| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Division of MR Imaging, University Hospitals of
Cleveland/Case Western Reserve University, 1110 Euclid Ave., Cleveland, OH
44106.
2
Department of Diagnostic Radiology, University of Ulm, Ulm 89075,
Germany.
3
Department of Urology, University Hospitals of Cleveland/Case Western Reserve
University, Cleveland, OH 44106.
4
Department of Biomedical Engineering, Case Western Reserve University,
Cleveland, OH 44106.
5
Department of Pathology, University Hospitals of Cleveland/Case Western
Reserve University, Cleveland, OH 44106.
6
Department of Oncology, University Hospitals of Cleveland/Case Western Reserve
University, Cleveland, OH 44106.
Received August 8, 2000;
accepted after revision January 24, 2001.
The University Hospitals of Cleveland/Case Western Reserve University
Interventional MR Program is supported in part through research collaborations
with Siemens Medical Systems and Radionics, Inc. This project was also
supported through grants from the Whitaker Foundation, American Cancer
Society, National Institutes of Health (1R01-CA81431-01A1), and the German
Research Foundation (DFG, As 116/1-1).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Twelve kidneys were treated in six pigs that were placed under general anesthesia in the MR suite, using a 0.2-T open C-shaped MR imaging system. A 4-cm-long, 14-mm-diameter balloon catheter was placed into the aorta using a transfemoral approach, and the balloon was positioned proximal to the renal arteries via guidance with MR imaging. A 2-cm exposed-tip MR-compatible 17-gauge radiofrequency electrode was placed into one kidney under MR fluoroscopy using fast imaging with steady-state free precession (FISP) sequences. Thermal ablation was performed with the electrode tip temperature maintained at 90 ± 2°C for 10 min. This procedure was repeated in the contralateral kidney. The balloon was inflated during one ablation. Postablation images were obtained, the pigs were sacrificed, and both kidneys of each animal were harvested for pathologic correlation.
RESULTS. Technical success was achieved in all animals. The lesion measured 14.2 ± 2.2 mm (mean ± standard deviation) for the ischemic kidney versus 8.0 ± 2.6 mm in the normally perfused kidney (p = 0.00002). No significant complications were noted. In all images, thermal lesions displayed low signal intensity with a sharp rim of high signal intensity best visualized using short tau inversion recovery (STIR) sequences with a mean accuracy of 1.3 ± 1.2 mm when compared with pathologic findings and a mean contrast-to-noise ratio of 4.9 ± 2.5.
CONCLUSION. We accept the hypothesis that temporary renal ischemia leads to a significantly increased radiofrequency ablation lesion size. We conclude that catheter-based balloon perfusion reduction is feasible, that the procedure does not lead to major complications, and that it can be performed using MR imaging as the sole imaging modality.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The size of the resulting necrosis is determined by various factors that have been extensively studied, including probe size, gauge, temperature, and ablation duration [6]. Advances have been made in the design of the probe and the radiofrequency generator to increase the tissue volume that can be ablated [7,8,9]. Still, larger target volumes are desirable, although one of the advantages of radiofrequency ablation over other treatment modalities, such as radiation therapy, is the option of an indefinite number of repeated applications [10]. One limiting factor for thermal damage is the perfusion of organs working essentially as a heat sink. Therefore, techniques that decrease perfusion of the target tissue will likely add to the effect of previously described developments [11].
Using CT for guidance, Goldberg et al. [11] proved in a porcine model that a decrease in liver perfusion leads to an increase in lesion size. To our knowledge, no use of perfusion modulation in kidney ablation has been reported in the current literature, despite the fact that the kidney is a target organ that is ideally suited for minimally invasive thermal ablation.
This study was performed to test the hypothesis that temporary renal ischemia will result in increased thermal lesion size during interstitial radiofrequency thermal ablation in the kidney.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animal Model
Using a protocol approved by the institutional animal care and use
committee, 12 kidneys were treated in six pigs that were placed under general
anesthesia. Each of the six farm pigs (males, 20-25 kg) was anesthetized using
a combination of acepromazine (0.25 mg/kg of body weight) and ketamine (7.5
mg/kg of body weight) by intramuscular injection. A 20-gauge, 1-inch (2.5 cm)
IV catheter (Surflo; Terumo Medical, Elkton, MD) was then inserted into a
dorsal ear vein, and thiopental sodium (15 mg/kg of body weight) was
administered IV to allow tracheal intubation. Mechanical ventilation was used
for the duration of the procedure, and anesthesia was maintained with repeated
doses of thiopental sodium. Each animal's lateral hindquarters were shaved
bilaterally, and one 8 x 12 cm wire-mesh grounding pad coated with
conductive gel was placed on each hind limb. Each pig was placed in a supine
position on the MR scanner table to allow dorsolateral access to the kidneys
through the open magnet.
A surgical cutdown was performed to expose the right femoral artery. An 8-French (2.7 mm) sheath (Input PS Percutaneous Catheter Introducer Set; Bard USCI Division, Billerica, MA) was inserted using Seldinger technique with the set's 0.038-inch (0.97-mm) wire and 5-French dilator. Then, a 0.035-inch 145-cm-long, 15-mm-diameter, J-shaped-tip Teflon-coated guidewire (Bard USCI Division, Tewskbury, MA) was carefully advanced into the descending aorta. A 5.8-French (1.9-mm) balloon dilatation catheter with a 4-cm-long, 14-mm-diameter balloon (XXL/14-4/5.8/75; Boston Scientific/Medi-Tech, Watertown, MA) was advanced to roughly the level of the renal arteries under MR-fluoroscopic visualization in the coronal plane using the FISP sequence described earlier. The artifacts from small metal rings at both ends of the balloon were readily identifiable and used to position the balloon so that the distal end of the balloon was immediately proximal to the renal artery origin from the aorta. The guidewire was removed. The balloon was inflated with 4 mL of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) diluted with saline (gadopentetate dimeglumine concentration 2 µm/mL) and then was deflated under continuous MR imaging to verify the position (Fig. 1).
|
Radiofrequency Ablation
Random targets for ablation were chosen in the renal cortex of either the
upper or lower pole using axial T2-weighted turbo spin-echo images (2405/96;
number of excitations, 3; echo train length, 7; slice thickness, 6 mm; field
of view, 250 mm2; matrix size, 196 x 256; acquisition time, 3
min 49 sec).
The electrode insertion site on the skin surface relative to the target was visualized using a fluid-filled syringe and the FISP sequences described earlier. A 2-cm exposed-tip custom-built MR-compatible 17-G radiofrequency electrode (Radionics, Burlington, MA) was placed into the pole of the kidney under further continuous MR fluoroscopy using FISP sequences (three parallel 5-mm slices centered on the predicted probe path). After placement of the electrode, the tip position was confirmed using a T1-weighted turbo spin-echo sequence (500/24; number of excitations, 3; echo train length, 5; slice thickness, 5 mm; field of view, 250 mm2; matrix size, 128x256; acquisition time, 1 min 19 sec) because turbo spin-echo sequences are superior to gradient-echo sequences in depicting accurate tip position [12].
Thermal ablation was then performed with the electrode tip temperature maintained at 90 ± 2°C by applying radiofrequency for 10 min using a 150-W radiofrequency generator operating at 500 kHz (modified model CC1; Radionics). This process was repeated in the contralateral kidney. The balloon was inflated during one ablation and deflated during the other in randomized order.
The electrode was withdrawn, and postablation images were obtained. Parameters for axial and coronal T1-weighted spin-echo sequences were 624/26; number of excitations, 3; slice thickness, 6 mm; field of view, 250 mm2; matrix size, 192x256; and acquisition time, 6 min 2 sec for 13 images. Parameters for T2-weighted turbo spin-echo sequences were 2405/96; number of excitations, 3; echo train length, 7; slice thickness, 6 mm; field of view, 250 mm2; matrix size, 196x256; and acquisition time, 3 min 49 sec for nine images. Parameters for turbo short tau inversion recovery (STIR) were 3874/48; inversion time, 110; number of excitations, 4; echo train length, 7; slice thickness, 6 mm; field of view, 250 mm2; matrix size, 140 x 256; and acquisition time, 5 min 19 sec for nine images. These sequences were chosen on the basis of prior work that showed the usefulness of T1-weighted, T2-weighted, and STIR images for visualizing thermally induced lesions [4, 5, 13, 14].
After the last scan, the animals were sacrificed using an overdose of IV thiopental sodium, and all kidneys were harvested for gross pathologic and histologic examinations. The formalin-fixed kidneys were cut in the axial plane perpendicular to the radiofrequency electrode tract, representative slices containing the radiofrequency lesion were digitally photographed, and the lesion diameters were measured using rulers. These slices were then embedded in paraffin, and 6-µm sections were serially cut and stained with H and E.
Image Analysis
To evaluate the relative visibility of the thermal lesions in each
postablation image, we first displayed each image on a freestanding
workstation (Magicview; Siemens). For each lesion, two regions of interest
were identified interactively by one of the investigators using the Evaluate
Statistics routine of the workstation's software (Numaris 3; Siemens). These
regions of interest were the core of the thermal lesion—ablated tissue
at the center of the affected area—and unaffected kidney tissue near the
lesion. Regions of interest were chosen in homogeneous, artifact-free areas
and made at the same anatomic level for each image sequence. Mean signal
amplitudes from each region of interest were recorded three times, and the
results were averaged. Regions of interest for signal amplitudes of tissues
included at least 20 pixels (
70 mm3). Regions of interest for
signal amplitudes of noise included at least 500 pixels (1735
mm3). Mean background noise was measured on each image ventral to
the kidneys, outside the body, along the phase-encoding axis.
Contrast between the regions of interest showing healthy kidney tissue and regions of interest showing the core of the thermal lesion was expressed in terms of contrast-to-noise ratio, calculated as the difference between the mean signal amplitudes from the two regions of interest, divided by the standard deviation of the noise. The direction of the difference calculation was selected to be that which would yield a positive value for the group mean for each particular contrast-to-noise ratio.
Statistical Analysis
Measurements of lesion diameter taken at gross pathology were compared with
the diameters obtained at MR imaging for both treatment groups—perfused
and nonperfused—in terms of mean difference, minimum and maximum
differences, standard deviations, and over- or underestimation.
Contrast-to-noise ratios among various MR imaging sequences were compared for both groups in terms of their minimum and maximum values, means, and standard deviations. Comparisons between specific values and between the two groups were performed with the Student's t test. A p value of less than 0.05 was considered significant.
The effect of perfusion on impedance and radiofrequency current measurements was tested using a repeated-measures analysis of variance with occluded versus perfused as the first within-subjects factor and time as a second within-subjects factor.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Radiofrequency current during the ablation averaged 325 ± 38 mA (mean ± standard deviation) in the perfused kidneys versus 254 ± 15 mA in the nonperfused kidneys. The average current applied in the first minute of the ablation was 402 ± 115.9 mA in the perfused kidneys versus 292 ± 90.9 mA in the nonperfused kidneys; the average current applied in the 10th minute of the ablation was 27 ± 74.8 mA in the perfused kidneys versus 239 ± 75.8 mA in the nonperfused kidneys (Fig. 2).
|
The baseline impedance averaged 83 ± 23.1 
in the group of
normally perfused kidneys versus 94±25.7 in the nonperfused
kidneys. The impedance dropped by an average of 5.0 ± 5.6 in
the first minute to an average of 77 ± 20.9 in the perfused
kidneys; it dropped by an average of 6.8 ± 3.3 to 87 ± 25.7
in the non-perfused kidneys. The impedance then leveled out to an
average of 82±22.7 in the perfused kidneys versus 87 ±
26.3 in the nonperfused kidneys
(Fig. 3).
|
The effect of perfusion on impedance and radiofrequency current measurements was tested using a repeated-measures analysis of variance with occluded versus perfused as the first within-subjects factor and time as a second within-subjects factor. Reduction in radiofrequency current was statistically significantly (F = 49.9, p = 0.006) in nonperfused kidneys compared with that in perfused kidneys. Impedance was higher in nonperfused kidneys, but the difference was not statistically significant (F = 1.446, p = 0.30).
The maximum lesion diameter perpendicular to the electrode as determined by gross pathology was 14.2 ± 2.2 mm (mean ± standard deviation) for the ischemic kidney versus 8.0 ± 2.6 mm in the normally perfused kidney (Fig. 4). This difference is statistically significant (one-tailed t test, p = 0.00002) and represents an average increase of 77% in the lesion diameter of the nonperfused kidneys when compared with the perfused kidneys.
|
No significant complications were noted. Self-limited bleeding was noted after ablation in three of the temporarily ischemic kidneys (50%) and in two of the normally perfused kidneys (33%, Fig. 5A,5B,5C,5D,5E,5F).
|
|
|
|
|
|
In all images, the thermal lesions displayed low signal intensity with a sharp rim of high signal intensity (edema), which was best visualized using turbo STIR sequences (Fig. 5A,5B,5C,5D,5E,5F). Turbo STIR had the highest renal cortex to radiofrequency thermal lesion contrast-to-noise ratio with an average of 4.9 ± 2.5 (mean ± standard deviation) for the perfused kidneys and 4.9 ± 2.4 for the nonperfused kidneys (Table 1). The higher contrast-to-noise ratios using STIR images were statistically significant compared with the ratios of both contrast-enhanced T1-weighted and turbo spin-echo T2-weighted images of both perfused and nonperfused groups (one-tailed t test; p = 0.03 for T1-weighted versus STIR for perfused and nonperfused, p = 0.02 for T2-weighted versus STIR for perfused, and p = 0.01 for T2-weighted versus STIR for non-perfused). The lowest contrast-to-noise ratio was obtained using contrast-enhanced T1-weighted sequences (3.5 ± 1.2 for both perfused and nonperfused). Turbo spin-echo T2-weighted images showed a contrast-to-noise ratio of 3.9 ± 1.7 for the perfused group and 4.0 ± 1.8 for the nonperfused group. No statistically significant differences between the perfused and the nonperfused group could be found for any sequence (one-tailed t test; p = 0.43 for T1-weighted, p = 0.23 for T2-weighted, and p = 0.50 for STIR).
|
All images were compared with the gross pathologic specimens in terms of largest diameter perpendicular to the radiofrequency electrode. The highest accuracy was found using STIR sequences with an average difference of 1.3±1.2 mm perfused and 1.2±1.0 mm nonperfused (Table 1). The next best accuracy was achieved using turbo spin-echo T2-weighted sequences (1.8±0.8 mm for perfused and 2.0±1.3 mm for nonperfused). The lowest accuracy was measured on contrast-enhanced T1-weighted images (2.3±2.3 mm for perfused and 2.3±1.0 mm for nonperfused).
Turbo spin-echo T2-weighted images over-estimated the true lesion size in two cases in the perfused group and one case in the nonperfused group, each by 1 mm. No overestimation was found on either contrast-enhanced T1-weighted or STIR sequences. Using the one-tailed t test, STIR sequences were significantly more accurate than contrast-enhanced T1-weighted sequences (p = 0.006 for perfused and nonperfused groups), whereas no significant difference in accuracy could be established between contrast-enhanced T1-weighted and turbo spin-echo T2-weighted sequences (p = 0.14 for perfused group, p = 0.18 for nonperfused group) nor between turbo spin-echo T2-weighted and STIR sequences (p = 0.10 for perfused, p = 0.05 for nonperfused). No statistically significant differences between the perfused and the nonperfused groups could be found for any sequence (one-tailed t test, p = 0.50 for T1-weighted, p = 0.31 for T2-weighted, and p = 0.50 for STIR).
The lesions showed a well-demarcated border on gross pathology corresponding to the high-signal-intensity rim on postprocedure images (Fig. 6A,6B,6C,6D,6E,6F). Grossly, the lesions appeared "targetlike," with an inner rim of white tissue adjacent to the electrode track and a well-circumscribed outer rim of red hemorrhagic tissue. Renal tissue not affected by the thermocoagulation appeared to be normal. Histologically, the thermocoagulation site consisted of a central cavity surrounded by an area with subtle features of evolving cell damage. The inner zone showed collapsed tubules and bloodless glomeruli, whose cells showed shrunken pyknotic nuclei. The outer hemorrhagic zone, which was evident grossly, showed pronounced vascular stasis within glomeruli associated with hemorrhagic interstitial edema, correlating with the hyperemic edematous rim visible on postprocedure images.
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Merkle et al. [4] published an elegant approach using radiofrequency as the thermal source and MR imaging for guidance and monitoring of the ablation in a pig model. Since then, this approach has been successfully used in patients in a pilot study and is currently being evaluated in a phase 2 trial (Lewin JS et al., presented at the Scientific Meeting of the International Society for Magnetic Resonance in Medicine [ISMRM, April 2000]).
The advantages and disadvantages of radiofrequency versus other energy sources as well as the potential of MR imaging versus other imaging modalities (such as sonography and CT) for guidance and monitoring have been extensively discussed by Merkle et al. [4]. Radiofrequency has been used for more than 30 years in stereotactic neurosurgical procedures [29]. With the development of reliable monitoring methods over the past decade, radiofrequency-induced tumor ablation has been used in clinical trials for treatment of hepatic [5, 30, 31] and cerebral metastases [13], hepatocellular carcinomas [32], and benign bony lesions [33].
Coagulative tissue necrosis results from the transfer of electrical energy and resistive heating in tissue caused by the passage of rapidly alternating current [29]. The size of the resulting necrosis is determined by various factors that have been extensively studied, including probe size, gauge, temperature, and ablation duration [6]. The designs of the probe and the radiofrequency generator have been improved so that the volume of tissue that can be ablated has been increased. These improvements include the development of multiprobe arrays [7], perfused tip electrodes [8], and pulsed high-current techniques [9]. Still, larger target volumes are desirable, although one of the advantages of radiofrequency ablation over other treatment modalities such as radiation therapy is the option of an indefinite number of repeated applications [10]. Increasing the energy deposition does not lead to an unlimited increase in lesion diameter because the higher energy results in tissue charring and vaporization, which, in turn, increases the tissue impedance and thus decreases radiofrequency heat generation at distance remote from the local tissue impedance increase [6].
One limiting factor for thermal damage is the perfusion of organs working
essentially as a heat sink, which can be deduced from the bioheat equation as
outlined by Pennes [34]. This
formula can be simplified in a first approximation to
[35]:
 |
Therefore, techniques that decrease perfusion of the target tissue will likely add to the effect of other developments discussed earlier. Using CT for guidance, Goldberg et al. [11] proved in a porcine model that a decrease in liver perfusion leads to an increase in lesion size. To our knowledge, no use of perfusion modulation in kidney ablation has been reported in the current literature, despite the fact that the kidney is a target organ ideally suited for minimally invasive thermal ablation. It can readily be dissected away from adjacent organs and most often gives rise to unifocal malignancy [22]. In addition, the incidence of satellite malignant tumors is small with renal cell carcinomas of 3 cm or smaller [36]. Hall et al. [37] reported the successful treatment of a patient with a hypervascular renal cell carcinoma using radiofrequency ablation in combination with arterial embolization, but a comparison of this approach with results from regularly perfused tumors was not performed.
The purpose of this study was to test the hypothesis that decreasing the kidney perfusion in pigs results in an increase of thermal lesion size and that these effects can be accurately monitored using MR imaging. A temporary occlusion of the renal arteries would seem feasible in a clinical setting, for example, by balloon occlusion.
The first part of the hypothesis can be accepted because the decrease in perfusion leads to a statistically highly significant increase in lesion size from a mean of 8 mm to a mean of 14.2 mm in diameter (p = 0.00002, Fig. 3). The measurements of lesion diameters for the regularly perfused kidneys agree well with the measurements reported by Merkle et al. [4], who found a mean lesion diameter of 9 mm using the same lesion temperature and duration as reported in our investigation.
In clinical applications, larger lesions are currently created by using cooled-tip electrodes (Lewin JS et al., ISMRM, April 2000) that feature a constant internal perfusion of the tip using 0°C saline as first described by Goldberg et al. [8]. In their article, Goldberg et al. reported an increase of lesion diameter (pig liver, 3-cm electrode) from a mean of 12 mm to a mean of 24 mm (100%). Because the combination of blood perfusion reduction (evaluated as 77%) and internal tip cooling (100%) should work synergistically according to the bioheat equation, a total increase in lesion diameter of approximately 177% seems reasonable. This increase would allow the treatment of 2-cm kidney tumors with a safety margin of at least 1 mm in a single session without repositioning of the electrode. No data on the combination of internally cooled electrodes with reduced organ perfusion have been published so far to our knowledge, but such research seems worth pursuing in future experiments.
The overall effect of perfusion reduction on solid tumors—not healthy kidney parenchyma—will, of course, depend on the vascularity of the mass. Hypervascular tumors would be ideal, but even in hypovascular tumors, an effect on the safety rim around the mass should be expected. This issue also seems worth pursuing in future experiments.
As with all experiments relying on animal data, further studies, including clinical phase 1 and 2 studies with carefully selected patients, are required before the final clinical effect can be properly evaluated.
The second part of the hypothesis underlying this study addresses the monitoring of the procedure using MR imaging as the sole imaging modality. This part of the hypothesis can also be accepted. The technical success rate under MR imaging guidance was 100%, including the placement of the catheter. All lesions could be visualized on MR images, and all three sequences compared in this study provide enough contrast-to-noise ratio (3.5 to 4.9) to adequately visualize the thermal lesion. STIR sequences allow a prediction of the true lesion size with a mean accuracy of 1.2 mm for the perfused group and 1.3 mm for the nonperfused group without the risk of overestimating the true lesion size (Table 1). Contrast-enhanced T1-weighted sequences are not quite as accurate as STIR sequences (mean accuracy, 2.3 mm). Turbo spin-echo T2-weighted images are more accurate than contrast-enhanced T1-weighted images but may pose the risk of lesion overestimation (3/12 lesions, 25%). It is important to note that performing the ablation in nonperfused kidneys did not cause deterioration of the image quality in terms of accuracy or lesion contrast-to-noise ratio compared with the images obtained with normally perfused kidney ablations.
The radiofrequency current required to reach and maintain 90°C was significantly lower for the nonperfused kidneys than for the perfused kidneys (p = 0.0006) as a result of the reduced heat sink. Both current and impedance showed the same tendency over time as described previously in the literature [4, 38]. The decrease in perfusion led to an increase in tissue impedance at all time points that was not statistically significant (p = 0.30), although it could be explained by the changes in tissue caused from temporary ischemia.
In conclusion, we accept the hypothesis that temporary renal ischemia leads to a significantly increased radiofrequency ablation lesion size based on the 77% increase in lesion diameter obtained during balloon occlusion. Furthermore, we conclude that catheter-based perfusion reduction is feasible and does not lead to major complications, and that the entire procedure can be performed and adequately monitored using MR imaging as the sole imaging modality.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E. M. Merkle, S. G. Nour, and J. S. Lewin MR Imaging Follow-up after Percutaneous Radiofrequency Ablation of Renal Cell Carcinoma: Findings in 18 Patients during First 6 Months Radiology, June 1, 2005; 235(3): 1065 - 1071. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Lewin, S. G. Nour, C. F. Connell, A. Sulman, J. L. Duerk, M. I. Resnick, and J. R. Haaga Phase II Clinical Trial of Interactive MR Imaging-guided Interstitial Radiofrequency Thermal Ablation of Primary Kidney Tumors: Initial Experience Radiology, September 1, 2004; 232(3): 835 - 845. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. J. Pijl, J. Doornbos, M. N. J. M. Wasser, H. C. van Houwelingen, R. A. E. M. Tollenaar, and J. L. Bloem Quantitative Analysis of Focal Masses at MR Imaging: A Plea for Standardization Radiology, June 1, 2004; 231(3): 737 - 744. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Duszak Jr Vascular Occlusion: Can We Push Radiofrequency Ablation into New Size Frontiers? Radiology, May 1, 2004; 231(2): 291 - 292. [Full Text] [PDF]
|
|
|
|
|
|
I. Chang, I. Mikityansky, D. Wray-Cahen, W. F. Pritchard, J. W. Karanian, and B. J. Wood Effects of Perfusion on Radiofrequency Ablation in Swine Kidneys Radiology, May 1, 2004; 231(2): 500 - 505. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. W. Mayo-Smith, D. E. Dupuy, P. M. Parikh, J. A. Pezzullo, and J. J. Cronan Imaging-Guided Percutaneous Radiofrequency Ablation of Solid Renal Masses: Techniques and Outcomes of 38 Treatment Sessions in 32 Consecutive Patients Am. J. Roentgenol., June 1, 2003; 180(6): 1503 - 1508. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
; average for nonperfused kidneys = 
.
