Why a Large Tip Electrode Makes a Deeper Radiofrequency Lesion

47

Why a Large Tip Electrode Makes a Deeper Radiofrequency Lesion: Effects of Increase in Electrode Cooling and Electrode-Tissue Interface Area KENICHIRO OTOMO. M.D.. WILLIAM S. YAMANASHI, PH.D., CLAUDIO TONDO. M.D., MATTHIAS ANTZ, M.D.. JONATHAN BUSSEY, B.S. JAN V. PITHA, M.D., PH.D., MAURICIO ARRUDA, M.D., HIROSHI NAKAGAWA, M.D., PH.D., FRED H.M. WITTKAMPE, PH.D.,* RALPH LAZZARA, M.D., and WARREN M. JACKMAN, M.D. From the Cardiovascular Section. Depatttiicnt of Medicitie. University of Oklahoma Health Sciences Center and the Department of Veterans Affairs Medical Center, Oklahoma City, Oklahoma; and the *Department of Cardiology, Heart Lung Institute, University Hospital Utrecht. Utrecht. The Netherlands

Increase in RF Lesion Depth with Larger Electrode. Introduction: Increasing electrode size allows an increase in radiofrequency lesion depth. The purpose of this study was to examine the roles of added electrode cooling and electrode-tissue interface area in producing deeper lesions. Methods and Results: In 10 dogs, the thigh muscle was exposed and superfused with heparinized blood. An 8-French catheter with 4- or 8-mm tip electrode was positioned against the muscle with a hlood flow of 350 mL/min directed around the electrode. Radiofrequency current was delivered using four methods: (1) electrode perpendicular to the muscle, using variable voltage to maintain the electrode-tissue interface temperature at 60°C; (2) .same except the surrounding hlood was stationary; (3) perpendicular electrode position, maintaining tissue temperature (3.5-mm depth) at 90°C; and (4) electrode parallel to the muscle, maintaining tissue temperature at 90°C. Electrode-tissue interface temperature, tissue temperature (3.5- and 7.0-mm depths), and lesion size were compared between the 4- and 8-mm electrodes in each method. In Methods 1 and 2, the tissue temperatures and lesion depth were greater with the 8-mm electrode. These differences were smaller without hlood flow, suggesting the improved convective cooling of the larger electrode resulted in greater power delivered to the tissue at the same electrode-tissue interface temperature. In Method 3 (same tissue current density), the electrode-tissue interface temperature was signiflcantly lower with the 8-mm electrode. With parallel orientation and same tissue temperature at 3.5-mm depth (Method 4), the tissue temperature at 7.0mm depth and lesion depth were greater with the 8-mm electrode, suggesting increased conductive heating due to larger volume of resistive heating heeause of the larger electrode-tissue interface area. Conclusion: With a larger electrode, hoth increased cooling and increased electrode-tissne interface area increase volume of resistive heating and lesion depth. (J Cardiovasc Electrophysiol, Vol. 9. pp. 47-54. January 199S) catheter ablation, radiofrequency, electrode size, lesion size, electrode cooling, electrode-tissue

interface area This research was supported in part by Grant ROl-HL-39670 from the National Institutes of Health and Grant HRI-iO4 from the Oklahoma Center for the Advancement of Science and Technology. Address for correspondence: Warreti M, Jackman. M.D,. Cardiovascular Section. Department of Medicine. University of Oklahoma Health Sciences Center. 920 S.L. Young Blvd. (WP3120), Oklahoma City, OK 73104. Fax: 405-271 -2619. Manuscript received 5 June 1997; Accepted for puhlication 20 October 1997.

Introduction In radiofrequency catheter ablation, radiofrequency current is delivered to the tissue to produce an area of necrosis. In the myocardium close to the electrode, the current density is sufficiently high to produce direct resistive heating (volume of resistive heating). Heat from this region is conducted to the adjacent myocardium, which pro-

48

Journal of Cardiovascular Electrophysiology

Vol. 9. No. I. January 1998

duces a thermal gradient radiating away from tbe volume of resistive beating.' All of the tissue reaching 47° to 49''C becomes necrotic.' - Lesion depth can be increased by increasing the power delivered to tbe tissue, increasing tbe volume of resistive heating. However, tissue power delivery is limited by suiface heating at the electrode-tissue interface. The electrode-tissue interface temperature must remain < IOO°C to prevent tissue desiccation and coagulum formation associated with an impedance rise.^"* Etirly in the development of radiofi^uency catheter ablation procedures, it was necessary to increase the ablation electrode size from a 6French, 2-mm length to a 7-Frencb, 4-mm length in order to produce adequate lesion size and depth for clinical efficacy.^ A further increase in ablation electrode size bas allowed the application of higher radiofrequency power, resulting in a further increase in lesion size and depth.^** Two tnechanisms have been proposed to explain the deeper radiofrequency lesions prodticed by larger ablation electrodes. The first explanation is an increase in convective cooling of the abiation electrode by the larger surface of the elecd'ode exposed to the blood flow,'"' wbicb maititains a lower electrode-tissue interface temperature allowing greater power to be delivered to the tissue (at the same electrode-tissue interface temperature) witb the result of higher tissue current density and deeper direct resistive heating. The second explanation relates to the increase in electrode-tissue interface ar^a.'-^'* Both effects sbould increase the volume of resistive heating and increase the lesion depth. The purpose of the present study was to validate the importance of each of these two mechanistiLS in ptxxJiicing a deeper lesion by a large ablation electrode. Using a canine thigh muscle preparation,'" we compared tbe tissue temperature at various depths and lesion depth between an 8-French, 4-mm tip electrode and an 8-French. 8-mm tip electrode oriented perpendicular and piirallel to the tissue surface and with different modes of radiofrequency energy delivery.

tained with supplemental doses of sodium pentobarbital. The right carotid artery and jugular vein were cannulated for continuous monitoring of arterial pressure and the administiation of drugs, respectively. The thigh muscle preparation was utilized in tbe same manner as previously described.'" After the dog was placed on its side, a 15-cm skin incision was made over the thigh muscle. Tbe skin, overlaying connective tissue, and thin superficial muscle were gently dissected, exposing the surface of the thicker underlying muscle. The fascia on the surface of the muscle was usually thin and transparent, but any thicker fascia was removed. The edges of the skin were raised to form a cradle that was filled with hepariinized canine blood from the same dog and maintained at 36" to 37°C (Fig. 1). An 8-French electrode catheter containing a 4- or 8-mm tip electrode with a tbermistor at the tip of the electrtxle but thennally insulated from tbe surrounding electrode (EP Tecbnologies. Mountain View. CA, USA) was positioned against the thigh muscle with a constant weigbt of 10 g using a custom balance. Blood was directed around the ablation electrode at a flow rate of 350 mL/min (Fig. I). Tbe 4- or 8-mm tip electrode was positioned perpendicular or paiallel to the thigh muscle in different expeiiments. Two fluoroptic thermal sensor probes (model 3000 [Luxtron, Santa Clara, CA, USA|; measut^ment range, 0° to 125°C; acctiracy

Skin Cradle

Heparinized Blood Reservoir {36-37*C)

Methods Experimental Preparations

The experimental protocol was approved by the University of Oklahoma Committee on Use and Cate of Animals. Ten mongrel dogs weighing 15 to 20 kg were anesthetized with 25 mg/kg sodium pentobarbital, intubated. and mechanically ventilated with room air. General anesthesia was main-

Figure 1. Schematic representation of the canine thigh mu.scle preparation.'" The skin, connective ti.ssue, and thin superficial muscle were dissected to expose the smooth, glistening .surface of the thick thigh muscle. The edges of the skin were raised to form a cradle, which was filled with heparinized canine blood at 36° to 37°C. An 8-French electrode catheter with ci 4- or H-mm tip electrode was positioned again.st the thigh muscle with a constant weight of 10 g. Blood was directed around the ablation electrode at a flow rate of 350 ml/min.

Otomo, et al. Increase in RF Lesion Depth with Larger Electrode

± 0.2°C) were bundled together with shrink tubing. One sensor tip extended 3.5 mm from the end of the shiink tubing and the other .sensor tip extended 7.0 mm. The sensor probes were inserted into the muscle directly adjacent to the ablation electrode up to the shrink tubing, placing one sen.sor 3.5 mm and the other 7.0 mm below the surface. The electrodetissue interface temperature was measured using the tip thermistor when the electrode was oriented peipendicular to the thigh muscle. When the electrode was placed parallel to the thigh muscle, an additional fluoroptic thennal sensor probe was placed between the middle of the electrode and the tissue to measure the electrode-tissue interface temperature at the center of the interface. Radiofrequency current (550 to 650 kHz) was produced hy a constant voltage generator (mode! LIZ-88, American Cardiac Ahlation Co.. Foxboro, MA. USA) and delivered between the catheter tip electrode and an adhesive electrosurgical dispersive patch applied to the shaved skin of the abdominal wall. During each application of radiofrequency current, the root mean square voltage, current, impedance, and the temperatures measured from the thermistor at the tip of the ablation electrode and the tissue probes were continuously monitored and recorded on optical disk (Baid LahSystem. Bard Electrophysiology, Billerica, MA, USA). Five to ten applications of radiofrequency current were delivered to separate sites on the thigh muscle. The skin incision was closed, the dog was turned onto its other side, and 5 to 10 applications of radiofrequency current were delivered at separate sites on the other thigh muscle. Ablation Protocols

The 4- and 8-mm tip electrodes were compared using four methods of radiofrequency current deMethod 1 4mm

amm

Method 2 4mm

8mm

49

livery (Fig. 2). Methods 1 through 3 were designed to detennine the effects of the difference in convective cooling of the ablation electrode on lesion depth. The 4- and 8-iTim tip electrodes were positioned perpendicular to the thigh muscle with the same electrode-tissue interface area. TTie 8-mm tip electrode had a larger electrode-hlood interface area for greater convective cooling by the blood. In Method I. radiofrequency current was delivered for 60 seconds using variable voltage (30 to 60 V) to maintain the electrode-tissue interface temperature at 6O''C. Method 2 was identical to Method 1, except that there was no pumping of blood around the electrode (i.e.. no blood flow). In Method 3. radiofrequency current was delivered for 60 seconds using variahle voltage (30 lo 60 V) to maintain a con.stant temperature of 90°C measured in the tis.stw at a depth of 3.5 mm, indicative of the same degree of the tissue heating. Method 4 was designed to determine the effects of an increase in electrode-tissue interface ai-ea on lesion depth. The 4- and 8-mm tip electrodes were positioned parallel to the thigh muscle. In the parallel orientation, the ratio of electrode surface area in contact with blood and tissue is essentially the same for the two electrodes. Therefore, the degree of electrode cooling of the electrode-tissue interface should be similar. RF current was delivered for 60 seconds using variable voltage (30 to 60 V) to maintain a constant temperature of 90°C measured in the tissue at a depth of 3.5 mm. After Ablation Two hours after the ahlation procedure was completed, 30 mL of 2% triphenyl tetrazolium chloride (TTC) was administered intravenously. TTC is a vital stain that forms a briek red precipitate in the presence of active intraeellular dehydroMethod 3 4mm

8mm

Method 4 4mm 3.5mm

i

90*C

60*C

^eo'C'^

8mm 3.5mm 901:;

Figure 2. Schematic representation of the four methods. Arrows in Methods i, 3, and 4 represent the pumped blood flow directed around the abtatton electrode, and the arrows in Mettwd 2 represent convective blood flow associated with heating of the electrode-tissue inteiface and the ablation electrode. See text for details.

50

Journal of Cardiovascular Electrophysiology

Vol. 9, No. I, January 1998

genase. staining viable but not necrotic (ablated) tissue. Ventricular fibrillation or asystole was induced by the TTC or potassium chloride injection, and shortly thereafter the thigh muscles were excised and fixed in a 10% formalin solution. Each radiofrequency lesion was sectioned along its longest axis. The maximal depth and maximal diameter of the lesion were measured.

q TT

+1 +1

d d +1 t l

d d tl tl

tl tl

tl tl

tl tl

tl tl

tl tl

U-. so

II

Statistical Analysis Data are expressed as mean ± SD. The electrical parameters of radiofrequency delivery (voltage, current, impedance, and power), temperatures at the electrode-tissue interface as well as at tissue depths of 3.5 and 7.0 mm, and lesion depths and diameters were compared among the four methods by ANOVA. Any significant differences were measured by Scheffe F-test for paii-wise comparison. P < 0.05 was considered statistically significant.

q d d tl tl

q r-^ •?. E

I*-

d d

w

O

tl

tl

tl

tl

—

0^

tl

tl tl

tl tl

tl tl

tl

tl tf

tl tl

tl tl

+1 II +1

in —

1— r*!

r- DO

ti tl

tl tl

tl +1

tl +1

tl tl

+1

tl

Results A total of 133 radiofrequency lesions were produced in the thigh muscles of the 10 dogs. The radiofrequency parameters, temperature values, and lesion dimensions obtained from the four ablation methods are shown in Table 1.

•i-

O S

«

IN r^

+1 II

q d tl —

q d tl o

c c

tl

+1 tl

tl

ri in

d d

Effects of Increase in Electrode Cooling (Methods 1 Through 3) In Method 1, the 4- and 8-mm electrodes were oriented perpendicular to the tissue, and radiofrequency power was varied to maintain the electrode-tissue interface temperature at 60"C. Although the electrode-tissue interface temperature was the same for the 4- and 8-mm electrodes, the radiofrequency applications with the 8-mm electrode had significantly higher tissue temperatures at depths of 3.5 mm (90° ± 9°C vs 74° ± 10°C) and 7.0 mm (63° ± TC vs 54° ± 5°C) and greater lesion depth (7.8 ± 0.8 mm vs 6.6 ± 0.5 mm). Compared to Method 1. the radiofrequency applications in Method 2 (without pumped blood tlow around the electrode) had markedly lower voltage, current, and power (4 ± I W vs 18 ± 5 W for the 4-mm electrode: 20 ± 3 W vs 35 ± 10 W for the 8-mm electrode) at the same electrodetissue interface temperature. As a result, tissue temperatures at depths of 3.5 and 7.0 mm were significantly lower and lesion depths were much smaller in Method 2 compared to Method 1 for the 4- and 8-mm electrodes (Table 1). However,

tl tl

tl tl

tl tl

•o o P •=!

C.0

•a I

3

k! UJ . | S H 5 k!

E

3 E

p 2 t^P

Otomo. et al.

compared to the 4-mm electrode, the 8-mm electrode still was associated with higher tissue temperatures at depths of 3.5 mm (55° ± 6°C vs 49° ± 4°C) and 7.0 mm (42" ± TC vs 39° ± 2°C) and greater lesion depth (3.5 ± 0.6 mm vs 2.5 ± 0.4 mm). Although a pumped blood flow was absent, motion of the bUxKl around the electrode was observed during radiofrequency applications, presumably convective motion due to heating at the electrode-tissue interface. In Method 3, radiofrequency power was varied to maintain a temperature of 90°C at 3.5 mm below the electrode-tissue interface with the 4- and 8-mm electrodes positioned perpendicular to the tissue. This method was designed to produce the same degree of tissue heating. The similar temperature at a depth of 7.0 mm (60° ± 4°C vs 61° ± 4°C) and similar lesion depth (7.4 ± 0.6 mm vs 7.4 ± 0.6 mm) suppoii the similar degree of direct resistive heating. Despite the similar degree of resistive heating, the radiofrequency applications using the 8-min electrode were associated with significandy lower electrode-tissue interface temperature (57° ± 3°C vs 65° ± 6°C). For the similar degree of resistive heating, the 8-mm electrode required significantly higher voltage (53 ± 7 V vs 42 ± 6 V). current (0.61 ± 0.08 A vs 0.46 ± 0.06 A), and power (32 ± 8 W vs 20 ± 7 W). Effects of Increase in Eleetrode-Tissue Interfaee Area (Method 4) In Method 4, the 4- and 8-mm electrodes were positioned parallel to the tissue, and nidiofrequency power was varied to maintain a temperatutis of 90°C at a tissue depth of 3.5 mm. The electrode-tissue interface temperatures for the 4- and 8-mm electrodes were similar (64° ± 11°C vs 63'' ± 9°C). Despite the similar temperatures at the electrode-tissue interface and 3.5-mm depth, the 8-mm electnxle resulted in significantly higher temperature at the 7.0mm depth (69° ± 4°C vs 61° ± 4°C) and greater lesion depth (8.4 ± 0.6 mm vs 7.2 ± 0.6 mm).

Increase in RF Lesion Depth with Larger Electrode

51

Effects of Increase in Electrode Cooling (Methods I Through 3) A radiofrequency lesion is generated by two coexistent heating mechanisms: direct resistive heating, which occurs close to the ablation electrode; and passive conductive heating, which emanates from the tissue heated electrically (Fig. 3).i-i