Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print January 19, 2006, 10.1148/radiol.2382050571
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2382050571v1
238/3/881    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Clasen, S.
Right arrow Articles by Pereira, P. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Clasen, S.
Right arrow Articles by Pereira, P. L.
(Radiology 2006;238:881-890.)
© RSNA, 2006

Experimental Studies

Multipolar Radiofrequency Ablation with Internally Cooled Electrodes: Experimental Study in ex Vivo Bovine Liver with Mathematic Modeling1

Stephan Clasen, MD, Diethard Schmidt, MD, Andreas Boss, MD, PhD, Klaus Dietz, PhD, Stefan M. Kröber, MD, Claus D. Claussen, MD and Philippe L. Pereira, MD

1 From the Department of Diagnostic Radiology, Eberhard-Karls-University, Hoppe-Seyler-Strasse 3, 72076 Tübingen, Germany (S.C., D.S., A.B., C.D.C., P.L.P.); and Department of Medical Biometry (K.D.) and Institute of Pathology (S.M.K.), University of Tübingen, Tübingen, Germany. From the 2004 RSNA Annual Meeting. Received April 6, 2005; revision requested June 3; revision received June 30; final version accepted July 20. Address correspondence to S.C. (e-mail: stephan.clasen{at}med.uni-tuebingen.de).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To evaluate the size and geometry of thermally induced coagulation by using multipolar radiofrequency (RF) ablation and to determine a mathematic model to predict coagulation volume.

Materials and Methods: Multipolar RF ablations (n = 80) were performed in ex vivo bovine livers by using three internally cooled bipolar applicators with two electrodes on the same shaft. Applicators were placed in a triangular array (spacing, 2–5 cm) and were activated in multipolar mode (power output, 75–225 W). The size and geometry of the coagulation zone, together with ablation time, were assessed. Mathematic functions were fitted, and the goodness of fit was assessed by using r2.

Results: Coagulation volume, short-axis diameter, and ablation time were dependent on power output and applicator distance. The maximum zone of coagulation (volume, 324 cm3; short-axis diameter, 8.4 cm; ablation time, 193 min) was induced with a power output of 75 W at an applicator distance of 5 cm. Coagulation volume and ablation time decreased as power output increased. Power outputs of 100–125 W at applicator distances of 2–4 cm led to a reasonable compromise between coagulation volume and ablation time. At 2 cm (100 W), coagulation volume, short-axis diameter, and ablation time were 66 cm3, 4.5 cm, and 19 min, respectively; at 3 cm (100 W), 90 cm3, 5.2 cm, and 22 min, respectively; at 4 cm (100 W), 132 cm3, 6.1 cm, and 27 min, respectively; at 2 cm (125 W), 56 cm3, 4.2 cm, and 9 min, respectively; at 3 cm (125 W), 73 cm3, 4.9 cm, and 12 min, respectively; and at 4 cm (125 W), 103 cm3, 5.5 cm, and 16 min, respectively. At applicator distances of 4 cm (>125 W) and 5 cm (>100 W), the zones of coagulation were not confluent. Coagulation volume (r2 = 0.80) and RF ablation time (r2 = 0.93) were determined by using the mathematic model.

Conclusion: Multipolar RF ablation with three bipolar applicators may produce large volumes of confluent coagulation ex vivo. A compromise is necessary between prolonged RF ablations at lower power outputs, which produce larger volumes of coagulation, and faster RF ablations at higher power outputs, which produce smaller volumes of coagulation.

© RSNA, 2006


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Radiofrequency (RF) ablation is a therapy option for the treatment of primary and secondary liver tumors. Percutaneous RF ablation is a minimally invasive technique that is safe and effective in local tumor therapy (1,2). One challenge in achieving complete RF ablation is to ensure that the thermally induced coagulation zone covers the entire tumor volume, with an appropriate safety margin in order to reduce relapse rate or residual tumor tissue after treatment. The rate of incomplete RF ablations, however, increases with tumor diameter (3,4). Thus, the maximum volume of induced coagulation that is obtained in one therapy session remains a major limitation.

A strategy to achieve complete coagulation in large tumors is to perform overlapping ablations (5). As is shown by computer analyses, however, the size of the composite volume of coagulation is small relative to the number of ablations performed (6). In a geometrically optimized model, a small change in the position of the ablation spheres with respect to the target sphere can leave potentially nonablated tumor tissue (7). Thus, one aim of the improvement in RF techniques is to extend the volume of coagulation that is produced by a single RF ablation. These strategies include the use of internally cooled electrodes (8), perfusion electrodes (9,10), multiprobe arrays (eg, clusters) (11), and multitined expandable electrodes (12), as well as the implementation of modifications in the algorithm of energy deposition (13) and the use of adjunct techniques to reduce perfusion-mediated tissue cooling (14).

The currently available RF devices are monopolar systems that apply an alternating electric current at a high-frequency range of 375–480 kHz (15). Energy is applied between the monopolar electrode that is placed in the target tissue and the large dispersive electrodes (grounding pads) that are placed on the body surface (16). A different strategy is to apply the RF energy exclusively into the target tissue by using bipolar electrodes (17). The electric circuit is closed between two RF electrodes placed in or at the periphery of the tumor, and no grounding pads are required. Thus, a design with two electrodes located on two different shafts (17) or on the same shaft (18) is possible.

The purpose of our study was to evaluate the size and geometry of thermally induced coagulation by using multipolar RF ablation and to determine a mathematic model to predict the volume of coagulation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
RF Technique and Equipment
RF ablations (n = 80) were performed at room temperature in eight fresh bovine livers that were obtained from an abattoir. Ablations were performed by using a 470-kHz ± 10 multipolar RF generator (CelonLab Power, with firmware version 1v12; Celon, Berlin, Germany) that provided a maximum power output of 250 W. RF energy was deposited by using internally cooled applicators with two electrodes (CelonPro Surge; Celon). When one bipolar applicator is connected to the RF generator, the energy is applied in bipolar mode (Fig 1a). When two or three bipolar applicators are connected, the RF system is operating in multipolar mode (Fig 1b–1f). The electrode shaft (diameter, 1.8 mm; length, 15 cm) contains two lumina that enable internal fluid circulation. In our study, distilled water (Ampuwa; Fresenius Kabi, Bad Homburg, Germany) was delivered at a rate of 30 mL/min at room temperature by using a peristaltic pump (CelonAquaflow III; Celon). The exposure tip of the bipolar applicator had a length of 4 cm and consisted of two noninsulated electrodes that were separated by 4 mm of insulation (Fig 2a). The parameters of tissue resistance, ablation time, and power output were transferred to a personal computer connected to the RF generator and were recorded by using a software program (CelonLab PowerTerm, v2.50; Celon).


Figure 1
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a: Schematic drawings demonstrate orientation of electric field for (a) single bipolar applicator and (b–f) the simultaneous use of a triangular array of bipolar applicators activated in multipolar mode. By using a single bipolar applicator with both electrodes located on the same applicator shaft, electric current is directed parallel to the applicator (a). If three bipolar applicators are used, they are activated in multipolar mode. In this mode, every possible pair of electrodes, which are not necessarily located on the same shaft, is activated one after the other for a short period of time. The electric current may pass between electrodes on the same or on different applicator shafts. Drawings show possible combinations between (b) pairs of electrodes located on the same shaft, (c) applicators 1 and 2, (d) applicators 1 and 3, and (e) applicators 2 and 3. (f) Drawing demonstrates that, altogether, activation can be switched between 15 different combinations in multipolar mode.

 

Figure 1
View larger version (44K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b: Schematic drawings demonstrate orientation of electric field for (a) single bipolar applicator and (b–f) the simultaneous use of a triangular array of bipolar applicators activated in multipolar mode. By using a single bipolar applicator with both electrodes located on the same applicator shaft, electric current is directed parallel to the applicator (a). If three bipolar applicators are used, they are activated in multipolar mode. In this mode, every possible pair of electrodes, which are not necessarily located on the same shaft, is activated one after the other for a short period of time. The electric current may pass between electrodes on the same or on different applicator shafts. Drawings show possible combinations between (b) pairs of electrodes located on the same shaft, (c) applicators 1 and 2, (d) applicators 1 and 3, and (e) applicators 2 and 3. (f) Drawing demonstrates that, altogether, activation can be switched between 15 different combinations in multipolar mode.

 

Figure 1
View larger version (38K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1c: Schematic drawings demonstrate orientation of electric field for (a) single bipolar applicator and (b–f) the simultaneous use of a triangular array of bipolar applicators activated in multipolar mode. By using a single bipolar applicator with both electrodes located on the same applicator shaft, electric current is directed parallel to the applicator (a). If three bipolar applicators are used, they are activated in multipolar mode. In this mode, every possible pair of electrodes, which are not necessarily located on the same shaft, is activated one after the other for a short period of time. The electric current may pass between electrodes on the same or on different applicator shafts. Drawings show possible combinations between (b) pairs of electrodes located on the same shaft, (c) applicators 1 and 2, (d) applicators 1 and 3, and (e) applicators 2 and 3. (f) Drawing demonstrates that, altogether, activation can be switched between 15 different combinations in multipolar mode.

 

Figure 1
View larger version (37K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1d: Schematic drawings demonstrate orientation of electric field for (a) single bipolar applicator and (b–f) the simultaneous use of a triangular array of bipolar applicators activated in multipolar mode. By using a single bipolar applicator with both electrodes located on the same applicator shaft, electric current is directed parallel to the applicator (a). If three bipolar applicators are used, they are activated in multipolar mode. In this mode, every possible pair of electrodes, which are not necessarily located on the same shaft, is activated one after the other for a short period of time. The electric current may pass between electrodes on the same or on different applicator shafts. Drawings show possible combinations between (b) pairs of electrodes located on the same shaft, (c) applicators 1 and 2, (d) applicators 1 and 3, and (e) applicators 2 and 3. (f) Drawing demonstrates that, altogether, activation can be switched between 15 different combinations in multipolar mode.

 

Figure 1
View larger version (36K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1e: Schematic drawings demonstrate orientation of electric field for (a) single bipolar applicator and (b–f) the simultaneous use of a triangular array of bipolar applicators activated in multipolar mode. By using a single bipolar applicator with both electrodes located on the same applicator shaft, electric current is directed parallel to the applicator (a). If three bipolar applicators are used, they are activated in multipolar mode. In this mode, every possible pair of electrodes, which are not necessarily located on the same shaft, is activated one after the other for a short period of time. The electric current may pass between electrodes on the same or on different applicator shafts. Drawings show possible combinations between (b) pairs of electrodes located on the same shaft, (c) applicators 1 and 2, (d) applicators 1 and 3, and (e) applicators 2 and 3. (f) Drawing demonstrates that, altogether, activation can be switched between 15 different combinations in multipolar mode.

 

Figure 1
View larger version (55K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1f: Schematic drawings demonstrate orientation of electric field for (a) single bipolar applicator and (b–f) the simultaneous use of a triangular array of bipolar applicators activated in multipolar mode. By using a single bipolar applicator with both electrodes located on the same applicator shaft, electric current is directed parallel to the applicator (a). If three bipolar applicators are used, they are activated in multipolar mode. In this mode, every possible pair of electrodes, which are not necessarily located on the same shaft, is activated one after the other for a short period of time. The electric current may pass between electrodes on the same or on different applicator shafts. Drawings show possible combinations between (b) pairs of electrodes located on the same shaft, (c) applicators 1 and 2, (d) applicators 1 and 3, and (e) applicators 2 and 3. (f) Drawing demonstrates that, altogether, activation can be switched between 15 different combinations in multipolar mode.

 

Figure 2
View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: (a) Applicator tip with two electrodes separated by insulation. (b) Three bipolar applicators placed in a triangular array in ex vivo bovine liver.

 
RF Protocols
Multipolar RF ablations (n = 80) were performed by using three bipolar applicators in an equidistant array (Fig 2b) (S.C. and D.S., with 2 and 4 years experience in ablation therapy, respectively). The placement of applicators close to large vessels and inside thin parts of the liver, being not capable to incorporate the zone of coagulation, was avoided. The maximum power output (75, 100, 125, 175, or 225 W) and the interapplicator spacing of the triangular array (2, 3, 4, or 5 cm) were systematically varied. The maximum power output of 250 W was not investigated because RF ablations at 225 W were already leading to a rapid increase in tissue resistance and to smaller zones of coagulation in this ex vivo setting. The duration of multipolar RF ablation at a power output of 75 W was long (up to 193 minutes). A power output of less than 75 W was not investigated because this would have further prolonged RF ablation.


Figure 2
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: (a) Applicator tip with two electrodes separated by insulation. (b) Three bipolar applicators placed in a triangular array in ex vivo bovine liver.

 
Four RF ablations were performed by using each combination of interapplicator spacing and power output settings. Precise spacing of the bipolar applicators was ensured by using an acrylic puncture aid. Typical tissue resistance in fresh liver ranged from 50 to 100 {Omega} for the electrodes used in our study. Therefore, the output characteristics of the RF generator were optimized for this resistance range. On the other hand, the final state of RF ablation is characterized by a steady desiccation of tissue, which is directly associated with a substantial increase in tissue resistance (13). For this reason, we selected a predefined resistance threshold of 500 {Omega} as an end point for RF energy application between the two bipolar electrodes. According to the RF generator characteristics, a value of 500 {Omega} was bound to a power output of approximately one-third of the power delivered at baseline resistance, which further decreased at higher resistances. Consequently, heating at resistance levels above the threshold was less effective, and a continuation of the ablation process was supposed to have no considerable effect on the coagulation volume.

In multipolar mode, every possible pair of electrodes, which were not necessarily located on the same shaft, was activated one after the other. The 15 possible pairs of electrodes were activated consecutively for 2 seconds each. In the first second of activation, power output was increased continuously from zero to either the preselected maximum power output or the maximum power output applicable to the actual tissue resistance. This ensured a gentle coagulation process without sudden vaporization of tissue water. The power output was then maintained for another second before the next pair of electrodes was activated, and the same scheme was applied. This procedure was optimized in pretests and ensured the best possible delivery of RF energy to the tissue. These findings were confirmed by those obtained in a study on monopolar RF ablation, which showed that alternating activation in intervals of 2 seconds, compared with simultaneous or sequential activation, is advantageous with respect to coagulation volume and shape (19).

After the RF generator had switched to every possible electrode pair, the cycle started again. If, in three cycles, the tissue resistance exceeded 500 {Omega} or the applicable power was less than 33% of preselected maximum power while a combination of electrodes was activated, the electrode pair was omitted from subsequent cycles. RF ablation was automatically stopped when the last possible pair of electrodes was excluded from the cycle. After 1 minute of tissue cooling, a second RF ablation was started by using the same settings. At the beginning of the second RF application, all 15 combinations of electrode pairs were again included.

Assessment of the Size and Geometry of Coagulation
After multipolar RF ablation was performed by using three bipolar applicators, the livers were sectioned (S.C., D.S.) in a plane that contained one of the three applicator tracks and the center of the triangular array (Fig 3a). We then assessed whether the zone of coagulation was well defined and had a regular shape. The diameters parallel (diameter a) and perpendicular (diameter b) to the applicator track of the white zone of coagulation were measured macroscopically by the consensus of three radiologists (S.C., D.S., A.B.) and one pathologist (S.M.K.). The short-axis diameter was defined as the shorter diameter of a and b.


Figure 3
View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: Two zones of coagulation after multipolar RF ablation. (a) Cross section of liver that was sliced in a plane containing one of three applicator tracks and the center of the triangular array. Diameters parallel (a) and perpendicular (b) to the applicator track were measured macroscopically. (b) Cross section perpendicular to applicator tracks shows confluent zone of coagulation. Line in b represents plane shown in a.

 
In an additional plane that was perpendicular to the three applicator tracks (Fig 3b), we assessed whether coagulation was confluent, partially confluent (clover leaf–shaped), or nonconfluent. The zone of coagulation was called confluent if the cross section that was perpendicular to the applicator shafts was approximately circular and if the cross section that was parallel to an applicator shaft was approximately oval or circular. If the shape of the cross section that was perpendicular to the applicator shafts was between that of a circle and triangle, the zone of coagulation was considered confluent when the variation of the cross-sectional diameters was less than 3 mm. In zones of coagulation that were classified as confluent according to these criteria, we calculated the coagulation volume (V) by using the formula for an ellipsoid, V = ({pi}/6)ab2, where a is the diameter parallel to the applicator track and b is the diameter perpendicular to the applicator track. The a/b ratio was then calculated to describe the geometry of confluent zones of coagulation, with a value close to 1.0 indicating a more spherical shape. In the case of partially confluent (clover leaf–shaped) or nonconfluent zones of coagulation, we did not assess the diameters and volume of coagulation.


Figure 3
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: Two zones of coagulation after multipolar RF ablation. (a) Cross section of liver that was sliced in a plane containing one of three applicator tracks and the center of the triangular array. Diameters parallel (a) and perpendicular (b) to the applicator track were measured macroscopically. (b) Cross section perpendicular to applicator tracks shows confluent zone of coagulation. Line in b represents plane shown in a.

 
Statistical Analysis
The model equations that were used will be described in the context of the reported results. Nonlinear fits (ie, higher order regression models) and linear regressions were performed with a statistical software package (JMP, version 5.1; www.jmp.com) (K.D., S.C.). To fit the power function y = axb, both y and x were log transformed in order to obtain the linear equation log y = log a + b(log x). For multivariate linear regression models, we tested the significance of the individual slope parameters by using the standard F test. Hyperbolic functions were fitted by using the classic method of least squares. The goodness of fit of a model was assessed by the parameter r2 = 1 – (mean squared error/variance), which can be interpreted as the proportion of the total variability explained by the model.

The probability of fusion was determined by using maximum likelihood as a function of both power output and applicator distance. The importance of the individual regression coefficients for power outputs and applicator distances was tested by using a univariate logistic regression model. P values of less than .05 were considered to indicate a statistically significant difference for all tests.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Induced coagulations were confluent for interapplicator distances of 2 cm (power output, 75–225 W), 3 cm (power output, 75–225 W), 4 cm (power output, 75–125 W), and 5 cm (power output, 75–100 W). Confluent zones of coagulation were well defined and had regular shapes. Areas of partially confluent and clover leaf–shaped coagulation necrosis were produced for applicator distances of 4 cm (power output, 175–225 W) and 5 cm (power output, 125–175 W). A power output of 225 W in combination with an interapplicator spacing of 5 cm resulted in nonconfluent coagulation necrosis.

The volume of coagulation and the duration of multipolar RF ablation were dependent on the preselected power output and interapplicator spacing. The volume of induced coagulation increased when RF ablation was performed at a lower maximum power output of 75 W and 100 W verses a higher maximum power output (Fig 4a). The zone of coagulation increased when the bipolar applicators had a more distant spacing. The maximum zone of coagulation (short-axis diameter, 8.4 cm ± 0.4 [standard deviation]; volume, 324 cm3 ± 63) was induced at a power output of 75 W and interapplicator spacing of 5 cm (Table).


Figure 4
View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: Graphs show the volume of coagulation and duration of multipolar RF ablation in relation to preselected power outputs and bipolar applicator distances. (a) Data points represent geometric mean of confluent volumes of coagulation at preselected power outputs and interapplicator distances. Line graph represents mathematic model for Equation (1). (b) Data points represent geometric mean of RF ablation time until an increase in tissue resistance occurred at preselected power outputs and interapplicator distances. Line graph represents mathematic model for Equation (2). Error bars indicate 95% confidence intervals. Coagulation volume and RF ablation time increased at lower power output levels and larger applicator distances.

 

View this table:
[in this window]
[in a new window]

  		Parameters for Multipolar RF Ablation

 
Parallel to the increase in coagulation when RF energy was applied at a lower power level, the duration of RF ablation became long and variable (Fig 4b)—that is, up to 193 min ± 46 at an interapplicator spacing of 5 cm. There was a marked increase in the duration of RF ablation when the preselected power output was reduced from 100 to 75 W. Increasing the preselected maximum power output resulted in smaller coagulation zones but shorter and more constant durations of RF ablation (Table). On the basis of our experimental data, two mathematic expressions were derived to determine the volume of coagulation (V, in cubic centimeters) and the duration of RF ablation (D, in hours) as a function of preselected power output (P, in watts) and interapplicator spacing (S, in centimeters). The first function is

Formula 1(1)
and explains 80% of the variability. The second function is

Formula 2(2)
and explains 93% of the variability. As concluded from Equations (1) and (2), the volume of coagulation (V) can be calculated as a function of the duration of RF ablation (D) and the interapplicator spacing (S):

Formula 3(3)
The line graph (Fig 5) of the volume of coagulation as a function of RF ablation time has a continuously decreasing slope. The short-axis diameter (X, in centimeters) as a function of power output (P) and interapplicator spacing (S) is given by the equation

Formula 4(4)
and explains 79% of the variability.


Figure 4
View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: Graphs show the volume of coagulation and duration of multipolar RF ablation in relation to preselected power outputs and bipolar applicator distances. (a) Data points represent geometric mean of confluent volumes of coagulation at preselected power outputs and interapplicator distances. Line graph represents mathematic model for Equation (1). (b) Data points represent geometric mean of RF ablation time until an increase in tissue resistance occurred at preselected power outputs and interapplicator distances. Line graph represents mathematic model for Equation (2). Error bars indicate 95% confidence intervals. Coagulation volume and RF ablation time increased at lower power output levels and larger applicator distances.

 

Figure 5
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5: Graph demonstrates the volume of coagulation in relation to the duration of multipolar RF ablation and applicator distance. Data points represent geometric mean of confluent volumes of coagulation at different preselected power outputs and interapplicator distances in relation to the duration of RF ablation. Line graph represents mathematic model for Equation (3). Error bars indicate 95% confidence intervals. During multipolar RF ablation, the volume of coagulation as a function of RF ablation time had a continuously decreasing slope.

 
On the basis of the Equations (1) and (2), the expected volume of coagulation (Fig 6a) and the duration of RF ablation (Fig 6b) were plotted relative to the different settings for power output and interapplicator spacing. The mathematic model (Eq [2]) for the duration of multipolar RF ablation has a singularity at 66.3 W. A preselected power output below this singularity would lead to an infinite application of RF energy without an automatic stop to RF ablation, which occurs with an increase of tissue resistance. Therefore, volumes of coagulation and durations of RF ablation for power outputs less than 66.3 W are not given in the presented plots.


Figure 6
View larger version (31K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6a: Graphs present (a) expected volume of coagulation and (b) duration of RF ablation. Graphs were calculated on the basis of mathematic models for Equations (1) and (2) and were plotted in relation to preselected power outputs and applicator distances.

 

Figure 6
View larger version (31K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6b: Graphs present (a) expected volume of coagulation and (b) duration of RF ablation. Graphs were calculated on the basis of mathematic models for Equations (1) and (2) and were plotted in relation to preselected power outputs and applicator distances.

 
To exclude the data points obtained for nonconfluent or partially confluent zones of coagulation from the presented plots, we calculated a linear function to separate these plots into two parts. This function is based on the extreme coordinates of data points, which reliably showed confluent zones of coagulation: y = 412.5 – (62.5 · distance), where interapplicator distance is measured in centimeters. The coordinates below this function represent confluent zones of coagulation and were therefore included in the presented plots. The combinations of power output and interapplicator spacing settings that did not reliably lead to confluent zones of coagulation were excluded from the presented plots. The probability that a zone of coagulation was not confluent was associated with a higher power output (P < .001) and a larger interapplicator spacing (P < .001) according to univariate logistic regressions. Mathematically, the probability (p) of fusion (F) is determined as a function of the expression F = distance + (0.02 · power output), where interapplicator distance is measured in centimeters and power output is measured in watts, and can be calculated as

Formula 5(5)
As F increases from 7.40 to 7.58, p declines from 99% to 1%. In conclusion, the probability (p) of creating a confluent zone of coagulation is 50% if F equals 7.48 and 99% or greater if F equals 7.40 or less. Mathematic modeling for Equations (1)–(3) and Equation (5) is summarized in Figure 7.


Figure 7
View larger version (49K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7: Graph presents relationship between the volume and fusion of coagulation and the duration of multipolar RF ablation as a function of preselected power output and applicator distance. The three-dimensional surface represents predicted volume of coagulation. Expected duration of RF ablation is color coded. The dark red line, which separates the color-coded area from the gray surface, corresponds to a probability that the zone of coagulation is 99% confluent, as predicted by Equation (4). The gray surface reflects the probability that the zone of coagulation is less than 99% confluent. For these combinations of power output and applicator distance, the predicted duration of RF ablation is not color coded; only the border of the different durations, as used for color coding, is given by the black lines.

 
In confluent zones of coagulation, the diameter parallel to the applicator shaft (diameter a) was slightly greater than the diameter perpendicular to the applicator shaft (diameter b) except for when a power output of 75 W and an interapplicator spacing of 5 cm was used (Table). The zone of coagulation was more spherical at interapplicator distances of 4 cm (a/b ratio, 1.1–1.2) and 5 cm (a/b ratio, 1.0) compared with interapplicator distances of 2 cm (a/b ratio, 1.3–1.5) and 3 cm (a/b ratio, 1.2–1.3) (Fig 8). Lower power output also induced coagulation zones that were more spherical. This finding is supported by the mathematic model that was calculated from our data, which showed that the logarithm of the a/b ratio is dependent on the logarithm of the power output (P < .05) and interapplicator spacing (P < .001). The model is given by the expression

Formula 6(6)
and explains 59% of the total variability.


Figure 8
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8: Graph presents the geometry of confluent zones of coagulation, which is described by the ratio between diameters parallel (diameter a) and perpendicular (diameter b) to the applicator shaft. Data points represent the mean a/b ratio for multipolar RF ablation at different preselected power outputs and applicator distances. An a/b ratio of close to 1.0 indicates a more spherical volume of coagulation. Line graph represents mathematic model for Equation (5), and error bars indicate standard deviations. Induced coagulation during multipolar RF ablation is more spherical at lower power outputs and larger interapplicator distances.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
The combination of three bipolar applicators in our study allowed the application of RF energy in multipolar mode. As in bipolar RF ablation, the electric circuit is closed between electrodes placed in close proximity. Bipolar and multipolar RF ablations are supposed to lead to a focal and more efficient deposition of RF energy in the target tissue and, subsequently, to an increased volume of thermally induced coagulation. Comparisons between bipolar and monopolar RF systems have shown that bipolar systems achieve greater energy efficiency, with considerably larger zones of coagulation and higher tissue temperatures (20,21). The thermodynamic effect may contribute to larger volumes of coagulation with bipolar RF ablation because one electrode is thermally shielded by the second electrode (20,21). This effect is supposed to be more pronounced in multipolar mode when RF energy is converted to heat in a large volume (owing to 15 different electric fields) instead of in a small tissue shell around one monopolar electrode. The presence of multiple independent heat sources is supposed to lead to less heat dissipation (21) and create a synergistic effect of additive heat diffusion, as has been shown during monopolar RF ablation with multiprobe arrays (11).

In a multipolar operating RF device, the application of RF energy can be controlled by using tissue resistance instead of impedance, which is influenced by the whole electric circuit. A separate measurement of the tissue resistance between every electrode pair enables monitoring of thermal effects in different parts of the target tissue; for example, a heat sink effect may lead to a less pronounced increase in tissue resistance between the adjacent pairs of electrodes. In multipolar mode, these pairs of electrodes can be activated separately, and RF energy can be applied selectively to untreated parts of the tumor after RF ablation has been completed in other parts of the tumor.

Depending on the variations in tissue resistance between the active and neutral electrode, monopolar RF ablation can produce an unpredictable electric field distribution in the patient's body (22). There is the possibility of heating metallic materials, like surgical clips, or of interference with pacemakers, thereby causing dysfunction (23). Moreover, an inadequate positioning of grounding pads can cause skin burns during monopolar RF ablation (24). These problems with the monopolar technique should not occur in bipolar or multipolar RF ablation because the electric circuit is closed inside the target tissue.

Our combined use of three bipolar applicators in multipolar mode produced large confluent zones of coagulation. The maximum volume of induced coagulation increased when multipolar RF ablation was performed at lower power output levels. Furthermore, an increased distance between the three applicators created an enlarged coagulation zone that was predominant in the direction perpendicular to the applicator shafts. The mathematic model for the volume of coagulation as a function of the duration of RF ablation has a continuously decreasing slope, which indicates that the multipolar mode is still effective for enlargement of the zone of coagulation, even when a large volume has already been coagulated. One prerequisite for achieving this enlargement of coagulation is to avoid a rapid increase in tissue resistance. The late increase in tissue resistance at low power outputs is supposed to be the reason that counterintuitive enlargement of the zones of coagulation is observed at lower power outputs. One drawback, however, is that a prolonged RF ablation (eg, mean duration of 193 minutes with an interapplicator spacing of 5 cm) is unsuitable for routine clinical application. In the ex vivo setting, power outputs in a middle range of 100–125 W in combination with an applicator spacing of 2–4 cm offer a reasonable compromise between the short-axis diameter and a clinically acceptable duration of RF ablation.

In our study, we performed a second ablation cycle to ensure that there was no preliminary exclusion of certain pairs of electrodes in the first ablation cycle. The second ablation cycle considerably increased the duration of RF ablation at lower power outputs, which probably had little effect on the overall volume of coagulation. Therefore, omitting the second ablation cycle might be possible, thereby shortening the duration of multipolar RF ablation. The probability that a zone of coagulation is nonconfluent is increased at higher power outputs and larger interapplicator distances. In our study, a wide range of different power outputs was investigated. Subsequently, the range of clinically relevant durations for RF application was intentionally expanded in both directions to obtain data from a broad range of different power outputs, which enabled more precise mathematic modeling. These mathematic functions enable an estimation of the coagulation volume, short-axis diameter, and duration of RF ablation, depending on preselected power output and applicator distance.

The combination of multiple bipolar applicators encourages larger coagulations with a single RF ablation. The simultaneous combination of two to four hybrid bipolar RF and cryoablation applicators yielded large and contiguous zones of ablation (25). When bipolar RF ablation was performed with both electrodes located on the same applicator shaft, the high-frequency electric current was oriented parallel to the applicator. In multipolar mode, the application of RF energy switches between electric fields that are parallel to the applicators and electric fields that cross the target tissue within the applicators. Therefore, when the three bipolar applicators are activated in multipolar mode rather than separately, the risk of partially confluent or nonconfluent volumes of coagulation should be reduced, and a more distant spacing between applicators should be possible. When more than one bipolar applicator was connected to the RF generator, the activation of pairs of electrodes could not be restricted to certain pairs of electrodes. Therefore, it was not possible to determine the portion in which certain pairs of electrodes contributed to the volume and fusion of coagulation (eg, pairs of electrodes located on either the same shaft or different shafts).

In our study, the maximum volume of coagulation at multipolar RF ablation was limited at higher power outputs. Additionally, applicable power outputs with saline-enhanced RF ablation are lower in bipolar mode than in monopolar mode (20). Modeling of sodium chloride–enhanced monopolar RF ablation showed that heat induction may be limited by the maximum power output of the RF generator (26). Consequently, a strategy for monopolar RF ablation would be to increase the power output of the generator to enlarge the volume of coagulation. Further improvement in multipolar RF ablation will possibly permit higher power outputs than those used in our study. Nevertheless, optimum power output, which is higher for monopolar RF ablation, seems to be the fundamental difference between monopolar and multipolar RF ablation.

For the clinical application of multipolar RF ablation, additional considerations are necessary. A major limitation of ex vivo RF ablation is that perfusion-mediated tissue cooling and heat sink effects are not taken into consideration (27–29). Thus, the volume of coagulation will be reduced in vivo, and the shape of coagulation could be altered by blood vessels (8,25,30,31). If tissue perfusion adjacent to bipolar electrodes is not homogenous, a different amount of RF energy will be converted into heat, which may lead to an asymmetric induction of coagulation (21). Perfusion-mediated tissue cooling could prevent coagulation from being completely confluent and may require a reduction in interapplicator distance. The relationship between the parameters of multipolar RF ablation is supposed to be altered in vivo. Nevertheless, ex vivo experiments are important for the initial evaluation of principles and new techniques of RF ablation.

As is shown in the modeling of temperature response to adjuvant sodium chloride treatment, a good correlation between mathematic modeling for ex vivo and in vivo settings is possible and the effects hold true, despite the greater physiologic variability in vivo (26). Mathematic modeling of multipolar RF ablation, however, must be adjusted by further investigations in vivo. An accurate placement of three bipolar applicators is necessary in multipolar RF ablation. This may lead to a longer RF procedure when cluster electrodes or multitined expandable electrodes are used. In cases for which critical anatomic structures or obstacles such as ribs may prevent parallel and equidistant applicator placement, this could lead to nonspherical and irregular volumes of coagulation.

The initial investigation of multipolar RF ablation in the ex vivo bovine liver is limited by several factors. Further in vivo evaluation of this technique is necessary, and verification of mathematic modeling, taking into consideration the presence of tumor tissue and blood flow, is mandatory. Our study was based on fixed, preselected power outputs, which led to a variable duration of RF ablation. For clinical application, an algorithm based on a predefined duration of RF ablation or on a certain amount of applied energy may be an alternative. The results of our study present an initial evaluation of multipolar RF ablation with several parameters that thus far have not been optimized. Hence, further evaluation and improvement of this technique is necessary.

In conclusion, findings from our ex vivo study demonstrate that multipolar RF ablation with internally cooled bipolar electrodes is capable of producing large and spherical zones of coagulation, with the drawback of a prolonged application of RF energy. On the other hand, multipolar RF ablation reduces the necessity for electrode replacement and sequential overlapping ablations. Nevertheless, the combination of a large interapplicator distance and high power output level leads to an increased risk of nonconfluent zones of coagulation.

Practical application: In ex vivo multipolar RF ablation, a compromise between prolonged RF ablation at lower power output levels, which produces larger volumes of coagulative necrosis, and faster RF ablation at higher power output levels, which produces smaller volumes of coagulative necrosis, is necessary. For clinical application of multipolar RF ablation, algorithms have to be defined. The mathematic modeling of multipolar RF ablation is based on ex vivo experiments and should be confirmed or modified with results from further investigations and clinical experience with this technique.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   FOOTNOTES
 

Abbreviations: RF = radiofrequency

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, S.C., C.D.C., P.L.P.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, S.C., D.S., P.L.P.; experimental studies, S.C., D.S., A.B., S.M.K., P.L.P.; statistical analysis, S.C., K.D.; and manuscript editing, S.C., K.D.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. McGahan JP, Dodd GD. Radiofrequency ablation of the liver: current status. AJR Am J Roentgenol 2001;176:3–16.[Free Full Text]
  2. Gazelle GS, Goldberg SN, Solbiati L, Livraghi T. Tumor ablation with radio-frequency energy. Radiology 2000;217:633–646.[Abstract/Free Full Text]
  3. Solbiati L, Livraghi T, Goldberg SN, et al. Percutaneous radio-frequency ablation of hepatic metastases from colorectal cancer: long-term results in 117 patients. Radiology 2001;221:159–166.[Abstract/Free Full Text]
  4. Livraghi T, Goldberg SN, Lazzaroni S, et al. Hepatocellular carcinoma: radio-frequency ablation of medium and large lesions. Radiology 2000;214:761–768.[Abstract/Free Full Text]
  5. Chen MH, Yang W, Yan K, et al. Large liver tumors: protocol for radiofrequency ablation and its clinical application in 110 patients—mathematic model, overlapping mode, and electrode placement process. Radiology 2004;232:260–271.[Abstract/Free Full Text]
  6. Dodd GD 3rd, Frank MS, Aribandi M, Chopra S, Chintapalli KN. Radiofrequency thermal ablation: computer analysis of the size of the thermal injury created by overlapping ablations. AJR Am J Roentgenol 2001;177:777–782.[Abstract/Free Full Text]
  7. Khajanchee YS, Streeter D, Swanstrom LL, Hansen PD. A mathematical model for preoperative planning of radiofrequency ablation of hepatic tumors. Surg Endosc 2004;18:696–701.[CrossRef][Medline]
  8. Goldberg SN, Solbiati L, Hahn PF, et al. Large-volume tissue ablation with radio frequency by using a clustered, internally cooled electrode technique: laboratory and clinical experience in liver metastases. Radiology 1998;209:371–379.[Abstract/Free Full Text]
  9. Schmidt D, Trubenbach J, Brieger J, et al. Automated saline-enhanced radiofrequency thermal ablation: initial results in ex vivo bovine livers. AJR Am J Roentgenol 2003;180:163–165.[Free Full Text]
  10. Livraghi T, Goldberg SN, Monti F, et al. Saline-enhanced radio-frequency tissue ablation in the treatment of liver metastases. Radiology 1997;202:205–210.[Abstract/Free Full Text]
  11. Goldberg SN, Gazelle GS, Dawson SL, Rittman WJ, Mueller PR, Rosenthal DI. Tissue ablation with radiofrequency using multiprobe arrays. Acad Radiol 1995;2:670–674.[Medline]
  12. de Baere T, Denys A, Wood BJ, et al. Radiofrequency liver ablation: experimental comparative study of water-cooled versus expandable systems. AJR Am J Roentgenol 2001;176:187–192.[Abstract/Free Full Text]
  13. Goldberg SN, Stein MC, Gazelle GS, Sheiman RG, Kruskal JB, Clouse ME. Percutaneous radiofrequency tissue ablation: optimization of pulsed-radiofrequency technique to increase coagulation necrosis. J Vasc Interv Radiol 1999;10:907–916.[Medline]
  14. Rossi S, Garbagnati F, Lencioni R, et al. Percutaneous radio-frequency thermal ablation of nonresectable hepatocellular carcinoma after occlusion of tumor blood supply. Radiology 2000;217:119–126.[Abstract/Free Full Text]
  15. Pereira PL, Trubenbach J, Schenk M, et al. Radiofrequency ablation: in vivo comparison of four commercially available devices in pig livers. Radiology 2004;232:482–490.[Abstract/Free Full Text]
  16. Rhim H, Goldberg SN, Dodd GD 3rd, et al. Essential techniques for successful radio-frequency thermal ablation of malignant hepatic tumors. RadioGraphics 2001;21(Spec Issue):S17–S35.[Abstract/Free Full Text]
  17. McGahan JP, Gu WZ, Brock JM, Tesluk H, Jones CD. Hepatic ablation using bipolar radiofrequency electrocautery. Acad Radiol 1996;3:418–422.[CrossRef][Medline]
  18. Burdio F, Guemes A, Burdio JM, et al. Bipolar saline-enhanced electrode for radiofrequency ablation: results of experimental study of in vivo porcine liver. Radiology 2003;229:447–456.[Abstract/Free Full Text]
  19. Lee JM, Rhim H, Han JK, Youn BJ, Kim SH, Choi BI. Dual-probe radiofrequency ablation: an in vitro experimental study in bovine liver. Invest Radiol 2004;39:89–96.[CrossRef][Medline]
  20. Lee JM, Han JK, Kim SH, et al. A comparative experimental study of the in-vitro efficiency of hypertonic saline-enhanced hepatic bipolar and monopolar radiofrequency ablation. Korean J Radiol 2003;4:163–169.[Medline]
  21. Haemmerich D, Staelin ST, Tungjitkusolmun S, Lee FT Jr, Mahvi DM, Webster JG. Hepatic bipolar radio-frequency ablation between separated multiprong electrodes. IEEE Trans Biomed Eng 2001;48:1145–1152.[CrossRef][Medline]
  22. Desinger K, Stein T, Tschepe J, Mueller G. Investigations on radio-frequency current application in bipolar technique for interstitial thermotherapy (RF-ITT). Minim Invasive Med 1996;7:92–97.
  23. Tong NY, Ru HJ, Ling HY, Cheung YC, Meng LW, Chung PC. Extracardiac radiofrequency ablation interferes with pacemaker function but does not damage the device [letter]. Anesthesiology 2004;100:1041.[CrossRef][Medline]
  24. Goldberg SN, Solbiati L, Halpern EF, Gazelle GS. Variables affecting proper system grounding for radiofrequency ablation in an animal model. J Vasc Interv Radiol 2000;11:1069–1075.[Medline]
  25. Hines-Peralta A, Hollander CY, Solazzo S, Horkan C, Liu ZJ, Goldberg SN. Hybrid radiofrequency and cryoablation device: preliminary results in an animal model. J Vasc Interv Radiol 2004;15:1111–1120.[Medline]
  26. Lobo SM, Afzal KS, Ahmed M, Kruskal JB, Lenkinski RE, Goldberg SN. Radiofrequency ablation: modeling the enhanced temperature response to adjuvant NaCl pretreatment. Radiology 2004;230:175–182.[Abstract/Free Full Text]
  27. Goldberg SN, Grassi CJ, Cardella JF, et al. Image-guided tumor ablation: standardization of terminology and reporting criteria. Radiology 2005;235:728–739.[Abstract/Free Full Text]
  28. Goldberg SN, Hahn PF, Tanabe KK, et al. Percutaneous radiofrequency tissue ablation: does perfusion-mediated tissue cooling limit coagulation necrosis? J Vasc Interv Radiol 1998;9:101–111.
  29. Lu DS, Raman SS, Vodopich DJ, Wang M, Sayre J, Lassman C. Effect of vessel size on creation of hepatic radiofrequency lesions in pigs: assessment of the "heat sink" effect. AJR Am J Roentgenol 2002;178:47–51.[Abstract/Free Full Text]
  30. Patterson EJ, Scudamore CH, Owen DA, Nagy AG, Buczkowski AK. Radiofrequency ablation of porcine liver in vivo: effects of blood flow and treatment time on lesion size. Ann Surg 1998;227:559–565.[CrossRef][Medline]
  31. Goldberg SN, Hahn PF, Halpern EF, Fogle RM, Gazelle GS. Radio-frequency tissue ablation: effect of pharmacologic modulation of blood flow on coagulation diameter. Radiology 1998;209:761–767.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 RadiologyHome page
O. Seror, G. N'Kontchou, M. Ibraheem, Y. Ajavon, C. Barrucand, N. Ganne, E. Coderc, J. Claude Trinchet, M. Beaugrand, and N. Sellier
Large (>=5.0-cm) HCCs: Multipolar RF Ablation with Three Internally Cooled Bipolar Electrodes--Initial Experience in 26 Patients
Radiology, July 1, 2008; 248(1): 288 - 296.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
P. F. Laeseke, T. M. Frey, C. L. Brace, L. A. Sampson, T. C. Winter III, J. R. Ketzler, and F. T. Lee Jr.
Multiple-Electrode Radiofrequency Ablation of Hepatic Malignancies: Initial Clinical Experience
Am. J. Roentgenol., June 1, 2007; 188(6): 1485 - 1494.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2382050571v1
238/3/881    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Clasen, S.
Right arrow Articles by Pereira, P. L.