summit of the left ventricular myocardium. This technique has the potential disadvantage of introducing a catheter into the arterial circulation and the production of a lesion in the left ventric- ular endocardium with the risk of thrombus formation and arrhythmogenesis.

Figure 8. Low power micrograph of the coronary sinus (CS) lesion in Dog 4 shown in Figure 5. Arrows delineate the extent of fibrosis and necrosis. Note the superficial involvement of the epicardium of both the left atrium (LA) and the left ventricle (LV). The fibrosis extends to the adventitia of the circumflex coronary artery (LC) but spares the media and intima. Thrombus in the coronary sinus is evident. Weigert-van Gieson stain; original magnification x IO. reduced by 26%.

Limitations of the study. Because accessory pathways do not occur in experimental animals, the utility of this tech- nique for ablation of accessory pathways in humans can be inferred only by analysis of the extent and orientation of the necrosis produced. However, anatomic studies of accessory pathways in humans (12) have shown that these pathways traverse the AV sulcus between the coronary sinus and the anulus fibrosis, an area damaged in all 14 dogs in this study, in which lesions were produced. On the basis of these studies, it is reasonable to assume that, if lesions of a similar extent could be produced in humans, they would effectively interrupt an accessory pathway.

Coronary sinus thrombosis was observed in 3 of 15 dogs. In each case, the thrombus was localized to the area of ablative injury and did not extend to the OS of the coronary sinus. The effects of coronary sinus thrombosis in humans are not well characterized, although this has been noted as an incidental complication of right heart catheterization and

496 LANGBERG ET AL. JACC Vol. 13, No. 2 CORONARY SINUS ABLATION USING RADIOFREQUENCY February 1989:4914

Figure 9. High power micrograph of the boundary of the lesions in the left ventricle. Note the inflammatory cells in the necrotic lesion and the sharp demarcations between healthy and ablated tissue. Hematoxylin-eosin stain; original magnification x 200, reduced by 26%.

has not been associated with clear-cut clinical complications (12-14).

Energy delivery variables did not correlate with resultant AV sulcus lesion size. In vitro studies (15) of radiofrequency catheter ablation show a good correlation between delivered energy and lesion volume. The lack of correlation in this study may be a result of variables that were not controlled, such as force of electrode tissue contact and the ratio of electrode surface area in contact with blood pool versus that in contact with the endothelial surface of the coronary vein. In addition, the energy that was delivered after the increase in impedance may not have contributed to lesion enlarge- ment. Large amounts of energy did not result in excessively large lesions or damage to vital structures.

Clinical implications. This study suggests that radiofre- quency catheter ablation can produce lesions in the coronary sinus that may be sufficient to interrupt left-sided accessory pathways in patients with the Wolff-Parkinson-White syn- drome. If this technique proves to be safe and effective, it will offer a less morbid alternative to surgical treatment of such patients with life-threatening arrhythmias.

I.

2.

3.

4.

5.

6.

7.

8.

9.

10.

II.

12.

13.

14.

References Gallagher JJ, Pritchett ELC, Sealy WC, Kasell I, Wallace AG. The preexcitation syndromes. Prog Cardiovasc Dis 1978;20:285-322.

Scheinman MM, Evans-Bell TE, and the Executive Committee of the Percutaneous Cardiac Mapping and Ablation Registry. Catheter ablation of the atrioventricularjunction: a report of the Percutaneous Mapping and Ablation Registry. Circulation 1984;70: 1024-9.

Morady F, Scheinman MM, Winston SA, et al. Efficacy and safety of transcatheter ablation of posteroseptal accessory pathways. Circulation 1985:72:17@-7.

Fisher JD, Brodman R, Kim SG, et al. Attempted nonsurgical electrical ablation of accessory pathways via the coronary sinus in the Wolff- Parkinson-White syndrome. J Am Coil Cardiol 1984;4:685-94.

Ward DE, Camm AJ. Treatment of tachycardias associated with the Wolff-Parkinson-White syndrome by transvenous electrical ablation of accessory pathways. Br Heart J 1985;53:w8.

Huang SK, Graham AR, Hoyt RH, Odell RC. Transcatheter desiccation of the canine left ventricle using radiofrequency energy: a pilot study. Am Heart J 1987;114:42-8.

Huang SK, Bharati S, Graham AR, Lev M, Marcus FI, Odell RC. Closed chest catheter desiccation of the atrioventricular junction using radiofre- quency energy-a new method of catheter ablation. J Am Coll Cardiol 1987;9:349-58.

Borggrefe M, Budde T, Podczeck A, Breithardt G. High frequency alternating current ablation of an accessory pathway in humans. J Am Coll Cardiol 1987;10:57&82.

Brodman R, Fisher ID. Evaluation of a catheter technique for ablation of accessory pathways near the coronary sinus using a canine model. Circulation 1983:67:923-9.

Huang SK, Graham AR, Bharati S, Lee MA, Gorman G. Chronic effect of radiofrequency catheter ablation of the coronary sinus (abstr). Circu- lation 1987:76(suppl IV):IV-406.

Jackman W. Kuck K-H, Naccarelli G, Pitha J, Carmen L. Catheter ablation at the mitral annulus using radiofrequency current in canines (abstr). PACE 1987:10:410.

Becker AE, Anderson RH, Path MRC, Durrer D, Wellens HJJ. The anatomical substrates of Wolff-Parkinson-White syndrome: a clinico- pathologic correlation in seven patients. Circulation 1978;57:870-9.

Kozlowski JH. Inadvertent coronary sinus occlusion by a pulmonary artery catheter. Crit Care Med 1986:14:649.

Hazan MB, Byrnes DA, Elmquist TH, Mazzara JT. Angiographic dem- ^ onstratton ot coronary smus thrombosis: a potenttal consequence of trauma to the coronary sinus. Catheterization Cardiovasc Diagn 1982;8: 405-8.

15. Hoyt RH. Huang SK, Jordan N, Marcus F. Factors influencing trans- catheter radiofrequency ablation of the myocardium (abstr). Circulation 1985:72tsuppl 111I:111-473.