| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the Departments of Radiology, Division of MRI (E.M.M., J.R.H., J.L.D., J.S.L.), Biomedical Engineering (J.L.D.), Pathology (G.H.J.), and Oncology (J.S.L.), University Hospitals of Cleveland, Case Western Reserve University, 11100 Euclid Ave, Cleveland, OH 44106, and the Department of Diagnostic Radiology, University of Ulm, Germany (E.M.M., H.J.B.). From the 1998 RSNA scientific assembly. Received September 9, 1998; revision requested November 3; final revision received February 12, 1999; accepted June 8. Supported in part by grants from the Whitaker Foundation, Siemens Medical Systems, Minrad, Radionics, American Cancer Society, Mary Ann S. Swetland Fund, and the M. E. and F. J. Callahan Foundation and by Deutsche Forschungsgemeinschaft grant Me 1593/1-1. Address reprint requests to J.S.L. (e-mail: lewin@uhrad.com)
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: MR imaging-guided RF ablation was performed in the pancreas of six pigs. A 17-gauge monopolar RF probe was inserted into the pancreas with MR imaging guidance, and RF was applied for 10 minutes. After postprocedural imaging (T2-weighted, short inversion time inversion-recovery [STIR], and T1-weighted imaging before and after intravenous administration of gadodiamide), the pigs were observed for 7 days and follow-up MR images were acquired. The pigs were sacrificed, and pathologic examination was performed.
RESULTS: Successful RF probe placement was accomplished in all pigs; the interventional procedure took 46–80 minutes. Thermal lesions were 12–15 mm perpendicular to the probe track and were best seen on STIR and contrast material–enhanced T1-weighted images with a radiologic and/or pathologic mean difference in RF lesion diameter of 1.7 mm ± 1.0 (SD) and 0.8 mm ± 1.2, respectively. Diarrhea was the only side effect during the 1-week follow-up; no clinical signs of pancreatitis occurred.
CONCLUSION: MR imaging–guided RF thermal ablation in the pancreas is feasible and safe. Induced thermal lesion size can best be monitored with STIR and contrast-enhanced T1-weighted images. In the future, RF ablation may offer an alternative treatment option for pancreatic cancer.
Index terms: Interventional procedures, experimental studies, 77.12, 77.12149, 77.1269 • Magnetic resonance (MR), guidance, 77.121411, 77.121412, 77.12143 • Pancreas, interventional procedures, 77.1269 • Pancreatitis, 77.291 • Radiofrequency (RF) ablation, 77.1293, 77.1269
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The variety of primary sites treated with interstitial thermotherapy is increasing. Over the past decade, encouraging preliminary clinical results have been reported regarding treatment of benign lesions such as osteoid osteomas (6) and benign prostatic hyperplasia (7), as well as primary and secondary malignant tumors in the brain (8,9), head and neck (10), liver (1,11), and prostate gland (12).
To our knowledge, the pancreas has not yet been a target for percutaneous interstitial thermotherapy despite the fact that pancreatic carcinoma was the fourth leading cause of estimated cancer death in the United States for both sexes in 1998, with a 5-year relative cancer survival rate of 5% (13). We therefore sought to determine (a) whether magnetic resonance (MR) imaging–guided RF thermal ablation in the pancreas is safe and feasible in a porcine model and (b) whether induced thermal lesion size can be predicted with MR imaging monitoring.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
First, an in-room, 1,024 x 1,280-pixel, RF-shielded liquid crystal monitor was installed for image viewing at the side of the magnet, as is used in an x-ray fluoroscopy suite. Second, control of the imager from within the imaging suite was made possible with an MR imager-compatible mouse and foot pedal. Third, a rapid gradient-echo fast imaging with steady-state precession (FISP) sequence was applied with 17.8/8.1 (repetition time msec/echo time msec), one signal acquired, and a flip angle of 90° to produce images with a clinically adequate signal-to-noise ratio and spatial resolution. Images were acquired at 2–3 seconds per section in a single-section mode or at 4–9 seconds in a three-section mode.
A belt-shaped, 21-cm-diameter, multiturn, solenoid receive-only surface coil (Siemens, Erlangen, Germany) was used for signal reception.
Animal Model
By following a protocol approved by the Animal Research Committee of our institution, six 20–25-kg female farm pigs were anesthetized with a combination of acepromazine maleate (0.25 mg per kilogram of body weight; Fermenta Animal Health, Kansas City, Mo) and ketamine hydrochloride (7.5 mg/kg; Ketaject; Phoenix Scientific, St Joseph, Mo) injected intramuscularly. A 20-gauge, 1-inch-long intravenous catheter (Terumo Medical, Elkton, Md) was inserted into a dorsal ear vein, and thiopentothal sodium (Pentothal; Abbott Laboratories, North Chicago, Ill) was administered (15 mg/kg) via intravenous injection to allow tracheal intubation. This was followed by machine-assisted gas anesthesia with halothane 1% (Halocarbon Laboratories, River Edge, NJ). Antibiotic prophylaxis was performed with a single intramuscular injection of 600,000 IU of penicillin G benzathine (Hanford U.S. Veterinary, Syracuse, NY). The animal's thighs and abdomen were then shaved. The pig was placed on its side on the MR imager table, with the right side down for pancreatic head procedures (n = 4) and with the left side down for pancreatic body and tail interventions (n = 2). Two 8 x 12-cm wire mesh grounding pads coated with conductive gel (Aquasonic 100; Parker, Orange, NJ) were placed, one on each thigh.
RF Ablation
The desired ablation site was determined on transverse turbo spin-echo (SE) T2-weighted images (2,800/96, four signals acquired, echo-train length of seven, one presaturation band placed over the anterior abdominal wall) and transverse SE T1-weighted images (621/26, three signals acquired, presaturation of the abdominal wall) on the basis of (a) location with as much surrounding pancreatic tissue as possible and (b) percutaneous accessibility.
The target was localized on the skin surface with a fluid-filled syringe. A custom MR-compatible 17-gauge monopolar titanium electrode with a 2-cm exposed tip (Radionics, Burlington, Mass) was advanced during a continuous imaging mode that consisted of the sequential acquisition, reconstruction, and display of multiple sets of three parallel 5-mm sections. This set of sections was centered on the predicted probe position to allow rapid detection of probe bending or deflection. After placement of the probe at the target site, its position was confirmed with a turbo SE T1-weighted sequence (500/24, three signals acquired, echo-train length of five).
In all pigs, RF was applied for 10 minutes with a 100-W RF generator operating at 500 kHz (modified model RFG-3C; Radionics) with the tip temperature maintained at 90°C ± 2 (SD). During RF application, impedance, voltage, current, and power levels were documented every 60 seconds. These values were provided as output parameters on a built-in display on the RF generator.
Postprocedural imaging included turbo SE T2-weighted (2,800/96, four signals acquired, echo-train length of seven, presaturation of the abdominal wall), turbo STIR (2,700/48/110 [repetition time msec/echo time msec/inversion time msec], three signals acquired, echo-train length of seven), and SE T1-weighted sequences (621/26, three signals acquired, presaturation of the abdominal wall). T1-weighted imaging was repeated after intravenous administration of 0.2 mmol/kg gadodiamide (Omniscan; Nycomed, Princeton, NJ). All images were obtained transversely.
Follow-up Studies
The pigs were observed for signs of illness by a radiologist (E.M.M.), an animal technician, and a full-time veterinarian daily for 1 week. All clinical side effects were documented and treated in an appropriate manner with 0.3 mg of an intramuscularly administered pain-relieving medication (buprenorphine hydrochloride; Reckitt & Colman Products, Hull, England) and 10 mg/kg oral antidiarrheal medication (tylasin tartrate [Tylan; Elanco Animal Health, Indianapolis, Ind]). Follow-up MR images were acquired with the same 0.2-T imaging system and postprocedural sequences used for RF ablation.
The maximum diameter of each coagulative lesion was measured perpendicular to the RF probe track on postprocedural images with electronic calipers on the workstation monitor by one radiologist (E.M.M.) blinded to the results of pathologic examination. Only the hypointense, nonenhancing area within the pancreas was considered a thermally induced lesion. The margin of the RF lesion was defined by a hyperintense rim that was detected visually. No automated segmentation algorithms were used. Image dimensions were compared with those obtained at gross pathologic examination during the tissue sectioning that followed fixation. Every effort was made to slice the specimen along the RF probe track in a plane corresponding to the transverse imaging direction. The maximum lesion diameter perpendicular to the probe track was recorded.
Image Analysis
Signal amplitude of the image background, or noise, the RF thermal lesion, and the noninvolved pancreas were measured by a radiologist (E.M.M.) for each post-RF sequence by defining regions of interest at the MR imaging workstation. Regions of interest for signal intensity of tissue in the RF thermal lesions and noninvolved pancreas were at least 20 mm2 and were chosen in homogeneous, artifact-free areas of the tissue being measured. Regions of interest for noise were at least 500 mm2. Each signal amplitude value was calculated as the mean of three separately sampled regions of interest. The pancreas-to-lesion contrast-to-noise ratio (CNR) was calculated for each sequence with the following equation (14): CNR = (SApancreas - SAlesion)/SAnoise, where SApancreas, SAlesion, and SAnoise represent the signal amplitudes of the noninvolved pancreas, of the RF thermal lesion, and of the background noise, respectively.
In the last two pigs, a total of four blood samples each were drawn to measure the enzyme levels of amylase and lipase. Sampling points were (a) before RF probe insertion, (b) after probe insertion and before the ablation procedure, (c) 20 minutes after finishing the interstitial thermotherapy, and (d) during the follow-up study 1 week after RF ablation.
Statistical analysis was performed with the Student t test. A P value less than .05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
|
|
|
(mean, 79 ± 7), and the decrease in impedance after RF application for 10 minutes was 6–11 (mean, 9 ± 2). On transverse contrast-enhanced SE T1-weighted images acquired within 50 minutes after ablation (Table 1), coagulation, characterized as a central hypointense nonenhancing zone, varied in diameter from 1.2 to 1.5 cm (mean, 1.4 cm ± 0.1) (Fig 1) (Table 2). The foci of coagulation were also visible on turbo SE T2-weighted and STIR images and were demarcated by the surrounding edema, but they were poorly detected on nonenhanced SE T1-weighted images. Calculated pancreas-to-lesion CNR values were highest for contrast-enhanced SE T1-weighted images and showed statistically significant differences when compared with nonenhanced SE T1-weighted (P < .05) and turbo SE T2-weighted (P < .01) images. Although somewhat higher, the CNR for contrast-enhanced T1-weighted images was not significantly different from that for STIR images (P = .28) (Table 3).
|
|
At follow-up imaging 7 days after interstitial thermotherapy, RF-induced lesions were again best visualized on STIR and contrast-enhanced SE T1-weighted images (Table 3), but differences in CNR were not statistically significant. On transverse contrast-enhanced SE T1-weighted images, lesions were 1.3–1.6 cm (mean, 1.5 cm ± 0.1) in diameter and were slightly larger when compared with lesions on the images acquired immediately after intervention but were not sufficiently different to achieve statistical significance (P = .44). Signs of neither necrotizing pancreatitis nor peritonitis were visible on follow-up images. No dilation of the common bile duct or pancreatic duct was observed. One pig demonstrated localized peripancreatitis of the fatty tissue; minor thermal damage of the upper pole of the left kidney (pancreatic tail ablation) was also visible in the same animal. In another pig, a 4 x 3 x 2-cm hematoma (pancreatic body ablation), as noted immediately after electrode insertion, was still visible.
In the last two pigs from which blood samples were taken, neither the amylase nor the lipase level demonstrated significant elevation over baseline at any time during the interventional procedure, at 20 minutes after ablation, or in the 7 days following interstitial thermotherapy. Baseline values were 3,514 U/L and 3,520 U/L for amylase and less than 20 U/L each for lipase. The maximum amylase level was 3,540 U/L as measured 20 minutes after ablation. Lipase levels were always lower than 20 U/L.
At gross pathologic examination, it was noted that the thermal lesion reached the serosa of the duodenal wall in one pig, without evidence of ulceration. Another pig demonstrated localized minor exudative inflammation of the pancreatic tail and thermal damage (3.5 x 1.5 cm) to the upper pole of the left kidney. Also, the hematoma noted at imaging was grossly visible at dissection. At pathologic examination, the lesion diameter was 1.2–1.5 cm (mean, 1.4 cm ± 0.1) (Table 2). At gross examination, specimens demonstrated a clearly demarcated needle track and surrounding yellow-colored necrosis with well-defined and sharp margins (Fig 3). At histologic examination, the central yellow zone correlated with tissue coagulation with sharp serpiginous edges and a marginal reactive zone with neutrophilic infiltration. This was surrounded by a broad zone of granulation tissue with inflammatory infiltration by neutrophils, lymphocytes, histiocytes, plasma cells, and eosinophils.
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Furthermore, this disease rarely occurs before the age of 45 years, but its occurrence increases sharply thereafter; the pancreas is the fourth leading cancer site for men older than 54 years and for women older than 75 years (13,15). Therefore, patients are often debilitated, with coexistent medical problems (13,15). These data clearly demonstrate the necessity for development of palliative, minimally invasive therapeutic strategies to treat the jaundice, duodenal obstruction, and pain associated with this disease.
RF ablation is a safe, minimally invasive modality that has been used in stereotactic neurosurgical procedures for more than 30 years (17). During the past decade, RF-induced tumor ablation has also been used in early clinical trials for treatment of hepatocellular carcinoma (11), hepatic (18–20) and cerebral metastases (8), and benign bone lesions such as osteoid osteoma (6). Typically, under cross-sectional imaging guidance, a thin electrode is placed percutaneously directly into the tumor. Coagulation necrosis results from resistive heating in the tissue adjacent to the noninsulated part of the electrode.
Two main factors have impeded the use of minimally invasive interstitial thermotherapy in the pancreas to date. First, the retroperitoneal, central location of the pancreas deep within the body requires a long access path. Second, use of laser fibers and cryoprobes requires introduction of a 7–10-F sheath (3,10), which increases the risk of bleeding complications.
With MR imaging guidance, the desired insertion site in the pancreas was accessed without difficulty by using the 17-gauge RF probe in all six pigs in our series. Excellent vessel conspicuity due to flow-related enhancement effects inherent in two-dimensional Fourier transform gradient-echo sequences for probe guidance allowed a safe intervention (Fig 1b). In one pig, however, a puncture-related hemorrhage occurred as a minor complication (Fig 2). This most likely relates to the targeted site for insertion and difficult planned pathway for access in this experimental model, rather than to the guidance modality used for insertion, as a previous study of 106 MR imaging-guided procedures did not encounter complication (21). In general, the complication rate is 3.3% for computed tomographically and ultrasonographically-guided biopsies of the pancreas with 16-22-gauge needles (22). In a series of 269 procedures, Brandt et al (22) reported minor complications that included vasovagal reaction (n = 4), pain requiring narcotics (n = 1), and one small hemorrhage (n = 1). Major complications included pancreatitis (n = 2) and a leak in the pancreatic duct that necessitated emergency surgery.
The time required for the MR imaging-guided components of the interventional procedure was 46–80 minutes, with a mean of 61 minutes (Table 1), and included RF probe insertion with gradient-echo sequences, confirmation of the probe tip with turbo SE T1-weighted imaging (acquisition time, 1 minute 19 seconds), the RF ablation procedure (10 minutes), and immediate postprocedural turbo SE T2-weighted imaging with the probe in situ (acquisition time, 4 minutes 57 seconds). Interventional procedure time was longest when RF ablation was performed in both the pancreatic body and tail because of the smaller size of the target compared with the pancreatic head and the more difficult access to the pancreas between the kidney and spleen.
The RF current necessary to reach 90°C ± 2 was 400–600 mA, half the value reported for treatment of hepatic lesions (4,20). The cause of this discrepancy is perfusion-mediated tissue cooling; that is, vascular flow within an organ, which is more noticeable in the liver. The decrease in impedance after RF application for 10 minutes was 6–11 (9 ± 2) and showed the same tendency as described for ex vivo liver tissue experiments (23).
At MR imaging, RF-induced thermal lesions could be clearly identified immediately after the procedure (Fig 1). Similar to its application in the liver (4), the ability to visualize the resultant thermal lesion during the MR imaging-guided thermotherapy procedure provides the opportunity to immediately reposition the RF electrode and repeat ablation to ensure that all of the targeted tumor and margins have been adequately treated. This aspect of the MR imaging-guided and monitored technique represents its major advantage over other modalities for procedural guidance.
Signal intensity characteristics of coagulation necrosis were the same as described for other organs such as the brain (8,24) and the liver (4,25). The thermal lesions appeared as hypointense regions on T2-weighted and STIR images and were surrounded by peripheral hyperintense areas suggestive of edema. With T1-weighted sequences, RF-induced lesions were clearly depicted only after gadodiamide administration and were characterized by their central hypointense nonenhancing zone with peripheral rim enhancement (Fig 1). Calculated pancreas–to–RF–induced lesion CNRs were highest for contrast-enhanced T1-weighted images; this was significantly different from ratios for nonenhanced SE T1-weighted (P < .05) and turbo SE T2-weighted (P < .01) images. Also, STIR imaging allowed clear and immediate identification of the thermal lesions.
During the 1-week follow-up, only minor side effects such as diarrhea were observed and were easily and successfully treated. Neither clinical pancreatitis, peritonitis, nor jaundice occurred. This was confirmed in two animals with blood chemistry analysis and in all animals with follow-up imaging 7 days after intervention. In addition, no thrombosis of the major peripancreatic vessels was observed.
The main pancreatic duct in pigs is a fine, 1-mm structure that opens directly into the duodenal wall approximately 10–20 mm distal to the pylorus (26). This duct was not visible either prior to or after RF ablation at MR imaging. Also at follow-up imaging 7 days after intervention, no dilatation of the pancreatic duct was observed. The reason for this lack of ductal obstruction, especially following interventional procedures in the pancreatic head, remains unclear, but porcine pancreatic tissue seems to be resistive to mechanical alteration. In clinical practice, however, obstructive pancreatitis due to duct occlusion remains a possibility, especially if benign pancreatic tumors such as endocrine neoplasms are the target of the interventional procedure.
Treatment of malignant tumors is a different scenario: Pancreatic cancer involves the lymphatic system at an early stage; therefore, a curative approach with RF ablation does not seem realistic. Pain treatment, however, might be an effective treatment option, especially in patients in whom the tumor infiltrates the celiac plexus and ethanol ablation is not considered an effective treatment option (27). In these patients, due to prior surgical treatment or prophylactic endoscopic stent implantation, the risk of pancreatic duct damage or occlusion seems smaller.
RF-induced lesions demonstrated the same signal intensity characteristics on follow-up images as on the immediate postthermotherapy images but were about 10% larger than those on the images acquired immediately after intervention (P = .44). These results agree with those of Tacke et al (3), who observed a small increase in lesion size 7 days after cryotherapy in the liver. In two patients, the RF lesion extended beyond the pancreas and caused thermal damage at the serosa of the duodenal wall and the upper pole of the left kidney. This might have been avoided by using temperature mapping, which is now also available for midfield- and low-field-strength systems and allows online monitoring of the interventional procedure (28,29). On the other hand, these sequences are based on gradient-echo techniques, and the susceptibility artifact size of the 17-gauge titanium RF probe used in this study is nearly 10 mm at gradient-echo imaging, which leads to an almost complete overlap of the temperature map.
Lesion size was 12–15 mm perpendicular to the probe track at gross pathologic examination; this size might not be sufficient for adequate tumor treatment. In the liver, the diameter of coagulation can be increased to approximately 24 mm (30) perpendicular to the probe track, without repositioning the device if technically modified perfused RF systems, developed to increase energy deposition into the tissue, are used. It is reasonable to expect that an increase in maximal lesion diameter could be achieved in the pancreas if perfused RF systems were applied, which might allow creation of somewhat larger thermal lesions at each electrode position during interstitial thermal treatment. Lesion diameter along the probe track is predetermined by the probe design and is closely related to the length of the noninsulated segment of the probe tip (31) but can easily be increased by slightly withdrawing the probe and repeating RF application.
With transverse contrast-enhanced SE T1-weighted images acquired immediately after intervention, the diameter of the lesion as determined at gross pathologic examination of the thermal lesion was slightly overestimated in two animals and underestimated in one, with a mean difference between measurements at radiologic and pathologic examination of RF lesion diameter of 0.8 mm ± 1.2 (Table 2). This slight error in prediction of lesion size, never more than 3 mm, may be a true manifestation of the MR imaging technique. However, this may also reflect the mild tissue shrinkage that may occur with fixation at pathologic examination. Differentiation between these two possibilities will require further investigation with more advanced pathologic examination techniques such as the use of feducial markers (32).
In conclusion, MR imaging-guided RF thermal ablation in the pancreas is feasible in a porcine model, and the induced thermal lesion size can be predicted with MR imaging monitoring.Practical application: MR imaging–guided RF thermal ablation in the pancreas of a porcine model is feasible and reasonable in terms of time, but minor complications may be encountered, especially when the target is located in the body or tail of the gland. Induced thermal lesion size can best be monitored with STIR and contrast-enhanced T1-weighted images acquired immediately after intervention, thereby allowing feedback and, when necessary, repositioning of the RF probe. RF ablation may offer a future alternative treatment option for pancreatic cancer, perhaps by becoming part of a multimodal therapeutic strategy (33,34).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Author contributions: Guarantors of integrity of entire study, J.S.L., E.M.M.; study concepts and design, J.S.L., E.M.M., J.L.D., H.J.B., J.R.H.; definition of intellectual content, J.S.L., E.M.M., J.L.D., H.J.B., J.R.H.; literature research, J.S.L., E.M.M.; experimental studies, J.S.L., E.M.M., G.H.J.; data acquisition and analysis, J.S.L., E.M.M., G.H.J.; statistical analysis, J.S.L., E.M.M.; manuscript preparation, editing, and review, all authors.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Adachi, T. Kaminou, T. Ogawa, T. Kawai, Y. Takaki, K. Sugiura, Y. Ohuchi, and M. Hashimoto Heat Distribution in the Spinal Canal during Radiofrequency Ablation for Vertebral Lesions: Study in Swine Radiology, May 1, 2008; 247(2): 374 - 380. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Dickfeld, R. Kato, M. Zviman, S. Lai, G. Meininger, A. C. Lardo, A. Roguin, D. Blumke, R. Berger, H. Calkins, et al. Characterization of Radiofrequency Ablation Lesions With Gadolinium-Enhanced Cardiovascular Magnetic Resonance Imaging J. Am. Coll. Cardiol., January 17, 2006; 47(2): 370 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Haaga, A. A. Exner, Y. Wang, N. T. Stowe, and P. J. Tarcha Combined Tumor Therapy by Using Radiofrequency Ablation and 5-FU-Laden Polymer Implants: Evaluation in Rats and Rabbits Radiology, December 1, 2005; 237(3): 911 - 918. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
|
|
S. G. Nour, J. S. Lewin, M. Gutman, C. Hillenbrand, F. K. Wacker, J. W. Wong, I. C. Mitchell, C. B. Armstrong, M. M. Hashim, J. L. Duerk, et al. Percutaneous MR Imaging- guided Radiofrequency Interstitial Thermal Ablation of Tongue Base in Porcine Models: Implications for Obstructive Sleep Apnea Syndrome Radiology, February 1, 2004; 230(2): 359 - 368. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Aschoff, E. M. Merkle, S. N. Emancipator, C. A. Petersilge, J. L. Duerk, and J. S. Lewin Femur: MR Imaging-guided Radio-frequency Ablation in a Porcine Model— Feasibility Study Radiology, November 1, 2002; 225(2): 471 - 478. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Nour, A. J. Aschoff, I. C. S. Mitchell, S. N. Emancipator, J. L. Duerk, and J. S. Lewin MR Imaging-guided Radio-frequency Thermal Ablation of the Lumbar Vertebrae in Porcine Models Radiology, August 1, 2002; 224(2): 452 - 462. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
