biophysics of ablation: : application to technology

S2

Biophysics of Ablation: Application to TechnologyDAVID HAINES, M.D.

From the William Beaumont Hospital, Royal Oak, Michigan, USA

Biophysics of Ablation. Introduction: The question of what happens to tissue during radiofrequency(RF) catheter ablation continues to be asked as we evolve into the use of newer delivery systems.

Methods and Results: Three assumptions are made about RF ablation. (1) Tissue injury is thermallymediated; (2) heat transfer in tissue should be a predictable biophysical phenomenon; and (3) large lesiontechnologies have more or less equivalent efficacies. Based on these assumptions, predictions are made anddiscussed. Many of the predictors were proven to be true while some surprisingly were not.

Conclusion: In conclusion, tissue-area injury occurs reproducibly at a temperature of about 50◦C. Heattransfer in tissue is a predictable phenomenon. And finally, new technologies for large lesions are all effective,but greater surface area of ablation was achieved with a 10-mm tip and greater depth was achieved with aChilli® cooled ablation catheter. (J Cardiovasc Electrophysiol, Vol. 15, S2-S11, Suppl. 1, October 2004)

catheter ablation, arrhythmias, biophysics, thermodynamics, ablation biophysics, radiofrequency ablation

Introduction

The question of what happens to tissue during radiofre-quency (RF) catheter ablation continues to be asked as weevolve into the use of newer delivery systems. This articlewill begin by stating some general assumptions about RF

Address for correspondence: David Haines, M.D., Director, Heart RhythmCenter, William Beaumont Hospital, 3601 West 13 Mile Rd., Royal Oak,MI 48073; E-mail: [email protected]

Manuscript received 9 June 2004; Accepted for publication 12 July 2004.

doi: 10.1046/j.1540-8167.2004.15102.x

Figure 1. Resting membrane depolarization with hyperthermic exposure. (With permission from reference 2.)

ablation, followed by the predictions that flow from theseassumptions and their research realities.

General Assumptions About RF Ablation

The first assumption we will make is that tissue injuryis thermally mediated. This was supported by the workof Simmers et al.1 in 1996. The second assumption isthat heat transfer in tissue should be a predictable bio-physical phenomenon. And finally, that large lesion tech-nologies have more or less equivalent efficacies. Basedon these assumptions, the following predictions may bemade.

Haines Biophysics of Ablation S3

Figure 2. Excitability during hyperthermic exposure. (With permission from reference 2.)

Predictions of Tissue Response to RF Ablation

Prediction 1: The Exposure of the Myocardium to HeatingShould Result in a Reproducible Biological Response

To explore this question, experiments were conducted tolook at the effect of thermal exposure on the myocardium.2

Figure 3. Thermodynamic model of RF catheter ablation. (With permission from reference 5.)

Nath et al. utilized an isolated guinea pig muscle preparation.The relationship between the resting membrane depolariza-tion versus temperature is illustrated in Figure 1. Little tissueinjury occurs at temperatures below 45◦C. In the transitionzone between 45 and 50◦C, depolarization of the cell is stillseen. At above 50◦C, however, there is a fairly reliable injury

S4 Journal of Cardiovascular Electrophysiology Vol. 15, No. 10, Supplement, October 2004

and kill of the tissue. This is supported by the work of Sim-mers et al. as well,1,3 showing dramatic drop in conductionthrough tissue at about 50◦C. Therefore, 45–50◦C is the tran-sition zone. Above 50◦C seems to be the temperature wherea reproducible irreversible injury to the tissue occurs with ashort hyperthermic exposure.

Figure 2 displays the temperature ranges again in theguinea pig preparation2 versus tissue excitability. The thresh-old temperature for irreversible loss of excitability is againshown as 50◦C.

Prediction 2: Heat Transfer in Tissue Should Be aPredictable Biophysical Phenomenon

Figure 3 is a model that was proposed in 19894 look-ing at steady-state thermal transfer in tissue. The modelstates that a steady state should produce a predictable ra-dial temperature gradient. This thermal transfer emanatesfrom the source and then spreads out in a radial fashion. Thetemperature should be expected to drop in an exponentialfashion.

Actual measurements of temperature in tissue and thenin an in vitro setting were performed (Fig. 4).4 At differentsource temperatures, a fairly predictable decrease in the tis-sue temperature was seen with increasing distance from thesource. The dotted line across at 50◦C represents the isothermof irreversible injury. The point where the curve intersectsthat isotherm would ultimately represent the depth of thelesion.

Figure 4. Radial temperature gradients with constant interface temperature. (With permission from reference 4.)

Prediction 3: Lesion Size Should Be Proportional to theElectrode–Tissue Interface Temperature and HigherTemperature at the Source Results in Deeper Heating

Temperature in the controlled in vitro setting was avery firm predictor of the depth of the lesion. Figure 5demonstrates a direct relationship between the depth ofthe lesion and the temperature. When temperature wascompared to other electrical parameters including current,power, and energy, it seemed to be the best predicator(Fig. 6).

This does not, however, translate directly to the in vivosetting because there are opposing forces to just heat trans-fer in tissue (i.e., the cooling effect of regional bloodflow). In a pure system of heat transfer to tissue, how-ever, the temperature at the source should predict the lesionsize.4

Prediction 4: Lesion Size Should Be Proportional to theRadius of the Heat Source

It is intuitive that if one starts out with a large heat source,a larger lesion should be formed. If a pinpoint source of heatis used, less energy transfers to the tissue and the lesion sizemight be anticipated to be smaller. Haines et al.5 made thisobservation in the controlled in vitro setting in 1990. Lesionsize was found to be directly proportional to the diameterof the electrode used. Figure 7 demonstrates the relationshipbetween electrode radius and depth and diameter of resultantlesion.


Figure 5. Effect of electrode–tissue interface temperature on lesion size in vitro. (With permission from reference 4.)

Prediction 5: Temperatures Reaching or Exceeding aCritical Temperature at the Electrode–Tissue InterfaceShould Result in Coagulum Formation with Sudden Risein Electrical Impedance

Threshold temperature as a phenomenon was an area ofgreat interest during the early years of RF ablation therapy.

Figure 6. Relationship between lesion depth and tip temperature, current, power, and energy. (With permission from Reference 4.)

Figure 8 demonstrates the relationship between temperatureand impedance. It can be seen that the sudden impedancerise (shown in green) corresponds to a peak temperature of100◦C.

Haines and Verow6 used a canine model to obtain in vitroand in vivo data on the relationship between the tempera-ture at electrode–tissue interface and the resultant effect on


Figure 7. Lession size vs. electrode radius. (With permission from reference 5.)

impedance (Fig. 9). An impedance rise occurred fairly uni-formly at the threshold of 100◦C.

Prediction 6: Greater Dissipation of the RF Energy Alongthe Transmission Line May Reduce the Lesion Size

As electricity flows through a circuit, every point of thatcircuit represents a drop in voltage. The point of greatest drop

Figure 8. Sudden impedance rise during RF ablation.

in line voltage represents the area of highest impedance, and iswhere most of that electrical energy gets dissipated as heat.Therefore, with excessive electrical resistance in the trans-mission line, the line will actually warm up and power willbe lost in the transmission line. Current laboratory equipmentuses good electrical conductors from the generator all the waythrough to the patient and from the dispersive electrode backto the generator thus minimizing this power loss. However,


Figure 9. Correlation of sudden rise in electrical impedance with temperature monitoring. (With permission from reference 6.)

dissipation of energy that may limit lesion formation maystill occur at the dispersive electrode site. There is a voltagedrop, a dissipation of energy at that contact point betweenthat electrode and the skin. The issues of dispersive electrodelocation, size, number, and so on have been discussed andexplored over the years.

Figure 10. Effects of increased dispersive electrode surface area. (With permission from reference 7.)

Nath et al.7 conducted a clinical study that looked at theeffect of dispersive electrode position and surface area on cur-rent, voltage, baseline impedance, and catheter tip tempera-ture during constant power RF delivery. They found that thelocation of the dispersive electrode seemed to have no effecton impedance, voltage, current delivery, and tip temperature.


Figure 11. Effects of convective cooling on radial temperature gradient.

In addition, doubling of the dispersive electrode surface arealed to lower impedance, higher current delivery, and increasedtip temperatures. This was especially true when patients hada baseline impedance greater than 100 �. These findingsare particularly relevant when the system is power limited,

Figure 12. Effect of tip irrigation on tissue temperature. (With permission from reference 8.)

as with a 50 W generator. Therefore, when ablating certainsites, the addition of a second dispersive electrode or op-timizing the preparation between the dispersive electrodeand skin should result in relatively more power deliveryto the target tissue. Some of these results can be seen inFigure 10.


Prediction 7: Dissipation of RF Energy into theCirculating Blood Pool (from Poor Contact or SlidingContact) Should Result in a Smaller Lesion Size

As the magnitude of convective cooling increases, thereis decreased efficiency of heating due to more energy beingcarried away in the blood and less energy delivered to thetissue. A major site of tissue cooling is on the endocardialsurface. Excess heating at the surface is decreased by cool-ing of the catheter tip. This allows the power to be turned upwithout the risk of coagulum and char formation. Increasedpower delivery means that heating occurs deeper in the tis-sue, making for greater depth of direct volume (resistive)heating into the myocardial target tissue, thus yielding largerlesions.

The effects of convective cooling on radial temperaturegradient can be seen in Figure 11. These curves representthe tissue temperature versus the distance from the electrode.The typical non-cooled ablation with a 70◦C electrode–tissueinterface temperature is shown in curve A. This curve in-tersects the 50◦C isotherm and produces a lesion of about3.7 mm depth. If the power is turned up and surface coolingis increased, the results are represented by curve B. If theelectrode is not cooled, the peak temperature would be offthe scale and an impedance rise would occur. However, sincecooling is occurring on the surface, turning up the powerdrives the heating deeper into the tissue and shifts the curvecompletely to the right, thus causing a dramatic increase inlesion size.

What happens, however, if just the surface is cooled andpower is not increased? This results in the wastage of energy,as it will dissipate into the circulating cooling system or intothe blood pool. This is seen in curve C, and may actuallyresult in a smaller lesion.

Figure 13. Effect of coronary blood flow on lesion formation. (With permission from reference 9.)

This phenomenon was described very nicely by Nakagawaet al.8 in their classic publication. Figure 12 shows that withsurface cooling, temperatures over 100◦C can be achievedwith lesions 3.5 mm deep.

There are important opposing effects of RF tissue heat-ing associated with concomitant convective cooling. Surfacecooling does reduce the risk of boiling and coagulum forma-tion; however it does not allow one to monitor the temperatureat the tip, and thus one loses some feedback about lesion for-mation. Higher power can be used with convective cooling,but higher power can cause superheating in the subendocar-dial layers and cause “pop lesions.” Higher power results ingreater depth of volume heating, but if the ablation is powerlimited, power dissipation into the circulating blood pool canactually result in decreased lesion depth. Therefore, there arepositive and negative features associated with cooled tip ab-lation.

Another aspect to consider with convective cooling is theeffect of coronary perfusion. There are few coronary compli-cations with conventional RF ablation, because the coronaryarteries act as a heat sink. It is very hard to heat the coro-naries to the point where the vessel is damaged because asignificant quantum of heat is carried away by the high ve-locity blood flow. Fuller and Wood published a report in Cir-culation9 demonstrating the phenomenon. A small marginalartery in a tissue preparation was perfused at various rates.The tissue around the perfused artery was actually protectedagainst ablation (Fig. 13). The two graphs on the right of thefigure show that the volume of preserved myocardium wasrelated to the perfusion rate as well as the arterial diameter.This is an advantage as that coronary arteries are being pro-tected, but it is also a potential problem if there is a largeperforating artery going through an area of tissue that is theablative target.


Figure 14. Five thermistor tip RF electrode. Data obtained from a five thermistor tip RF electrode showing the variability in temperatures recorded aroundthe ablation electrode during ablation. (With permission from reference 10.)

Prediction 8. Tissue Heating Contiguous to LargeElectrodes with Variable Tissue Contact Will BeNonuniform

Finite element analysis of ablation from a 12-mm tip elec-trode can be done. It may show hot spots on the edges, andthe middle of the electrode appears cooler. This can result innonuniformity of the lesion, particularly if the contact of theelectrode with tissue is not stable.10

One can use multiple temperature sensors to assess tem-perature throughout lesion creation. Figure 14 shows dataobtained from a five thermister tip RF electrode showing thevariability in temperatures recorded around the ablation elec-trode during ablation.11

Comparison of Electrode Technologies

To summarize, what is the best RF technology to cre-ate large lesions? With the big tips there is the problem ofnon-uniform heating. However, this may be improved with amultisensor electrode. There is also difficulty in monitoringlesion formation. It is also difficult to monitor lesion forma-tion with cool tips.

The author conducted a study that examined differentcatheter design efficacy in a canine preparation. The 10-mmtip electrode, the 10-mm multisense electrode that has ther-misters or thermocouples both at the tip of the electrode aswell as at the base to improve control, and the Chilli (BostonScientific, Inc., San Jose, CA, USA) catheter were studied.One hundred watts and 70◦C temperature feedback powercontrol were used for the 10-mm tip electrodes. With abla-tion using the Chilli catheter (Boston Scientific, Inc.), thepower was titrated up with continuous impedance monitor-ing. The 10-mm multisense electrode produced the small-est lesions overall. The 10-mm tip yielded intermediate le-

sion depths, and the Chilli catheter (Boston Scientific, Inc.)gave the greatest depth of lesion formation. However, be-cause there was more contact surface area with a 10-mm tipelectrode catheter, the lesion surface area was greatest withthis design.

Conclusion

In conclusion, tissue-area injury occurs reproducibly ata temperature of about 50◦C. Heat transfer in tissues is apredictable biophysical phenomenon. And finally, new tech-nologies for large lesions are all effective, but greater surfacearea of ablation was achieved with a 10-mm tip and greaterdepth was achieved with a Chilli catheter (Boston Scientific,Inc.).

References

1. Simmers TA, DeBakker JMT, Wittkampf FHM, Hauer RNW: Effects ofheating with radiofrequency power on myocardial impulse conduction:Is radiofrequency ablation exclusively thermally mediated? J CardiovascElectrophysiol 1996;7:243-247.

2. Nath S, Lynch C, Whayne JG, Haines DE: Cellular electrophysiologicaleffects of hyperthermia in isolated guinea pig papillary muscle: Impli-cations for catheter ablation. Circulation 1993;88(4 Pt1):1826-1831.

3. Simmers AA, DeBakker JMT, Wittkampf FHM, Hauer RN: Effects ofheating on impulse propagation in superfused canine myocadium. AmJ Coll Cardiol 1995;25:1457-1464.

4. Haines DE, Watson DD: Tissue heating during radiofrequency catheterablation: A thermodynamic model and observations in isolated perfusedand superfused canine right ventricular free wall. Pacing Clin Electro-physiol 1989;12:962-976.

5. Haines DE, Watson DD, Verow AF: Electrode radius predicts lesionradius during radiofrequency energy heating. Validation of a proposedthermodynamic model. Circ Res 1990;67(1):124-129.

6. Haines DE, Verow AF: Observations on electrode-tissue interface tem-perature and effect on electrical impedance during radiofrequency ab-lation of ventricular myocardium. Circulation 1990;82:1034-1038.


7. Nath S, DiMarco JP, Gallop RG, McRury ID, Haines DE: Effects ofdispersive electrode position and surface area on electrical parametersand temperature during radiofrequency catheter ablation. J Am CollCardiol 1996;77:765-767.

8. Nakagawa H, Yamanashi WS, Pitha JV, Arruda M, Wang X, Ohtomo K,Beckman KJ, McClelland JH, Lazzara R, Jackman WM: Comparison ofin vivo tissue temperature profile and lesion geometry for radiofrequencyablation with a saline-irrigated electrode versus temperature control ina canine thigh muscle preparation. Circulation 1995;91:2264-2273.

9. Fuller IA, Wood MA: Intramural coronary vasculature prevents trans-mural radiofrequency lesion formation: Implications for linear ablation.Circulation 2003;107:1797-1803.

10. McRury ID, Mitchell MA, Panescu D, Haines DE: Non-uniform heatingduring radiofrequency ablation with long electrodes: Monitoring theedge effect. Circulation 1997;96:4057-4064.

11. McRury ID, Mitchell MA, Haines DE: Efficacy of multiple ring andcoil electrode radiofrequency ablation catheters for the creation of longlinear lesions in the atria. Med Engineer Phys 1998;20:551-557.

biophysics of ablation: : application to technology

Documents