| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Experimental Studies |
1 From the Departments of Surgery A (F.B., A.G., A.N., R.S., R.L.) and Pathology (T.C.), Hospital Clínico Universitario Lozano Blesa, Zaragoza, Spain; and Departments of Electric Engineering and Communications (J.M.B.) and Animal Pathology and Surgery, Veterinary Faculty (I.C., O.B.), University of Zaragoza, Spain. Received August 5, 2002; revision requested September 24; final revision received February 26, 2003; accepted March 11. Supported by Spanish Government grant FIS 00/255 for medical research. Address correspondence to F.B., Tapis 155, 17600 Figueres (Girona), Spain (e-mail: fburdio@comll.es).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: RF ablation (480 kHz) of the liver was performed in 24 healthy pigs by means of laparotomy: group A (n = 5), 4-cm distance between electrodes 1 and 2; group B (n = 7), 4-cm distance and the Pringle maneuver; group C (n = 5), 2-cm distance; and group D (n = 7), 2-cm distance with the Pringle maneuver. Twenty percent NaCl solution was infused continuously at a rate of 100 mL/h via each electrode during the procedure. The pigs were followed up, and they were euthanized on the 7th day. Livers were removed for histologic assessment. Time, impedance, current, power output, specific voltage of the contacts, energy output, temperatures in the liver, volume of the lesion, and energy delivered per lesion volume were determined and compared among groups. Predictability of lesion volume was evaluated with the coefficient of variability. Mean values of the variables were compared among the groups by means of one-way analysis of variance or Kruskall-Wallis test.
RESULTS: Impedance at the end of the RF ablation procedure was almost twofold lower than the corresponding initial value in all groups. In Pringle groups B and D, regular ellipsoids of coagulation necrosis were created (mean lesion volume, 149.50 cm3 ± 34.26 and 69.43 cm3 ± 15.48, respectively). In non-Pringle groups A and C, the shape of coagulation necrosis was influenced by the vessels encountered, and mean lesion size was lower than that in the Pringle groups (P < .01). The coefficient of variability of lesion size was lower in the Pringle groups (23% and 22%, respectively) than that in the non-Pringle groups (75% and 30%, respectively).
CONCLUSION: The bipolar saline-enhanced RF ablation method produces homogeneous and predictable areas of coagulation necrosis between two electrodes, regardless of the distance between them, preferably with vascular inflow occlusion.
© RSNA, 2003
Index terms: Animals • Experimental study • Liver, interventional procedures, 761.1269 • Liver neoplasms, 761.1269 • Radiofrequency (RF) ablation, 761.1269
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
One of the most effective strategies to increase the extent of coagulation necrosis is to infuse an NaCl solution through a monopolar electrode prior to or during RF application. Three mechanisms might account for the improved tissue heating and increased volume of necrosis with this technique: (a) enlargement of the effective electrode surface area by means of augmented tissue tonicity and therefore better electric conductivity to permit greater RF energy deposition; (b) improvement of tolerance for sustained high generator output as a result of tissue cooling, decreased impedance, or both; and (c) diffusion of heated saline into tissue (7). The resulting foci of coagulation necrosis are somewhat irregular in shape, however, and the volume of tissue necrosis is often difficult to predict (7–11). Renewed interest in this technique is attributable mainly to technical optimization in both NaCl solution concentration and infusion volume to provide the most tissue heating and coagulation with better predictability (12).
Burdío et al (11,13) designed an approach for RF ablation with two needles that are perfused with saline in a bipolar way. First, they demonstrated ex vivo that this method could create ellipsoids of coagulation necrosis that were as large as the distance between the two electrodes (11). These lesions were larger and more regular than those obtained with the monopolar method. The mean volume achieved with this method ex vivo was 144.88 cm3 ± 59.8 (more than threefold larger than that with currently available methods in similar conditions). Second, a dramatic increase in lesion size attainable with this method combined with or without the Pringle maneuver (temporary vascular inflow occlusion) (13) was also demonstrated in vivo (in a pig liver model). The mean lesion volume was 123.22 cm3 ± 49.62 in this model with the Pringle maneuver. Nevertheless, relatively low predictability of the lesion was demonstrated, especially in the non-Pringle group. This suboptimal predictability was attributed primarily to inaccuracies of electrode placement, uneven diffusion of saline into the tissue, or both. The purpose of our study was to evaluate whether a bipolar saline-enhanced RF ablation system embedded in one needle is able to consistently produce homogeneous and predictable areas of coagulation necrosis with or without the Pringle maneuver of vascular inflow occlusion.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
Catheter Design
The needle electrode used in the current study is a stiff straight 2.6-mm-diameter dual-lumen hollow stainless steel cannula that contains the two bipolar electrodes and the two perfusion systems in a single device (Fig 1). Electrodes 1 and 2 have an exposed 2.5-cm-long tip with three side grooves for saline infusion into tissue. The distance between the two electrodes can be selected before the needle is inserted by using an acrylic blocking device that offers the advantage of operational mobility to the required position. The inner channel is coated with fluorine-containing resins, which act as an electric insulator up to the exposed tip. With this approach, current flows between the two uninsulated RF electrodes in the liver without relying on a distant grounding pad. RF exposure is always accompanied by simultaneous infusion of NaCl solution through each channel up to the electrodes. The effective length of coagulation necrosis that can be expected is the total distance between the exposed electrode tips regardless of their separation. The height and width of the thermal lesion depend on time exposure; in optimal conditions, a homogeneous ellipsoid of coagulation necrosis of virtually any size can be created.
|
|
A 100-W RF-current (480-kHz) generator (model CC1; Radionics, Burlington, Mass) was used for each experiment. In all cases, the needle electrode was inserted up to the landmark in electrode 2 to ensure that the same length was in contact with the liver for both electrodes 1 and 2. Hypertonic saline (20% NaCl) at room temperature was infused through the needles into the liver by using two infusion pumps (model P1000; IVAC, Hampshire, England) at a steady rate of 100 mL/h. Four thermocouples (model K; Farnell, Leeds, England) were inserted in the tissue and connected to a temperature acquisition unit (Hydra; Fluke, the Netherlands).
RF Ablation Protocol
Four sets of experiments were performed to study our method (Fig 2). One experiment was conducted with each animal. Group A (n = 5), 4-cm distance between the two electrodes (total length of 9 cm, including electrode lengths); group B (n = 7), 4-cm distance (total length of 9 cm) and the Pringle maneuver; group C (n = 5), 2-cm distance (total length of 7 cm); and group D (n = 7), 2-cm distance (total length of 7 cm) with the Pringle maneuver. The Pringle maneuver was applied by using a Rummel tourniquet to obstruct the inflow of portal venous and hepatic arterial blood. The inferior surface of the liver was isolated from the stomach and the gut with cool wet gauzes.
|
The RF electrode was placed directly into the center of the largest lobe of the liver along its longest axis (A.G., R.S). Hypertonic saline (20% NaCl) was infused continuously through each electrode at a rate of 100 mL/h by using two infusion pumps. An RF current (480 kHz) was delivered between the two electrodes 30 seconds after initiation of the perfusion of saline. Power output was set manually with the following protocol: In groups A and B, initial power output was set manually at 30 W at the 1st minute, increased to 60 W at the 2nd minute, increased to 100 W at the 3rd minute, and maintained at 100 W thereafter. In group C, initial power was set at 30 W at the 1st minute and increased 10 W every 2 minutes up to a maximal power output of 80 W. In group D, power output was set at 30 W at the 1st minute. Thus, lesions in groups A and B were created with a high-power and high-energy protocol, whereas lesions in groups C and D were created with a low-power and low-energy protocol to assess this effect.
To appraise the specific electric contact of electrodes 1 and 2 with the tissue, the specific voltage of the contacts (V1 and V2, respectively) were registered by using an oscilloscope (model THS720P; Tektronik, Bearenton, Oregon) connected to each pole of the circuitry. An indifferent metal grounding pad (200-cm2 total surface area) was affixed to the lower back of the animal and connected to the oscilloscope for a reference control voltage.
Four thermocouples were placed into the tissue at a depth of 1 cm for remote thermometry: Thermocouple 1 was placed over the needle tip of electrode 1 at 1.25 cm from the puncture site, thermocouple 2 was placed over the needle tip of electrode 2 at 7.75 cm in groups A and B and at 5.75 cm in groups C and D, thermocouple 3 was placed between thermocouples 1 and 2 at 4.5 cm in groups A and B and at 3.5 cm in groups C and D, and thermocouple 4 was placed in the hepatic lobe closest to the targeted lesion for control. The end point of ablation was determined by taking into account the marked variability of the shape and volume of the liver in the pig, as well as the similarity to clinical application, where the end point of ablation is considered to be treatment of the targeted lesion. Thus, experiments were stopped when the hyperthermic lesion reached the superior surface of the targeted lobe of the liver (yellow appearance). Time of the procedure, impedance, current, power output, energy output, temperatures (T1–T4) in the liver, central temperature (Tc) of the animal, and volume of the lesion were determined. Additionally, three ratios were calculated for each case: (a) energy delivered per lesion volume (Evol); (b) impedance ratio, defined as initial impedance divided by final impedance; and (c) mean specific voltage difference, defined as V2 - V1, taking into account the specific voltage of each electrode as registered with the oscilloscope during the procedure.
Pathologic Assessment
After the experiments, the pigs were allowed to recover from anesthesia. They were euthanized 7 days later. All complications related to the technique were evaluated then. Livers were removed en bloc and processed for gross and pathologic examination. Then, livers were sliced approximately 1 cm along the axis that joined electrodes 1 and 2. When a regular ellipsoid lesion was encountered, diameters a, b, and c were measured directly. When a nonregular ellipsoid lesion was observed, then for each diameter a, b, and c, a mean diameter was calculated to obtain the mean a, b, and c diameters for each lesion. For each experiment, lesion volume (Vol) was calculated with the formula of the real ellipsoid: Vol = (1/6)
abc.
Macroscopic measurements were performed with the consensus of two investigators (F.B., A.N.). Histologic samples were submitted for microscopic review. Specimens were fixed in formalin, embedded in paraffin, cut, and stained with hematoxylin-eosin. Microscopic findings of coagulative necrosis (cytoplasmic shrinkage, densification, and variable degree of chromatin condensation with focal formation of apoptotic-like bodies) were sought. Representative specimens were also stained with reticulin or by means of a histochemical technique with a tissue oxidative enzyme, nicotinamide adenine dinucleotide, to assess viability of cells. The histologic study was performed by a pathologist (T.C.) in the presence of another author (A.N.) to compare results at gross analysis with those at histologic examination.
Statistical Analysis
Mean values of the procedure time, impedance, current, power output, energy output, temperatures (T1–T4 and Tc), volume, and energy delivered per lesion volume were compared and correlated among groups. Values were expressed as the mean ± SD. Additionally, the variability and predictability of the lesion volume in each group was studied by taking into account the coefficient of variability (15), which is defined as the SD of volume divided by the mean volume, expressed as a percentage. The Kolmogorov-Smirnov test was used to determine whether values followed a normal distribution. The Levene test was used to study the equality of variances. Mean values of the variables were compared among the groups by means of one-way analysis of variance or Kruskall-Wallis test. A general linear model of analysis of variance (repeated measures) was used to study differences between the mean final temperatures within the lesion. Apriori contrast tests and post hoc (Bonferroni) tests were conducted. Differences in variables were considered to be significant at a threshold of P < .05. The Pearson correlation coefficient was determined for the height and width of the lesion. All statistical analyses were performed with statistical software (SPSS, version 8.0; SPSS, Chicago, Ill).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
± 14.91. Values of impedance decreased linearly during the procedure in all groups in a similar manner. Therefore, final impedance at the end of the RF ablation procedure was almost twofold lower than the corresponding initial value, and no significant differences were found between the groups (see impedance ratios in Table 1). Spontaneous increases in selected power were linked to decreasing impedances during the procedures. Nevertheless, in three cases (one in group A, one in group B, and one in group D), an increase in impedance was observed with subsequent decrease of deposited power.
|
For best control of global impedance of the circuitry, the specific voltage of each prong (V1 for electrode 1 and V2 for electrode 2) was registered, and the mean specific voltage difference (SVD) between them was calculated: SVD = V2 - V1. Significant differences (P < .01) were observed between the high- and low-power groups, showing less symmetric energy dissipation in the high-power groups. Furthermore, the absolute value of the difference between the instantaneous specific voltages was maximum when impedance increased.
No significant differences were found among groups in the temperatures in the tissue around thermocouples 1 and 2 at the end of the procedure. Nevertheless, significant differences (P < .01) were observed in the temperatures of thermocouple 3, at the center of the targeted tissue, between the non-Pringle groups A and C compared with the Pringle groups B and D (Table 1).
Results of the general linear model analysis of differences among the mean final temperatures (T1–T3) of the lesions showed no significant differences either globally in groups A–D together or in non-Pringle groups A and C together. Conversely, in Pringle groups B and D together, the mean final temperature T3 was even greater than T1 and T2 (91.72°C ± 11.68, 70.35°C ± 26.56, and 79.21°C ± 15.96, respectively; P < .05) (Fig 3). Nevertheless, no significant differences were found among the final mean values of T1–T3 in groups B and D independently.
|
|
Evaluation of Volume and Shape
The mean lesion volumes in Pringle groups were much larger than those in the corresponding non-Pringle groups (P < .01) (Table 2). The mean lesion volume in group B (149.50 cm3 ± 34.26 [SD]) is, to our knowledge, the largest mean lesion size described in the literature for an in vivo procedure in pig liver with any RF ablation method. Unlike the huge amount of necrosis with vascular inflow occlusion (groups B and D), the mean lesion size achieved without it (groups A and C) was much smaller (26.03 cm3 ± 19.53 and 23.41 cm3 ± 7.15, respectively).
|
The coefficient of variability of volume size was also better in Pringle groups B and D than in non-Pringle groups A and C (23%, 22%, 75%, and 30%, respectively), which shows worse predictability of the thermic lesion for well-perfused livers. This predictability was markedly worse (coefficient of variability, 75%) with greater separation of the electrodes (4 cm) without the Pringle maneuver (group A).
Gross and Histologic Examinations
Gross examination revealed the lesions to be elliptic with a pale or yellowish homogeneous appearance (Fig 4). In two cases in group A, a central 1-mm-wide area of char was found near the path for electrode 2. In five of seven (71%) pigs in group B, homogeneous crumbly tissue was encountered in the treated lesion (Fig 5).
|
|
Complications and Adverse Effects
Three major complications occurred in three of 24 (12%) pigs, including two deaths (both in group B). One pig experienced moderate inactivity, diminished appetite, and cyanosis the day after the procedure and died on the 2nd postoperative day. A burn in the diaphragm without pulmonary involvement or hemoperitoneum was found at autopsy. In the other pig, a small burn was found on the lesser curve of the stomach at the end of the ablation procedure. Large lobar necrosis with special hepatic inferior surface involvement was likely the cause of the burn. The animal died on the 6th postoperative day. At autopsy, gastric dilatation and a deep burn on the lesser curve of the stomach were found. The last major complication was biliary peritonitis caused by a disrupted intrahepatic bile duct in 
crumbly
coagulation necrosis.
Three minor complications occurred. In a pig in group A, a small burn on the diaphragm was found at autopsy, but the postoperative course was uneventful. In a crumbly lesion in a pig in group B, a 3-cm-diameter hematoma was found at autopsy, but there was no apparent repercussion. Finally, moderate inactivity and diminished appetite were observed in the postoperative course of one pig in group D, but recovery was good. No signs of involvement in the vicinity of organs were found at autopsy on the 7th postoperative day.
All animals that experienced a complication were included in statistical analysis.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
An RF lesion is the result of tissue destruction caused by resistive heating in a narrow rim of tissue that surrounds the uninsulated part of the electrode because heating decreases as the inverse of the fourth power of the distance, which is a steep exponential decay (33,34). Deeper tissue heating occurs as a result of passive heat conduction from that rim. A further limitation in lesion size is the very high temperature at the tissue-electrode interface. This leads to tissue desiccation and coagulation, which results in a rapid and significant increase in impedance around the electrode. These effects prevent further RF energy conduction beyond the desiccated tissue and halt further tissue coagulation. Furthermore, flowing blood in the vessels acts as a heat sink and substantially limits the necrotizing effect of RF treatment in adjacent tissue.
Interstitial continuous saline infusion through a monopolar electrode is another strategy to increase the efficacy of RF ablation. In this way, the interstitial electrolyte perfuses the RF current further into the tissue away from the surface of the electrode. This allows a greater amount of RF energy to be delivered to the tissue without the critical current density being reached and to avoid, at least at the beginning of the procedure, the desiccation and char formation at the tissue-electrode interface (35). Therefore, a substantial increase in the volume of necrosis is achieved (up to 60 cm3) (11). Nevertheless, irregularly shaped areas of coagulation have been observed previously with this technique and attributed to nonuniform saline distribution. Moreover, a decreasing gradient of temperature has been described between tissue close to the electrode and that distant from the electrode, with uneven char formation around the electrode at the end of the procedure (11).
Bipolar RF ablation without infusion of saline has been described by several authors (36,37). This has the decided advantage of increasing local current density and therefore tissue heating and lesion size. The disadvantage of a dry bipolar system is that as a result of the increasing impedance around the two electrodes during the procedure, there will be low current density between them, which leads to cool spots between the electrodes. Therefore, lesions surrounding these electrodes may never become confluent if the electrodes are separated by more than 2–3 cm, even with use of two expandable devices (37).
Results in the current study underscore the fact that with our method, with use of two electrodes that are both perfused with hypertonic saline by using independent pumps, almost homogeneous and significant resistive heating is present in any area of the tissue between the two electrodes. This creates predictable ellipsoids of coagulation necrosis of virtually any size when the influence of blood cooling is limited. In other words, with our method, the targeted tissue between the two electrodes is converted into electric resistance. This allows dissipation of energy beyond the electrodes themselves with less reliance on passive heat conduction. Thus, the current density is no longer concentrated around the electrode. Instead, it, and consequently the power source, is spread over the region interposed between the two electrodes. Five experimental evidences from this study support this concept.
Evidence 1.—We demonstrated that our method is able to achieve uniform lesions between the two electrodes with variable distance (2 or 4 cm), especially with vascular inflow occlusion, and no lesion was observed outside this area. Furthermore, in our laboratory, continuous lesions even longer than 15 cm were obtained ex vivo with further separation between the two electrodes. This can be explained only by direct heat delivered into the tissue by means of electric current flowing through the targeted tissue. Neither passive heating conduction nor diffusion of heated saline alone can account for that phenomenon at those distances between the electrodes.
Evidence 2.—The increasing impedance values that are seen with dry RF ablation are not seen with our method. In fact, with our method, we demonstrated in each group that tissue impedance actually decreased while current increased during the procedure. This is probably a result of the electrolyte coupler spreading the RF energy away from the electrode tips, which changes the inherent conductivity of the tissue and turns the tissue itself into a virtual electrode. Decreases in impedance that occurred during our procedure can also be observed with monopolar interstitial saline infusion, but the effect with the monopolar interstitial saline method is frequently interrupted by an abrupt impedance increase with an audible pop or steaming that limits further energy deposition, especially after a certain time of RF application (8,11).
Evidence 3.—The amount of energy used per volume of coagulation necrosis is a key factor in the way the method works. In a similar pig liver model with the monopolar saline-enhanced method without the Pringle maneuver, Livraghi et al (7) used 1,500 J/cm3. Chinn et al (38), with an expandable electrode in a pig liver model and the Pringle maneuver, used 1,666 J/cm3. Curley et al (24), who treated primary and secondary tumors in the human liver with an expandable electrode combined with the Pringle maneuver, used 2,032 J/cm3. In general, between 1,000 and 2,000 J/cm3 are used with the methods described in the literature, regardless of the method. With our method in group D, 245 J/cm3 ± 63 were used; therefore, approximately 10-fold less energy was used per unit volume of coagulation necrosis.
This feature can only be explained by overtreatment or overheating (charring) of the zones of tissue nearer the electrodes with the conventional electrodes in contrast to our method. Furthermore, we demonstrated that if we delivered more than 300 J/cm3 with the Pringle maneuver (group B), then the targeted area of the liver might be converted into homogeneous crumbly tissue, which showed a high deposition of energy per volume of tissue. This phenomenon also correlates with the higher mean final temperatures attained at all thermocouples in group B compared with those in group D.
Evidence 4.—With almost any monopolar approach for RF ablation (even for monopolar saline-enhanced RF ablation), tissue temperature decreases rapidly with increasing distance from the electrode (26,28,39,40). In our study, as well as in a previous report with our method (11), no significant differences between the temperatures adjacent to the electrodes (T1 and T2) compared with temperatures at the center of the targeted tissue (T3) were observed globally in all groups. Hence, a similar pattern of energy dissipation occurred in any area of targeted tissue.
Evidence 5.—Finally, three distinguishable zones are observed at gross and histologic examination of pig liver on the 7th to 9th postoperative days after conventional approaches, including monopolar saline-enhanced RF ablation (41,43). Zone 1 is the central area of cavitation, with black and ragged borders of char surrounded by fibrin and hemorrhage. Zone 2 is a pale area containing shrunken hepatocytes with several grades of disruption. Zone 3 is also pale or whitish but with little hepatocyte disruption. This pathologic gradation through the lesion was not found with our method, regardless of the amount of energy delivered in the tissue; instead, the appearance was homogeneous. Pathologic examination in perfused groups A and C demonstrated features similar to those in nonperfused groups B and D, but the lesions were shaped irregularly, depending on the vessels encountered.
Several limitations of this study must be addressed.
Limitation 1.—The large volume of coagulation created may not always be beneficial or desired, and a detrimental effect in the surrounding tissues might occur. The majority of complications encountered in our study were attributed to failure to isolate the hyperthermic lesion. These complications emphasize the importance of proper electrode placement and remind us that caution should be exercised when large hepatic ablations are performed, especially with interstitial infusion of saline, in animal and clinical applications.
Limitation 2.—It is conceivable that the zone of necrosis obtained with our method will approximate an ellipse rather than a sphere. Results may be different in neoplastic tissue because of tissue composition. Although the height and width of the thermal lesion can be predicted by taking into account the time exposure and a good correlation between these two diameters, as was also described in our study (r = 0.77, P < .01), the volume of coagulation might not completely match the tumor geometry.
Limitation 3.—The predictability of thermal lesions achieved in normal perfused liver with our method in the current study is better than that achieved in our prior study (13), but it is still deemed suboptimal. The heat sink effect as a result of nearby vessels and several inhomogeneities in pig liver (too slim anteriorly and too segmented) can account for unpredictable lesions. Uneven diffusion of saline into the tissue can explain the rest of the inaccuracy. In the current study, the coefficient of variability of lesion size for Pringle groups B and D (23% and 22%, respectively) was even better than that (38%) in the study by Chinn et al (38) in similar conditions with use of expandable electrodes and the Pringle maneuver. In our study, non-Pringle group C had a coefficient of variability of 30%. This percentage is similar to those observed by De Baere et al (15) in a similar nonPringle pig liver model: 31% with a cool-tip electrode and 32% with an expandable electrode. In our study, however, the predictability of lesion necrosis with greater separation of the electrodes (4 cm) without the Pringle maneuver was less (75%), which led to lesion necrosis that was not completely predictable or was even unpredictable.
Limitation 4.—A theoretic limitation of any bipolar method is that the power dissipation of the two electrodes cannot be controlled independently in response to different conditions in the vicinity of each electrode (37). This may result in nonuniform lesions and decreased lesion size, which is why we recorded the specific voltage of both contacts of the electrode during each procedure. Nevertheless, we proved that an uneven electric contact between the two electrodes is not as relevant when the emitted power is low (usually less than 40 W).
Limitation 5.—The porcine model, which is currently used for RF ablation research, has been used extensively for experiments in this technique, but it has some limitations that must be considered to avoid substantial bias. To extrapolate observations to human clinical practice could be hazardous because of the unique vascular anatomy of pigs. In our study, for example, the Pringle maneuver was in place for about 8 minutes after which hepatic perfusion (and blood pressure) rapidly returned to normal. Longer occlusion times may be associated with hepatic infarction and death, as described by Denys (44). Furthermore, the relatively small size of porcine liver could explain some procedure-related complications in our study, especially when the distance between the electrodes was relatively large.
Limitation 6.—Saline infusion into a tumor has the theoretic possibility of increasing tumor cell seeding, which could result in an increased risk of local and distant recurrences. Nevertheless, this possibility is outside the scope of this article.
In conclusion, our bipolar saline-enhanced RF ablation system, embedded in only one needle (and preferably with vascular inflow occlusion), is consistently able to produce homogeneous and predictable areas of coagulation necrosis between the two electrodes, regardless of the distance between them.
Practical application: The method described in this study could be used either percutaneously or laparoscopically, preferably combined with vascular inflow occlusion, for the treatment of either small or large liver malignancies in a single application. Further experimental studies are needed to confirm these results before clinical application of the method.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, F.B.; study concepts, J.M.B., R.S., T.C., I.C., O.B.; study design, F.B., A.G., A.N., R.L.; literature research, F.B., A.N., A.G.; experimental studies, F.B., A.G., A.N., I.C., O.B.; data acquisition, A.G., A.N.; data analysis/interpretation, F.B., R.S., I.C., O.B.; statistical analysis, F.B., A.G.; manuscript preparation and editing, F.B., A.G., J.M.B.; manuscript definition of intellectual content, J.M.B., A.G., A.N., R.L.; manuscript revision/review and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Mulier, Y. Ni, L. Frich, F. Burdio, A. L. Denys, J.-F. De Wispelaere, B. Dupas, N. Habib, M. Hoey, M. C. Jansen, et al. Experimental and Clinical Radiofrequency Ablation: Proposal for Standardized Description of Coagulation Size and Geometry Ann. Surg. Oncol., April 1, 2007; 14(4): 1381 - 1396. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Clasen, D. Schmidt, A. Boss, K. Dietz, S. M. Krober, C. D. Claussen, and P. L. Pereira Multipolar Radiofrequency Ablation with Internally Cooled Electrodes: Experimental Study in ex Vivo Bovine Liver with Mathematic Modeling Radiology, March 1, 2006; 238(3): 881 - 890. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Lee, J. K. Han, S. H. Kim, S. H. Choi, S. K. An, C. J. Han, and B. I. Choi Bipolar Radiofrequency Ablation Using Wet-Cooled Electrodes: An In Vitro Experimental Study in Bovine Liver Am. J. Roentgenol., February 1, 2005; 184(2): 391 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J M Lee, J K Han, S H Kim, J Y Lee, S H Choi, and B I Choi Hepatic bipolar radiofrequency ablation using perfused-cooled electrodes: a comparative study in the ex vivo bovine liver Br. J. Radiol., November 1, 2004; 77(923): 944 - 949. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
