Super-micro-bland particle embolization combined with RF-ablation: Angiographic, macroscopic and microscopic features in porcine kidneys

Materials and methods In ten pigs, super-micro-bland particle embolization combined with RF-ablation was carried out. Super-micro-bland embolization was performed with spherical particles of very small size and tight calibration (40 ± 10 μm). In the left kidneys, RF-ablations were performed before embolization (I). In the right kidneys, RF-ablations were performed after embolization (II). The animals were killed three hours after the procedures. Angiographic (e.g. vessel architecture), macroscopic (e.g. long and short axes of the RF-ablations) and microscopic (e.g. particle distribution) study goals were defined. Results Angiography detected almost no vessels in the center of the RF-ablations in I. In II, angiography could not define the RF-ablations. Macroscopy detected significantly larger long and short axes of the RF-ablations in II compared to I (52.2 ± 3.2 mm vs. 45.3 ± 6.9 mm [ P < 0.05] and 25.1 ± 3.5 mm vs. 20.0 ± 1.9 mm [ P < 0.01], respectively). Microscopy detected irregular particle distribution at the rim of the RF-ablations in I. In II, microscopy detected homogeneous particle distribution at the rim of the RF-ablations. Microscopy detected no particles in the center of the RF-ablations in I and II. Conclusion The sequence of the different procedural steps of super-micro-bland particle embolization combined with RF-ablation impacts angiographic, macroscopic and microscopic features in kidneys in the acute setting. Keywords Radiofrequency ablation Embolization Particles Embozene Kidney 1 Introduction Among the different ablative tumor therapies, radiofrequency- (RF-) ablation is clinically established, effective and safe [1] . Especially cancer patients barely accessible to open surgery profit from this approach. To focus on RF-ablation for malignancies in the kidney, recurrence free survival rates range between 71% and 100% [2,3] . In small, well-circumscribed tumors, RF-ablation can be as effective as partial nephrectomy [4] . However, in bigger (e.g. beyond a diameter of 3 cm) or infiltrating (e.g. pelvic system) tumors, local tumor control rates decrease significantly [5,6] . Therefore, additional strategies have to be found to effectively treat advanced cancer in a minimally invasive fashion. The combination of transarterial embolization and RF-ablation is a concept to combine oncological synergies. Animal experiments in livers detected bigger and more spherical RF-ablations after embolization [7,8] . In vivo porcine kidney studies demonstrated larger RF-ablations with less variation in size after transarterial embolization [9,10] . Clinical studies described the combination of transarterial embolization and subsequent RF-ablation as feasible with local control rates up to 100% [3,11] . In the reports on the combination of embolization and RF-ablation, virtually always embolization was performed as the first procedural step. However, no systematic analysis was found comparing the differences when the sequence of the single procedural steps is changed: embolization with subsequent RF-ablation versus RF-ablation with subsequent embolization. The rational for this comparison might be unclear at first view. The regular sequence in the clinical routine constitutes embolization with subsequent RF-ablation [3,11] . However, the opposite approach with RF-ablation with subsequent embolization might be also beneficial. For example, RF-ablation creates a reactive hyperperfused rim surrounding the ablation zone. Subsequent free-flow controlled embolization with small particles might concentrate the ischemic effect into the hyperperfused rim and, if the RF-ablation was positioned properly, also into the hardly destructible tumor periphery. Additionally, the available studies used older embolization materials such as iodized oil and non-spherical, uncalibrated particles [7–9,12] . In this context, the purpose of this study was defined: To describe angiographic, macroscopic and microscopic features of super-micro-bland particle embolization combined with RF-ablation in porcine kidneys with a special focus on the impact of the sequence of the different procedural steps. Thereby, super-micro-bland embolization was performed with spherical particles of a very small size (40 μm) and tight calibration (±10 μm). 2 Materials and methods The study was performed in accordance with the Guide for the Care and Use of Laboratory Animals and approved by our State Animal Care and Ethics Committee. 2.1 Experimental set-up 2.1.1 Study animals In ten Landrace pigs (body weight between 30 and 35 kg), super-micro-bland particle embolization combined with RF-ablation was carried out. After sedation with an intramuscular cocktail consisting of 10 mg ketamine and 1 mg midazolam per kg of body weight, general anesthesia was induced and maintained with isoflurane. The femoral artery was surgically exposed and punctured with a Seldinger needle (Super 4 Needle; Bard, Karlsruhe, Germany). After introduction of a 0.35“guidewire (PTFE-coated Guidewire J-curved; Optimed, Ettlingen, Germany), a 4F introducer sheath was placed (Pinnacle; Terumo, Tokyo, Japan). Subsequently, the abdomen was surgically opened and the kidneys dissected. Thereby, extreme care was taken to avoid unnecessary manipulation at the hilum. 2.1.2 Super-micro-bland particle embolization combined with RF-ablation Selective angiograms of the left and right renal artery were performed (field of view 14 cm, matrix 1024 × 1024, image frame 3 images/s; Polystar Top, Siemens, Forchheim, Germany) applying a 4F Berenstein catheter (Cook, Bloomington, USA). After removal of the catheter, a RF-ablation of the left kidney was carried out (I). RF-generator parameters included an ablation time of 8 min, an ablation temperature of 105 °C and an ablation power of 150 W (Model 1500X RF Generator; AngioDynamics, Queensbury, USA). A monopolar RF-probe was used as ablation device with manually adjustable antennas with a maximum expansion of 5 cm (StarBurst XL; AngioDynamics, Queensbury, USA). A special method was used to position the RF-probe in the kidney. With retracted antennas, the shaft of the RF-probe penetrated the renal tissue perpendicular to its surface and along the long axis of the organ. The shaft was advanced to a depth of approximately 2.5 cm and the antennas expanded to 3 cm during shaft fixation. Finally, the shaft was retracted in such a manner, that the remaining 2 cm of the antennas were expanded. This resulted in a maximum antenna expansion of 5 cm, however with a more ellipsoid than spherical configuration ( Fig. 1 A) . The rational for this proceeding was to create relatively large RF-ablations in the relatively small porcine kidneys, since large RF-ablations are of great importance for human tumor destruction. A selective angiogram of the left renal artery followed. Three hours later, super-micro-bland particle embolization of the entire left and right kidney was performed with a 2.8F microcatheter (Progreat; Terumo, Tokyo, Japan) advanced over the Berenstein catheter into the renal artery. As embolization material, spherical particles of a very small size (40 μm) and tight calibration (±10 μm) (Embozene 40; CeloNova BioSiences, Peachtree City, USA) were used. To get the optimal particle suspension, the recommendations of the manufacturer were followed accurately. The embolization procedure itself was performed at least over 25 min until complete stasis of the segmental and subsegmental renal arteries was observed. This happened after selective injection of two to four milliliters of pure particles. A RF-ablation of the right kidney followed three hours later comparable to the left side (II). Finally, selective angiograms of the left and right renal artery were performed. 2.1.3 Pathology Three hours later, the animals were killed. The kidneys were harvested and preserved in formalin 4% for 4 weeks to achieve adequate tissue fixation. With a sausage cutter (Futura F1; Graef, Arnsberg, Germany), the kidneys were cut in 2–3 mm thin axial macroscopic slices. Representative samples through the RF-ablation and through the non-ablated kidney were embedded in paraffin, cut in microscopic slices and stained with hematoxylin–eosin. Macroscopic and microscopic slices were digitized. 2.2 Study goals Angiographic, macroscopic and microscopic study goals were defined. Angiographic study goals : Selective angiograms of the renal arteries were used to evaluate the angiographic study goals. The major appearance of the vessel architecture in the center, at the rim and outside of the RF-ablations was described after the first and second procedural step. Macroscopic study goals : Digitized macroscopic slices uploaded with an image processing software were used to evaluate the macroscopic study goals (Adobe Photoshop 10.0.1; Adobe Systems Incorporated, San Jose, USA). Qualitative study goals : Those included the description of the macroscopic morphology (hemorrhagic rim infiltrates, demarcation, shape and homogeneity) in the center, at the rim and outside of the RF-ablations. Quantitative study goals : Those included the determination of the long axis, short axis, volume and circularity of the RF-ablations. Thereby, only the yellow-white zone, the coagulation necrosis, was considered [14,15] . A red-brown zone surrounding the coagulation necrosis, the hemorrhagic rim infiltrate, was not included in the measurements [7,15,16] . The long axis (in mm) was defined as the longest diameter of the coagulation necrosis along the axis of the RF-probe. The number of axial macroscopic slices on which the coagulation necrosis occurred was counted and multiplied with the mean axial macroscopic slice thickness of 2.5 mm. The short axis (in mm) was defined as the shortest diameter of the largest coagulation necrosis circumscribable on an axial macroscopic slice. The volume (in ml) was determined [17] : (1) Volume = 4 π 3 × long     axis × short   axis 2 8000 . The circularity was calculated to describe the roundness of the RF-ablations [18] : (2) Circularity = short     axis long     axis . Thereby, a perfect circle has a circularity of 1. Microscopic study goals : Digitized microscopic slices uploaded with an image processing software were used to evaluate the microscopic study goals (Adobe Photoshop 10.0.1; Adobe Systems Incorporated, San Jose, USA). Those included the description of the tissue characteristics and major particle characteristics (distribution and morphology) in the center, at the rim and outside of the RF-ablations. 2.3 Statistical analysis Prism software (Version 4.00, GraphPad Software, LaJolla, USA) was used to calculate the statistics. Descriptive statistics of the quantitative data is presented as mean ± standard deviation as well as range. Comparative statistics of the quantitative data for I and II was performed with the non-parametric Wilcoxon signed rank test. The reproducibility of the RF-ablations was analyzed applying crossed variance component analysis by F -test. Thereby, the variance was defined as follows: (3) Variance = ( standard     deviation ) 2 . P < 0.05 was regarded as statistically significant. 3 Results In all animals, super-micro-bland particle embolization combined with RF-Ablation was finished as projected. Complications or adverse effects did not occur. 3.1 Angiographic study goals In I, RF-ablation was performed as first procedural step. Thereby, angiography detected almost no vessels in the center of the RF-ablations as well as irregular and spastic vessels at the rim of the RF-ablations ( Table 1 ) ( Fig. 1 B). Super-micro-bland embolization followed as second procedural step. Thereby, a more inhomogeneous embolization of the segmental and subsegmental renal arteries at the rim of the RF-ablations was observed ( Fig. 1 C). In II, super-micro-bland embolization was performed as first procedural step. Thereby, angiography detected homogeneous and complete embolization of the segmental and subsegmental renal arteries ( Table 1 ) ( Fig. 1 D and E). RF-ablation followed as second procedural step. Thereby, no additional angiographic aspects were found, especially the RF-ablations were not definable ( Fig. 1 E). 3.2 Macroscopic study goals 3.2.1 Qualitative study goals In I, hemorrhagic rim infiltrates around the RF-ablations existed regularly, the demarcation of the RF-ablations was unsharp, the shapes of the RF-ablations were asymmetrical and the homogeneity of the RF-ablations was intermediate ( Table 1 ) ( Fig. 2 A) . In II, hemorrhagic rim infiltrates around the RF-ablations existed exceptionally, the demarcation of the RF-ablations was sharp, the shapes of the RF-ablations were symmetrical and the homogeneity of the RF-ablations was high ( Table 1 ) ( Fig. 2 B). 3.2.2 Quantitative study goals Long and short axes as well as volume of the RF-ablations were significantly larger in II compared to I (52.2 ± 3.2 mm vs. 45.3 ± 6.9 mm [ P < 0.05], 25.1 ± 3.5 mm vs. 20.0 ± 1.9 mm [ P < 0.01], and 17.4 ± 4.8 ml vs. 10.2 ± 3.4 ml [ P < 0.01], respectively) ( Table 2 ) ( Fig. 3 A–D) . The circularity was comparable in I and II (0.45 ± 0.16 vs. 0.48 ± 0.05 [n.s.]). Only the long axis of the RF-ablations was significantly more reproducible in II compared to I (10.2 vs. 47.8 [ P < 0.05]) ( Table 3 ). 3.3 Microscopic study goals 3.3.1 Tissue characteristics In I, cell lysis and cell damage were detected in the center of the RF-ablations ( Table 1 ). Whereas intense and broad hemorrhagic infiltrates were found at the rim of the RF-ablations, no cell lysis but early signs of cell death and some hemorrhagic infiltrates were observed outside of the RF-ablations. At the rim of the RF-ablation with orientation more to the center of the RF-ablation, the arteries were thrombosed ( Fig. 4 A–E ). In II, cell lysis and cell damage were detected in the center of the RF-ablations ( Table 1 ). Whereas few and narrow hemorrhagic infiltrates were found at the rim of the RF-ablations, no cell lysis but early signs of cell death with some hemorrhagic infiltrates were observed outside of the RF-ablations ( Fig. 4 F–J). 3.3.2 Major particle characteristics In I, no particles were detected in the center of the RF-ablations ( Table 1 ). Whereas a more irregular particle distribution was found at the rim of the RF-ablations, a homogeneous particle distribution was observed outside of the RF-ablations. At the rim of the RF-ablations, the particles were more distant to the RF-ablation. In II, no particles were detected in the center of the RF-ablations. A homogeneous particle distribution was detected at the rim and outside of the RF-ablations. The particles were close to the RF-ablations. In I and II, the particles were round and intact. 4 Discussion In this study, the sequence of the different procedural steps of super-micro-bland particle embolization combined with RF-ablation impacted angiographic, macroscopic and microscopic features in porcine kidneys. Thereby, super-micro-bland embolization as first procedural step with subsequent RF-ablation created markedly sharper, more symmetrical, and more homogeneous as well as significantly larger RF-ablations compared to RF-ablation as first procedural step with subsequent super-micro-bland embolization. 4.1 Angiographic study goals There is a lack of information in the literature on the appearance of the vessels after RF-ablation. Nakai et al. demonstrated occluded small portal vein branches after RF-ablation of porcine livers with a compensatory increase of the hepatic arterial flow [8] . However, after additional arterial embolization, even the larger portal vein branches disappeared. Our results demonstrated, that RF-energy has the power to affect non-embolized arteries. While the absence of the vessels in the center of the RF-ablations corresponded most likely vessel destruction and thrombosis, the vessels at the rim of the RF-ablations seemed to have also a reversible spastic component [13,19,20] . This is in line with a more inhomogeneous embolization at the rim of the RF-ablations [21] . On the other side, when super-micro-bland embolization was performed as first procedural step, final angiography could not define the RF-ablations. This suggested a very high potential of super-micro-bland embolization for effective acute vessel occlusion in the kidney [22] . 4.2 Macroscopic study goals 4.2.1 Qualitative study goals When RF-ablation was performed as first procedural step with subsequent super-micro-bland embolization, the overall appearance of the lesions corresponded more to either in vivo lesions in kidneys and livers without interruption of the perfusion or in vivo lesions in livers with occlusion of either the portal vein or the hepatic artery with patent flow in the other vessel [7–10,23] . Blood flow in a renal artery creates hyperperfusion with hemorrhage at the rim of the RF-ablations in the acute phase [10,24] . Transferable results were found in in vivo porcine liver studies. Chung et al. described a narrower zone of hemorrhagic rim infiltrates in RF-ablations after transarterial embolization (iodized oil) compared to RF-ablations with normal flow [7] . After interruption of the hepatic flow using transarterial embolization (degradable starch microspheres) or Pringle's maneuver, hemorrhagic rim infiltrates were minimal or even absent [23] . Iwamoto et al. demonstrated iodized oil accumulation outside the RF-ablations after embolization of the hepatic artery with a combination of iodized oil and gelatine sponge particles [12] . These aspects point out, that the type of embolization material matters (material [e.g. iodized oil, gelatine, starch], formulation [e.g. liquid, particulate], shape [non-spherical, spherical] and size [macro, micro]). When super-micro-bland embolization was performed as first procedural step, the overall appearance of the RF-ablations is similar more to ex vivo lesions in kidneys and livers. The RF-ablations were very homogeneous and sharply demarcated as shown previously [19] . After interruption of the blood flow, hemorrhagic rim infiltrates were exceptional and very few [8,9] . Moreover, the shapes of the RF-ablations were more symmetrical [7,10] . Acute vessel occlusion from super-micro-bland embolization may be the reason for the decrease in hemorrhagic rim infiltrates. 4.2.2 Quantitative study goals Our results were in line with reports demonstrating significantly larger RF-ablations after reduction of the tissue perfusion [19,21] . The major explanation for this is minimization of the “vessel heat-sink effect”. The kidney is an organ with a very high vascular density and therefore high perfusion affects RF-ablation geometry [25,26] . In an in vivo porcine kidney study, Chang et al. concluded that vascular clamping and transarterial embolization can effectively overcome the cooling effects of the blood flow [9] . Thereby, the RF-ablation width measured 0.86 cm in kidneys with normal flow compared to 1.4 cm in kidneys after clamping ( P < 0.001). Another pig kidney study demonstrated significantly larger RF-ablation volumes as well as rounder lesions after transarterial embolization compared to the control group [10] . In our study, roundness and reproducibility of the RF-ablations were comparable in both study groups. A possible explanation for this might be the aggressive parameters used to create our RF-ablations (e.g. long ablation time, maximum antenna expansion of 5 cm). As we know from some publications, the study design (animals, target organs, RF-systems and method of vessel occlusion [e.g. clamping or embolization, embolization material, type of occluded vessel (e.g. renal artery, portal vein)]) can affect the outcome of RF-ablations [10] . 4.3 Microscopic study goals 4.3.1 Major tissue characteristics The basical histopathological features of RF-ablations in kidneys are well-known. These correspond to our observations when RF-ablation was performed as first procedural step. In the center of the RF-ablations, cell lysis (loss of cell integrity) and cell damage (frayed and indistinct cell boarders) occurred [13,27] . At the rim of the RF-ablations, intense and broad hemorrhagic infiltrates were found [24,27] . Outside of the RF-ablations, super-micro-bland embolization-induced early signs of cell death (blurred nuclear chromatin and increased eosinophilia of the cytoplasm) were detected [16,27] . The major differences compared to the other study group with super-micro-bland embolization with subsequent RF-ablation represented the hemorrhagic infiltrates. Those were few and narrow at the rim of the RF-ablations. Some authors define the hemorrhagic rim infiltrate as potentially viable [24] . Marcovich et al. detected viable rims in 80% of the RF-ablations performed in non-ischemic kidneys, compared to 40% in ischemic kidneys [16] . In another study, hemorrhagic rim infiltrates were discreet after renal artery embolization (large, imprecisely calibrated spherical particles) [9] . The degree of hemorrhagic infiltrates might depend on the mechanism of vessel occlusion. While hilar clamping can result in backward bleeding from blood-filled and congested veins, exsanguination of the kidney with occlusion of the supplying arteries (but unaffected venous drainage) should minimize hemorrhagic infiltrates. 4.3.2 Major particle characteristics When super-micro-bland embolization was performed as first procedural step, homogeneous particle distribution was observed at the rim and outside of the RF-ablations. Particles were detected close to the RF-ablations. This fits to another report attributing this favourable characteristic to the uniformity of the particle size [22,28] . The standardized level of vessel occlusion is important to minimize perfusion-mediated “vessel heat-sink effects”. This should result in larger as well as more predictable RF-ablations [10] . When RF-ablation represented the first procedural step, the particle distribution was irregular at the rim of the RF-ablations [20,21] . Particles were more distant to the RF-ablations. This could have been triggered by thrombosis, vascular destruction as well as reversible spasm as observed by other groups [21] . In the center of the RF-ablations, no particles were found in both study groups. Whether this is due to particle instability to the high ablation temperatures, or just a phenomenon in line with the complete tissue destruction (with loss of the leading structures important to identify the particles), is unclear. The morphology of the particles was comparable in both study groups. At the rim and outside of the RF-ablations, the particles were round and intact. Therefore, the type of embolization material used in this study seems to be effective for embolization in combination with RF-ablation. 4.4 Study limitations Several limitations have to be listed. First, this experimental trial evaluated effects of super-micro-bland embolization and RF-ablation in non-tumor-bearing kidneys of the pig and not in primary or secondary cancers of human beings. Second, our method to position the RF-probe in the kidney is not standard use. The protocols of the manufacturer are designed for regular expansion of the antennas. However, we selected this way since we wanted to create large RF-ablations in relatively small pig kidneys. Accordingly, the RF-ablation dimensions and shapes were different compared to the specifications of the manufacturer and to other reports using comparable system parameters. Third, the microscopic study goals were evaluated in a qualitative fashion with light microscopy in selected tissue samples. Additional stainings (e.g. nicotinamide adenine dinucleotide or triphenyl tetrazolium chloride for light microscopy as well as uranyl acetate/lead citrate for electron microscopy) and semi-quantitative or even quantitative analysis of tissue and particle characteristics would have given more specific information [16,22,29] . Finally, all study goals were evaluated in an acute setting. Survival trials are necessary to get important information on the long-term outcome such as shrinkage, inflammation, encapsulation and skip lesions. To conclude, the sequence of the single procedural steps of super-micro-bland particle embolization combined with RF-ablation impacts angiographic, macroscopic and microscopic features in porcine kidneys. Thereby, super-micro-bland embolization as first procedural step with subsequent RF-ablation created markedly sharper, more symmetrical, and more homogeneous as well as significantly larger RF-ablations compared to RF-ablation as first procedural step with subsequent super-micro-bland embolization. Funding No funding. Conflict of interest The authors declare that they have no conflict of interest. References [1] A. Gillams Tumour ablation: current role in the kidney, lung and bone Cancer Imaging 9 Spec No A 2009 S68 70 [2] J.S. Lewin S.G. Nour C.F. Connell Phase II, clinical trial of interactive MR imaging-guided interstitial radiofrequency thermal ablation of primary kidney tumors: initial experience Radiology 232 2004 835 845 [3] A.H. Mahnken D. Rohde D. Brkovic R.W. Gunther J.A. Tacke Percutaneous radiofrequency ablation of renal cell carcinoma: preliminary results Acta Radiol 46 2005 208 214 [4] J.M. Stern R. Svatek S. Park Intermediate comparison of partial nephrectomy and radiofrequency ablation for clinical T1a renal tumours BJU Int 100 2007 287 290 [5] E.E. Rutherford J.E. Cast D.J. Breen Immediate and long-term CT appearances following radiofrequency ablation of renal tumours Clin Radiol 63 2008 220 230 [6] J. Tacke Interventional oncology in urology Radiologe 47 2007 1089 1096 [7] H. Chung M. Kudo Y. Minami T. Kawasaki Radiofrequency ablation combined with reduction of hepatic blood flow: effect of lipiodol on coagulation diameter and ablation time in normal pig liver Hepatogastroenterology 54 2007 701 704 [8] M. Nakai M. Sato S. Sahara Radiofrequency ablation in a porcine liver model: effects of transcatheter arterial embolization with iodized oil on ablation time, maximum output, and coagulation diameter as well as angiographic characteristics World J Gastroenterol 13 2007 2841 2845 [9] I. Chang I. Mikityansky D. Wray-Cahen W.F. Pritchard J.W. Karanian B.J. Wood Effects of perfusion on radiofrequency ablation in swine kidneys Radiology 231 2004 500 505 [10] c.M. Sommer N. Kortes S. Zelzer Renal artery embolization combined with radiofrequency ablation in a porcine kidney model: effect of small and narrowly calibrated microparticles as embolization material on coagulation diameter, volume, and shape Cardiovasc Intervent Radiol 2010 [Epub ahead of print] [11] K. Arima K. Yamakado H. Kinbara A. Nakatsuka K. Takeda Y. Sugimura Percutaneous radiofrequency ablation with transarterial embolization is useful for treatment of stage 1 renal cell carcinoma with surgical risk: results at 2-year mean follow up Int J Urol 14 2007 585 590 [12] T. Iwamoto N. Kawai M. Sato Effectiveness of hepatic arterial embolization on radiofrequency ablation volume in a swine model: relationship to portal venous flow and liver parenchymal pressure J Vasc Interv Radiol 19 2008 1646 1651 [13] J.D. Crowley J. Shelton A.J. Iverson M.P. Burton N.C. Dalrymple J.T. Bishoff Laparoscopic and computed tomography-guided percutaneous radiofrequency ablation of renal tissue: acute and chronic effects in an animal model Urology 57 2001 976 980 [14] S.N. Goldberg G.S. Gazelle C.C. Compton P.R. Mueller K.K. Tanabe Treatment of intrahepatic malignancy with radiofrequency ablation: radiologic-pathologic correlation Cancer 88 2000 2452 2463 [15] F. Burdio A. Navarro R. Sousa Evolving technology in bipolar perfused radiofrequency ablation: assessment of efficacy, predictability and safety in a pig liver model Eur Radiol 16 2006 1826 1834 [16] R. Marcovich J.P. Aldana N. Morgenstern A.I. Jacobson A.D. Smith B.R. Lee Optimal lesion assessment following acute radio frequency ablation of porcine kidney: cellular viability or histopathology? J Urol 170 2003 1370 1374 [17] J.M. Lee J.K. Han S.H. Kim In vivo efficiency of multipolar radiofrequency ablation with two bipolar electrodes: a comparative experimental study in pig kidney J Vasc Interv Radiol 18 2007 1553 1560 [18] P.F. Laeseke F.T. Lee Jr. L.A. Sampson D.W. van der Weide C.L. Brace Microwave ablation versus radiofrequency ablation in the kidney: high-power triaxial antennas create larger ablation zones than similarly sized internally cooled electrodes J Vasc Interv Radiol 20 2009 1224 1229 [19] A. Hacker F. Risse K. Peters Magnetic resonance imaging for assessment of radiofrequency lesions in kidney tissue immediately after ablation: an experimental study J Endourol 20 2006 312 317 [20] H.U. Kasper C. Bangard A. Gossmann H.P. Dienes D.L. Stippel Pathomorphological changes after radiofrequency ablation in the liver Pathol Int 60 2010 149 155 [21] E.J. Patterson C.H. Scudamore D.A. Owen A.G. Nagy A.K. Buczkowski Radiofrequency ablation of porcine liver in vivo: effects of blood flow and treatment time on lesion size Ann Surg 227 1998 559 565 [22] S. Stampfl N. Bellemann U. Stampfl Arterial distribution characteristics of Embozene particles and comparison with other spherical embolic agents in the porcine acute embolization model J Vasc Interv Radiol 20 2009 1597 1607 [23] J.P. Ritz K.S. Lehmann C. Isbert In-vivo evaluation of a novel bipolar radiofrequency device for interstitial thermotherapy of liver tumors during normal and interrupted hepatic perfusion J Surg Res 133 2006 176 184 [24] S.Y. Nakada T.J. Jerde T.F. Warner Bipolar radiofrequency ablation of the kidney: comparison with monopolar radiofrequency ablation J Endourol 17 2003 927 933 [25] M.G. Lubner C.L. Brace J.L. Hinshaw F.T. Lee Jr. Microwave tumor ablation: mechanism of action, clinical results, and devices J Vasc Interv Radiol 21 2010 S192 203 [26] L.R. Williams R.W. Leggett Reference values for resting blood flow to organs of man Clin Phys Physiol Meas 10 1989 187 217 [27] T.H. Hsu M.E. Fidler I.S. Gill Radiofrequency ablation of the kidney: acute and chronic histology in porcine model Urology 56 2000 872 875 [28] S. Stampfl N. Bellemann U. Stampfl Inflammation and recanalization of four different spherical embolization agents in the porcine kidney model J Vasc Interv Radiol 19 2008 577 586 [29] Y. Gao Y. Wang Y. Duan 915 MHz microwave ablation with high output power in in vivo porcine spleens Eur J Radiol 75 2010 87 90