Surgical treatment of atrial fibrillation using radiofrequency ablation

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (http

s

://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Khargi, K.

Publication date

2005

Link to publication

Citation for published version (APA):

Khargi, K. (2005). Surgical treatment of atrial fibrillation using radiofrequency ablation.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)

and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open

content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please

let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material

inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter

to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

will be contacted as soon as possible.

(2)

The Saline Irrigated Cooled Tip Radiofrequency

Ablation was an Effective technique to perform the

Maze Procedure; a prospective randomized study

6 '

7 Krishna Khargi, Thomas Deneke, Bernd Lemke, Helmut Haardt,

Klaus-Michael Muller, Axel Laczkovics

Cardiothoracic Surgery, Cardiology and Pathology, Berufsgenossenschaftliche Kliniken BergmannnsheiTUniversity Hospital Bochum, Germany.

6 Oral Presentation at the 7th annual CTT Meeting, Cardio thoracic and Technologies, Current trends in Thoracic Surgery VII,

New Orleans, USA, January 24-27, 2001.

(3)

CHAPTER 4

Saline-Irrigated, Cooled-Tip Radiofrequency

Ablation Is an Effective Technique to Perform the

Maze Procedure

Krishna Khargi, MD, Thomas Deneke, MD, Helmut Haardt, ME, Bernd Lemke, MD,

Peter Grewe, MD, Klaus-Michael Muller, MD, and Axel Laczkovics, MD

Departments of Cardiothoracic Surgery, Cardiology, and Pathology, Berufsgenossenschaftliche, Kliniken Bergmannnsheil-Univcrsity Hospital Bochum, Bochum, Germany

Background. We evaluated the effectiveness of the

s a l i n e - i r r i g a t e d - c o o l e d - t i p - r a d i o f r e q u e n c y a b l a t i o n (SICTRA) to produce linear intraatrial lesions.

Methods. Thirty patients with chronic atrial fibrillation

and mitral valve disease were consecutively randomized to have mitral valve operation either with a Maze proce-dure (group A) or without (group B). Intraatrial linear lesions were made with an SICTRA catheter (20 to 32 W; 200 to 320 mL/h saline). An echocardiography and 24-h o u r e l e c t r o c a r d i o g r a m w e r e o b t a i n e d 12 m o n t 24-h s postoperatively.

Results. The cumulative frequencies of sinus rhythm

in group A and B were 0.80 and 0.27 (p < 0.01). Restored biatrial contraction was present in 66.7% (6 of 9) of the

T

he "cut and s e w " technique is most effective to create intraatrial transmural lesions, which act as electro-physiological barriers, interrupting the multiple wavelet reentry circuits, resulting in extinction of atrial fibrilla-tion. The Maze procedure, as described by Cox a n d colleagues [11, is a blueprint comprising the precise localization of the various biatrial incisions. However, this operation is, in our opinion, extensive and complex. Substantial surgical experience is required to match the excellent results reported by Cox and colleagues. To facilitate t h e operative p r o c e d u r e , we used saline-irrigated, cooled-tip radiofrequency ablation (SICTRA) to create intraatrial linear lesions. The rationale of the SICTRA is the ability to create intraatrial transmural functional linear lesions, which act as electrophysiologi-cal barriers, but without causing any atrial tissue dehis-cence, therefore obviating the need for "sewing the cut edges." However, the effectiveness of the SICTRA to abolish atrial fibrillation and safety of the technique need to be evaluated. Therefore this prospective randomized study was designed and initiated at our institution.

Presented at the Seventh Annual Cardiothoracic Techniques and Tech-nologies Meeting 2001, New Orleans, LA, jan 24-27, 2001.

Address reprint requests to Dr Khargi, Berufsgenossenschaftliche Klini-ken Bergmannsheil-University Hospital Bochum, Buerkle de la Camp Platz 1, 44789 Bochum, Germany; e-mail: krishna.khargi@ruhr-uni-bochum.de.

group A patients in sinus rhythm. One patient from each group received a permanent pacemaker because of bra-dycardia. A fatal renal bleeding and mediastinitis oc-curred in 2 group A patients, 6 weeks postoperatively. O n e group A patient had sudden cardiac death at home, 4 months after operation. One patient from each group had lethal respiratory failure, 7 and 10 months after operation. Survival after 12 months for group A and B was 73% and 93% (p = 0.131).

Conclusions. The SICTRA appeared to be an effective

technique to perform the Maze procedure.

P a t i e n t s a n d M e t h o d s

Between February 1998 and October 1999, 30 patients with d o c u m e n t e d chronic atrial fibrillation, preexisting for more than 1 year, and mitral valve disease were randomized consecutively to have mitral valve opera-tions either with a Maze procedure (group A) or without (group B). The SICTRA was used to create the intraatrial linear lesions. The clinical history, an electrocardiogram, an echocardiography, and a 24-hour electrocardiogram were obtained at 6 and 12 m o n t h s postoperatively. The medical ethical committee of our institution approved the study and informed consent was obtained from each patient.

SICTRA Set Up

The SICTRA set-up constituted of a SICTRA catheter (Sprinklr; Medtronic, Minneapolis, MN), which had a 7F (2.33 mm) diameter, a 4-mm tip length, and 13 irrigation holes. The catheter was connected by an infusion p u m p with a 0.9% NaCl infusion bag. The flow rate for the first 4 patients was 200 mL/h, but was altered to 250 mL/h in next 4 patients, and was eventually changed to 320 mL/h in the last 7 patients. The reason for these changes was that the ablation took too much time; we believed that a higher irrigation volume could accelerate the ablation procedure. The catheter was also connected to a radio-frequency generator (CardioRythm-ATAKR, Medtronic). In the first 4 patients the ATAKR was programmed to

0003-4975/01/S20.00 PII S0003-4975(01 )02940-X

(4)

Ann Thorac Surg 2001;72:S1090-5

CTT SUPPLEMENT KHARGI ET AL S1091 COOLED-TIP RADIOFREQUENCY MAZE PROCEDURE

B

•

-S:

**'"*—m**

V

R1 R2

Fig 1. Outer aspect of the right atrium as seen from the surgeon's side. The surgical incisions are SI, S2, and S3. The saline-irrigated, cooied-tip radiofretjuency ablation lesions are Rl and R2. AV = atrioventricular; IVC - inferior caval vein; RA — right atrium;

RAA = right atrial appendage; SA = sinoatrial; SCV = superior

caval vein.

provide 20 W. Because we thought that the energy delivery was too low, we increased the energy supply to 32 VV for the last 8 patients. The ATAKR was also connected to a 300 X 175 mm indifferent electrode, which was placed firmly between both scapulas of the patient, and to a foot pedal to switch the ATAKR generator on.

Maze Procedure

A standard median sternotomy was performed. The

aorta, the superior caval vein (SCV), and the inferior caval vein (ICV) were cannulated. The "cut a n d s e w " incisions (abbreviated as S-lesions) and the S1CTRA lesions (abbreviated as R-lesions) are shown in Figures 1, 2, and 3. The right a p p e n d a g e was excised (SI). A perpendicular incision 3 to 4 cm long was m a d e from the middle of the Si lesion, traversing over the lateral free wall of the right atrium (S2). A curved incision was m a d e from the atrioventricular (AV) groove, about 2 to 3 cm cranially and anterior from the ICV, and continuing posterocranially behind the sulcus terminalis (S3). Then the Rl and R2 lesions were created from, respectively, the posterocranial end of S3 into the SCV and from the posterocaudal end of S3 into the ICV (Fig 1). Lesion R3 was m a d e from the anterior edge of S3, close to the AV groove, traversing over the endocardium to the middle of the posterior part of the tricuspid annulus. Lesion R4 was m a d e from the medial cut edge from the right a p p e n d a g e to the anteroseptal commissural area of the tricuspid valve (R4). The R5 lesion was m a d e from the posterocra-nial edge of S3 traversing to the posterior area of the foramen ovale continuing to the posterior rim of the coronary sinus orifice, then curving toward the inferior caval vein, ablating the so-called isthmus (Fig 2).

The aorta was cross-clamped and cold antegrade blood cardioplegia was administered. The left atrium was o p e n e d in the interatrial groove and its dome (S4) (Fig 3).

K4 <4

* • R5

Fig 2. The inner aspect of the right atrium as opened through the curved incision (S3). View from the surgeon's side (see text). TJie saline-irrigated, cooled-tip radiofrequency ablation (SICTRA) lesions are R3, R4, and R5 ("white lines). AV = atrioventricular; FO = foramen ovale; IVC = inferior caval vein; TV = tricuspid valve;

SC = coronary sinus; SCV = superior caval vein; TK = triangle of

Koch.

The endocardial rim of the ostium of the right superior and inferior (R6) as well as the left superior and inferior (R7) pulmonary vein was ablated. Lesion R8 was the connection between R6 and R7. From the left inferior pulmonary vein, lesion R9 was created to the midportion of the posterior mitral annulus. Then lesion RIO was

Fig 3. The inner aspect of the left atrium after opening its dome and the interatrial septum (S4). The d a s h e d line is the posterior part of the mitral annulus. The saline-irrigated, cooled-tip radiofre-quency ablation (SICTRA) lesions are R6-R10, S4, and S5 (white

lines) (see text). AMV - anterior mitral valve; CX = circumflex; FO = foramen ovale; LAA = left atrial appendage; LIPV = left

inferior pulmonary -vein; LSPV = left superior pulmonary vein;

RIPV = right inferior pulmonary vein; RSPV = right superior

(5)

CHAPTER 4

SI 092 CTT SUPPLEMENT KHARGI ET AL

COOLED-TIP RADIOFREQUENCY MAZE PROCEDURE

A n n Thorac Surg 2001 ;72:S1090-5

m a d e from the left lateral rim of the orifice of the left inferior pulmonary vein to the rim of the left atrial a p p e n d a g e orifice. The left atrial a p p e n d a g e was resected (S5) or ablated, if adhesions prevented its resection, and closed with Prolene 4-0 suture (Ethicon, Somerville, NJ). The left atrium was closed with a single row of running Prolene 4-0 suture. The cross-clamp was removed and the incisions of the right atrium were closed with a pledget-buttressed mattress Prolene 4-0 running suture. The heart was paced with the Atrium atrium inhibition m o d e (AAI), if possible, or DDD mode. The patient was w e a n e d from cardiopulmonary bypass and the chest closed in a standard way.

Postoperative Care

The patients were kept on atrium atrium inhibition mode (AAI) or double double double mode (DDD) pacing if the heart rate was slower than 75 beats per minute during the first 7 postoperative days. If an atria! fibrillation per-sisted, a cardioversion was performed during the first 24 postoperative hours. However, this early cardioversion was a b a n d o n e d in the last 10 patients because the pro-cedure did not contribute to any long-term cardiac rhyth-mic stability. Sotalol 40 mg twice a day (bid) was started on the first postoperative day. The dose was increased to 80 mg bid on the third postoperative day and eventually to 160 mg bid if no bradyarrhythmia was noticed. All patients received Coumadin (warfarin sodium), started on the first postoperative day.

Folïow-Up

All data were collected between February 1998 and October 2000. Data acquisition was obtained for each patient on the first postoperative day, 12th postoperative day (predischarge), and after the third, sixth, and ninth postoperative month. The medical history, clinical exam-ination, and an electrocardiogram (ECG) were obtained at each visit. A 24 hour-ECG analysis was performed after 6 and 12 months. A transthoracic echocardiography, including t r a n s m u r a l and transtricuspid Doppler exami-nation, was obtained on the 12th postoperative day, after 6 and 12 months. Sotalol, at least 80 mg bid, was contin-ued for 6 m o n t h s and replaced by metoprolol, at least 95 mg per day. All patients received Coumadin targeting an international normalized ratio value between 2.2 and 2.5. If an atrial fibrillation was observed after the 12th postoperative day, 3rd, or 6th month, a cardioversion with 240 to 360 J was performed twice with two different defibrillator pad positions, twice during the follow-up period.

Analysis

Continuous variables were expressed as mean (standard deviation (median). Student's unpaired r test (two-tailed) was used for comparison in between the two groups. Differences were considered significant at a p value less than 0.05. The survival rate and maintenance rate of sinus rhythm (SR) were calculated according to Kaplan-Meier method and groups were compared using log rank-test (significant difference postulated at p < 0.05).

Table J. Operative Data

O p e r a t i v e t i m e (min) CPB time (min) X-time (min) M V P ( n ) Mechanical valve (n) Biological valve (n) G r o u p A (n - 15) 270(232-323) 188(165-230) 103(86-134) 14 1 G r o u p B (n = 15) 190(128-314) 127(60-197) 84(38-112) 2 13 P < 0 . 0 5 < 0.05 < 0 . 0 5 CPB = cardiopulmonary bypass; time _ aortic cross-clamp time.

M V P = mitral valve plasty;

R e s u l t s

The patient characteristics, except for age (64.7 versus 69.7 years; p ~ 0.05) and sex distribution (female to male ratio: 9:6 versus 12:3), were similar in both groups. Table 1 shows the operative data. The 30-day mortality was zero. Postoperative morbidity included respiratory insuf-ficiency (1 patient in group B), superficial wound infec-tion (3 patients in group A, 1 patient in group B), and sternal instability (1 patient in group A, 1 patient in group B patient). During follow-up, fatal renal bleeding and mediastinitis occurred in 2 patients in group A, 6 weeks postoperatively. O n e group A patient had a s u d d e n cardiac death at home, 4 months after the operation. O n e patient from each group had lethal respiratory failure, 7 respectively and 10 months after the operation. O n e patient from each group received a p e r m a n e n t pace-maker because of a bradycardia. The 12-month follow-up was complete, although 2 group B patients were unable to revisit our outpatient cardiology clinics. Both patients, however, had an ECG, which showed atrial fibrillation. Survival after 12 m o n t h s for group A and B was 73% (11 of 15 patients) and 93% (14 of 15 patients) (p = 0.131). The respective cumulative frequencies of SR after 6 and 12 m o n t h s for the group A and B patients were 0.733 and 0.267 for group A and 0.800 and 0.267 for group B (p = 0.005) (Fig 4). The n u m b e r of group A patients who were in SR with an atrial contraction (transmitral A-wave) was 5 of 8 (62.5%) after the 12th postoperative day, 7 of 10

» Post Operative Days Fig 4. Postoperative cumulative frequencies of sinus rhythm (SR), (Y-axis = the cumulative frequency of SR; X-axis = postoperative days).

(6)

Ann Thorac Surg 2001 ;72:S1090-5

(70%) after 6 months, and 6 of 9 (66.7%) after 12 months. For the group B patients who were in SR these corre-sponding figures were 4 of 4 (100%), 4 of 4 (100%), and 3 of 3 (100%). Right atrial contraction (transtricuspid A-wave) was detected in all patients in SR.

C o m m e n t

The SICTRA is an alternative source of energy to produce intraatrial linear functional transmural lesions. It is per-formed with a cooled porous radiofrequency ablation catheter, which was originally used by Wittkampf and colleagues [2]. The use of such a catheter impedes an impedance upstroke because the electrode-tissue surface temperature will drop due to the continuous saline irrigation. Therefore, the total a m o u n t of radiofrequency energy that can be delivered will be higher and conse-quently a deeper tissue lesion can b e created. Sie and colleagues [3] proposed the intraoperative use of the SICTRA to treat patients with chronic atrial fibrillation. Nakagawa and associates [4] investigated and compared t h e i n d u c e d lesion g e o m e t r y of t h e t e m p e r a t u r e -controlled, radiofrequency ablation, the conventional " d r y " radiofrequency ablation, and the SICTRA. They reported that the SICTRA induced the largest and the deepest tissue lesions, which was predominantly caused by direct resistive heating, which occurred in the deeper tissue layers. The lesion was not created primarily by heat conduction from the tissue surface to the deeper layers.

The size of the SICTRA-induced tissue lesion is deter-mined by the a m o u n t of delivered energy through the cooled-tip radiofrequency catheter. The power (Watts), the saline irrigation speed (milliliters per minute), the electrode diameter, and the application delivery time are the main factors, which determine the total amount of the delivered radiofrequency energy. The higher the power, the more energy per second will be given. However, this will lead to a higher tissue surface t e m p e r a t u r e upstroke with a subsequent i m p e d a n c e rise, which is associated with a higher risk of tissue carbonization (charring). Once the tissue surface is carbonized, no ablation of the un-derlying tissue layers is possible because of the ex-tremely high impedance of the carbonized tissue surface. Therefore carbonization, in our opinion, should always be avoided. However, the lower the power, the longer the ablation time will be. Initially we used 20 W with a saline irrigation speed of 220 mL/h in the first 4 patients. The time to create an intraatrial lesion length of 3 cm was estimated to be about 60 to 100 seconds. Therefore, we increased the power to 25 W, which reduced the appli-cation time to approximately 45 to 75 seconds to create a similar tissue lesion. Ultimately we increased the power to 32 W. However, we noticed a higher risk of carboniza-tion and were therefore forced to increase our saline irrigation speed from 220 to 250 mL/h. The risk of carbonization formation was reduced, but not enough, in our opinion. Ultimately the irrigation speed was further increased to 320 mL/h. At the same time, however, the risk of "tissue-popping" increased as well. Tissue

pop-CTT SUPPLEMENT KHARGI ET AL S 1 0 9 3 COOLED-TIP RADIOFREQUENCY MAZE PROCEDURE

ping is a s u d d e n release of steam, which is formed in the d e e p e r tissue layers. This steam finds its way out of the tissue in an explosive manner, causing tissue cracks or even a complete tissue breakdown. We speculated that increasing the irrigation speed would lead to a higher formation of resistive heat in the deeper tissue layer, which in turn would increase the size of the induced lesion. However, the risk of tissue " p o p p i n g " with rup-ture would be higher. We ultimately used a 32-W power supply with a saline irrigation speed of 320 mL/h, which proved to be satisfactory, because a smooth linear lesion could be created without causing any charring or "tissue popping." The application time was about 10 to 25 seconds to create a 3-cm linear lesion. We used the Sprinklr catheter, with a 2.33-mm diameter, throughout the entire study. During intraoperative handling, the catheter was moved slowly u p and down over the atrial endocardium until a whitish blistering of the superficial endocardia] cell layer was observed. This whitish blister-ing reflected the acute swellblister-ing of the atrial myocyts, which occurred because of the resistive heating, which was formed in the d e e p e r atrial tissue layers. Once this whitish blistering was visible, the radiofrequency appli-cation was stopped and the catheter was advanced to another area.

Rationale of the Lesion Pattern Design as Used in This Study

Figures 1, 2, and 3 shows the lesion pattern, which was used in this study- The aim was to block the multiple wavelet r e e n t r a n t circuits to extinguish atrial fibrillation, but still preserving the sinoatrial and AV conduction pathway [1]. But we also tried to target and exclude the reinitiating pulmonary trigger zones to prevent the re-currence of atrial fibrillation. Haissaguerre and col-leagues [51 reported the occurrence of ectopic beats, located in the pulmonary veins and around their orifices, especially in the superior left and right pulmonarv veins. These foci trigger atrial fibrillation. The most significant changes between our SICTRA lesion pattern and the Maze III blueprint is the way in which the pulmonary vein orifices were isolated. Whereas, the Maze III proce-dure isolates all four pulmonary vein orifices as an entire one-piece-tissue island, our lesion pattern isolated each pulmonary vein orifices separately, but over its complete orifice circumference. We did not ablate within the pul-monary vein orifices to avoid any risk of pulpul-monary vein stenosis. The superior and inferior orifices on either side were then interconnected. These two lesions patterns, one on the left side and the other on the right, were again interconnected with an additional SICTRA lesion. Ulti-mately a figure of " H " was created. Whether this varia-tion will have any significant effect on the postoperative results remains unclear.

Spontaneous Conversion From Atrial Fibrillation to Sinus Rhythm

GROUP A. Between the 12th postoperative day and the 6th postoperative month, 3 patients showed a spontaneous conversion from atrial fibrillation to SR. A potential

(7)

CHAPTER 4

S I 094 CTT SUPPLEMENT KHARGI ET AL

COOLED-TIP RADIOFREQUENCY MAZE PROCEDURE

explanation is provided by Cox and colleagues [1], who described that the refractory time of the atria during the postoperative period will be temporarily shortened b e -cause of the surgical trauma with edema and inflamma-tion and the elevated circulating catecholamines. There-fore, multiple but smaller reentrant circuits can occur and sustain. An additional explanation is provided by Pasic and colleagues [6], who observed a recovery of the sinus node function and atrial contraction for 1 year, which they attributed to a reinnervation of the perioper-ative, d a m a g e d autonomic nervous system. This finding corroborates our impression that the ultimate success rate should b e evaluated after at least 1 year.

GROUP B. A spontaneous conversion occurred in 4 of 15 patients (26.6%). This finding is in accordance with the reported spontaneous conversion rate as reported by Obadia and colleagues [7], who analyzed 191 mitral valve repair patients and found a spontaneous conversion rate of 36% (5 of 14) in patients with chronic atrial fibrillation, preexisting longer than 1 year. Our patients had a pre-operative duration of atrial fibrillation for at least 1 year (mean = 3.6 years). The spontaneous conversion rate of our study group was expected to be between 4.5% and 35.7%, according to the data of Obadia. W h e r e a s the group A patients converted between the 12th postoper-ative day and 6th postoperpostoper-ative month, the group B patients converted almost immediately after operation, within the 12th postoperative day. This difference suggested that the beneficially changed postoperative h e m o -dynamics contributed to the spontaneous conversion in these patients, whereas the reinnervation of the autono-m o u s nervous systeautono-m and the change of the refractory time probably played a key role in the spontaneous conversion in the group A patients.

Postoperative Morbidity

Three group A patients had a superficial w o u n d infec-tion, which was successfully treated without any invasive surgical treatment. The prolonged operation time was a risk factor. Sternal instability, necessitating surgical re-fixation, occurred in 1 patient from each group.

Follow-Up

O n e group A patient died from mediastinitis on the 45th postoperative day. The prolonged operative time in this obese, diabetic patient was a clear risk factor. Another group A patient had a fatal Coumadin-related renal bleeding. In our opinion, this prosthesis-associated com-plication is not related to the SICTRA p r o c e d u r e itself. O n e group A patient experienced sudden cardiac death at home, possibly due to the proarrhythmic effects of sotalol, although a procedure-related a d v e r s e event could not be excluded. Our postoperative prescription of sotalol was changed from 6 months to 1 month and then replaced by metoprolol. One group A and one group B patient died after 7 and 9 months, both as a result of pulmonary complications related to their preexisting chronic obstructive pulmonary disease. In both patients, the dose of 0-blocker was, from the immediate

postop-42

Ann Thorac Surg 20GT;72:S1090-5

erative time on, reduced or even deleted. One group A patient received a DDD pacemaker because of a sinus bradycardia, although the beta blockade was omitted. O n e group B patient had a ventricle-ventricle-inhibition mode pacemaker because of a bradyarrhythmia.

Treatment Failures and Study Limits

At 12 months 2 patients remained in atrial fibrillation, indicating that the SICTRA was ineffective in these 2 patients. We speculated that a nontransmural functional lesion was created in these 2 patients, which potentially reflected the technical inadequacies of the SICTRA sys-tem or the surgical technique. The energy delivery and irrigation speed in these 2 patients were 20 W and 220 mL/h and 25 and 250 mL/h, respectively. The lack of a postoperative mapping was a drawback in these patients. The difference in age (64.7 versus 69.7 years; p — 0.05) and sex distribution (female to male ratio: 9:6 versus 12:3) between both groups might have affected the results as well. Finally, the small group of patients inevitably en-compasses potential statistical errors, which can b e avoided only if larger groups of patients are studied.

Comparison With the International Literature

The "cut and sew" technique is the golden standard to create linear transmural lesions. The unsurpassed success of 99%, reported by Cox and colleagues in 346 patients, is unique [1]. Melo and colleagues [8] used the Cerablate (Sulzer Osypka, Grenzach-Wyhlen, Germany) radiofre-quency ablation catheter intraoperatively in 43 patients undergoing mitral valve operations who had chronic atrial fibrillation with a mean duration of 6 ± 5 years. This catheter contained four electrodes, which were firmly at-tached to the atrial endocardium. Two oval lesions around the superior-inferior pulmonary vein orifices on either side were created. Before starting the ablation, Melo and col-leagues infused cold saline into and outside the atrium, apparently to acquire a certain level of cooling and to improve the tissue-electrode contact. The left atrial append-age was closed from the inside. The 3-month follow-up was complete for 33 patients; 10 (30%) patients had SR with atrial contraction, 12 patients remained in atrial fibrillation (36%), and 11 had various types of supraventricular rhythms. Compared with our data, the incidence of post-operative SR in Melo's study was lower. The absence of a continuous stable irrigation saline flow to create a deep atrial wall lesion is, in our opinion, a potential explanation for the lower postoperative SR rate. In addition, the varia-tion in atrial wall thickness is not taken in consideravaria-tion when the Cerablate catheter is applied for a fixed time on every part of the atrial wall. Moreover, the simple surgical closure of the left atrial appendage, does not, in our opin-ion, automatically lead to an electrophysiological isolation of the left appendage.

Melo's group, thereafter, also used a different temper-ature-controlled, radiofrequency ablation catheter, Ther-malin-Ep technologies, to produce similar lesions in a group of 46 patients with chronic atrial fibrillation. At the 6 month follow-up of the 25 patients, 13 patients re-mained in atrial fibrillation, 8 had SR with atrial

(8)

contrac-Ann Thorac Surg CTT SUPPLEMENT KHARCI ET AL S1095 2001;72:S1090-5 COOLED-TIP RADIOFREQUENCY MAZE PROCEDURE

tion, and 4 patients had an undefined supraventricular rhythm.

Benussi and colleagues 19] used epicardial radiofre-quency ablation in conjunction with mitral valve opera-tions in 40 patients with chronic atrial fibrillation with a mean duration of 43 ± 51.9 months. The ablation was performed with a temperature-controlled, multipolar ra-diofrequency catheter to produce, from the epicardial side, two encircling lesions a r o u n d the orifices of the right and left pulmonary veins. Then the left atrium was opened and these two epicardial lesions were intercon-nected with an endocardial ablation line. The mitral valve procedure was then performed and the left a p p e n d a g e was sutured. At a mean follow-up of 11.6 months, 77% (30 of 39 patients) were in SR with atrial contraction. In contrary to the dry ablation, the temperature-controlled radiofrequency ablation can create deeper tissue lesion, which will certainly contribute to a higher success rate. The intraoperative epicardial handling of the radiofre-quency catheter to secure a constant a n d firm tissue-electrode surface contact on a beating heart can be technically difficult, however, especially if a substantial a m o u n t of epicardial fat is present.

We thank Hauw Sie, MD, cardiothoraeic surgeon at the Weezen-landen Hospital Zwolle, the Netherlands, who helped us and taught us the operation technique.

References

1. Cox JL, Jaquiss RDB, Schuessler RB, Boineau JP. Modification of the Maze procedure for atrial flutter and atrial fibrillation. Surgical technique of the Maze III procedure. J Thorac Car-diovasc Surg 1995,110:485-95.

2. Wittkampf FH, Hauer RN, Robles de Medina EO. Radiofre-quency ablation with a cooled porous electrode catheter [Abstract]. J Am Coll Cardiol 1988;11:17.

3. Sie HT, Ramdat Misier AR, Beukema WP. Radiofrequency ablation of atrial fibrillation in patients undergoing mitral valve surgery: first experience. Circulation 1996;94:1-675. 4. Nakagawa H, Yamanashi WS, Pitha JV, et al. Comparison of

in vivo tissue temperature profile and lesion geometry for radiofrequency ablation with a saline-irrigated electrode ver-sus temperature control in a canine thigh muscle preparation. Circulation 1995;91:2264-73.

5. Haissaguerre M, Jais P, Shah DC, etal. Spontaneous initiation of atrial fibrillation bv ectopic beats originating in the pulmo-nary veins. N Engl J Med 1998;339:659-66.

6. Pasic M, Musci M, Siniawski H, et al. The Cox Maze II procedure: parallel normalization of the sinus node dysfunc-tion, improvement of atrial function and recovery of the cardiac autonomic nervous system. J Thorac Cardiovasc Surg 1999;118:287-96.

7. Obadia JF, el Farra M, Bastien OH, Lievre M, Martelloni Y, Chassignolle JF. Outcome of atrial fibrillation after mitral valve repair. J Thorac Cardiovasc Surg 1997;114:179-85. 8. Melo J, Adragao P, Neves J, et al. Surgery for atrial fibrillation

using radiofrequencv catheter ablation: assessment of results at one year. Eur J Cardiothorac Surg 1999;15:851-5. 9. Benussi S, Pappone C, Nascimbene S, et al. A simple way to

treat chronic atrial fibrillation during mitral valve surgery: the epicardial radiofrequency approach. European J Cardiotho-racic Surg 2000;17:524-9.