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


     

Published online before print October 2, 2002, 10.1148/radiol.2252001500
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2252001500v1
225/2/471    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Aschoff, A. J.
Right arrow Articles by Lewin, J. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Aschoff, A. J.
Right arrow Articles by Lewin, J. S.
(Radiology 2002;225:471-478.)
© RSNA, 2002

Experimental Studies

Femur: MR Imaging–guided Radio-frequency Ablation in a Porcine Model— Feasibility Study1

Andrik J. Aschoff, MD, Elmar M. Merkle, MD, Steven N. Emancipator, MD, Cheryl A. Petersilge, MD, Jeffrey L. Duerk, PhD and Jonathan S. Lewin, MD

1 From the Departments of Radiology (A.J.A., E.M.M., C.A.P., J.L.D., J.S.L.), Pathology (S.N.E.), and Oncology (J.S.L.), University Hospitals of Cleveland/Case Western Reserve University, 11100 Euclid Ave, Cleveland, OH 44106; Department of Diagnostic Radiology, University Hospital of Ulm, Germany (A.J.A., E.M.M.); and Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio (J.L.D.). From the 2000 RSNA scientific assembly. Received September 6, 2000; revision requested October 20; final revision received May 7, 2002; accepted May 29. Supported in part by 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 and Me 1593/1-1), and by research collaborations with Siemens Medical Systems and Radionics. Address correspondence to J.S.L. (e-mail: lewin@uhrad.com).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: To determine the feasibility of magnetic resonance (MR) imaging–guided and –monitored radio-frequency (RF) ablation of bone.

MATERIALS AND METHODS: Seven femurs were treated in five pigs with use of a 0.2-T open MR imager. An 11-gauge bone marrow needle was percutaneously inserted into the distal femur metaphysis with MR fluoroscopy (fast imaging with steady-state precession, or FISP, sequences) to introduce an RF electrode into the bone with further image guidance. Thermal ablation was performed for 10 minutes (90°C ± 2 [mean ± SD]). MR follow-up was performed immediately after ablation and again at 7 and 14 days after the procedure (with contrast material–enhanced T1-weighted, T2-weighted, and fast short inversion time inversion-recovery, or STIR, sequences). The animals were sacrificed at day 14. The femurs were sliced, decalcified, and stained. Image analysis was performed to measure lesion diameter and contrast-to-noise ratio (CNR) and to evaluate complications.

RESULTS: Technical success was obtained in all animals. The lesion diameter perpendicular to the electrode was 15.4 mm ± 2.7. No significant complications were noted. The thermal lesions displayed low signal intensity with a sharp rim of high signal intensity. T2-weighted images demonstrated the highest CNR and the lowest error in predicting the lesion size immediately after ablation (2.7 mm ± 1.3). Contrast-enhanced T1-weighted images demonstrated the highest accuracy at day 14 (1.0 mm ± 1.0).

CONCLUSION: RF ablation of bone with MR imaging as the sole imaging modality is feasible and allows monitoring of the ablation.

© RSNA, 2002

Index terms: Animals • Femur, MR, 445.121411, 445.121412, 445.121413 • Femur, neoplasms, 445.3122, 445.33 • Magnetic resonance (MR), guidance, 445.121411, 445.121412, 445.121413 • Radiofrequency (RF) ablation, 445.1299


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Image-guided radio-frequency (RF) interstitial thermal ablation is an innovation in minimally invasive treatment of selected tumors (1–7), although bone is rarely a target tissue. Rosenthal et al (8–10) and other groups (11–14) reported success in treating osteoid osteomas with RF ablation with computed tomographic (CT) guidance. In addition, preliminary data suggest that RF ablation may be useful in the treatment of those patients with painful osseous metastatic disease (15,16).

To our knowledge, no data have been reported in the current literature on the use of magnetic resonance (MR) imaging for either guidance or monitoring of RF ablations in bone. Tacke et al (17) demonstrated the use of cryoablation in femurs in a porcine model with MR monitoring, but they relied on fluoroscopy for guidance. For other organs, the use of MR imaging as the sole imaging modality for both guidance and procedure monitoring for minimally invasive percutaneous tumor treatment with RF was successful in clinical trials (1) and in experimental animal studies involving the liver (18), brain (19), pancreas (20), and kidneys (21). As a result of the design and construction of nonferromagnetic electrodes, MR imaging can be performed without artifacts for guiding electrode insertion and monitoring thermal lesion development throughout the ablation procedure (1).

The purpose of our study was to test the hypothesis that RF interstitial thermal ablation with MR imaging as the sole imaging modality for both guidance and monitoring is feasible in bone.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
MR Imaging System
The entire procedure was performed in the MR imaging suite with a 0.2-T open C-shaped MR imaging system (Magnetom Open; Siemens Medical Systems, Erlangen, Germany). The following three modifications facilitated image-guided interventions:

First, an in-room 1,024 x 1,280-pixel RF-shielded liquid crystal display monitor was installed at the side of the magnet for image viewing. Second, an MR-compatible mouse and foot pedal allowed control of the imager from within the imaging suite. Third, a rapid gradient-echo sequence (fast imaging with steady-state precession, or FISP) was applied to produce images with clinically adequate spatial resolution and signal-to-noise ratio (17.8/8.1 [repetition time msec/echo time msec], two signals acquired, 90° flip angle, section thickness of 5 mm, field of view of 250 x 250 mm, matrix of 128 x 256, and acquisition time per frame of 4.6 seconds). The whole-body transmit coils incorporated in the imager were used for RF transmission, and a belt-shaped 21-cm-diameter solenoid receive-only surface coil was used for signal reception.

Animal Model
In the protocol for this study, which was approved by the institutional animal care and use committee, seven femurs were treated in five male farm pigs (weight range, 20–25 kg) with use of general anesthesia. The lateral hindquarters of each animal were shaved bilaterally before the first procedure, and one 8 x 12-cm wire mesh grounding pad that was coated with conductive gel was placed on each hind limb. Each pig was positioned supine on the MR imager table to allow access to both femurs within the open magnet.

Only one femur was treated in the first three pigs. Both femurs were treated in the last two pigs after our observation that the first three pigs had not experienced substantial discomfort after the procedure. All five 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 (7.5 mg/kg) (Ketaject; Phoenix Scientific, St Joseph, Mo) by means of intramuscular injection. A 20-gauge 1-inch-long intravenous catheter (Surflo; Terumo Medical, Elkton, Md) was then inserted in a dorsal ear vein, and thiopental sodium (15 mg/kg) (Pentothal; Abbott Laboratories, North Chicago, Ill) was administered intravenously to allow tracheal intubation. Mechanical ventilation was used for the duration of the procedure. Anesthesia was maintained with fluothane (dosage for repeated doses of pentobarbital sodium [Nembutal; Abbott Laboratories], 50–150 mg as needed) (Halothane 1%; Halocarbon Laboratories, River Edge, NJ) during the two initial procedures for each pig (RF ablation and postablation day 0 imaging, and postablation day 7 imaging) and with repeated doses of pentobarbital sodium (Euthasol; Diamond Animal Health, Des Moines, Iowa) during the final procedure (postablation day 14 imaging and sacrifice).

RF Ablation
Targets for ablation were chosen in the metaphysis of the proximal femur on the basis of transverse T2-weighted fast spin-echo MR imaging (2,405/96, three signals acquired, echo train length of seven, section thickness of 6 mm, field of view of 250 mm2, matrix size of 196 x 256, and acquisition time of 3 minutes 49 seconds). The electrode insertion site on the skin relative to the target was localized by moving the target a tap water–filled syringe across the skin surface and monitoring its position on the in-room liquid crystal diode monitor by using the fast imaging with steady-state precession sequence. An 11-gauge (3 x 100-mm) MR-compatible bone marrow needle (Somatex, Berlin, Germany) was inserted percutaneously into the metaphysis of the distal femur with continuous MR fluoroscopy guidance with fast imaging with steady-state precession (three parallel 5-mm-thick sections centered on the needle path) sequences, by one of the authors (A.J.A., E.M.M., J.S.L.). This needle was then used as a sheath to introduce a 17-gauge MR-compatible RF electrode (Radionics, Burlington, Mass) with a 2-cm-long exposed tip into the bone marrow by means of continued image guidance (Fig 1). After the electrode was placed, the tip position was confirmed with a T1-weighted fast spin-echo sequence (500/24, three signals acquired, echo train length of five, section thickness of 5 mm, field of view of 250 mm2, matrix size of 128 x 256, and acquisition time of 1 minute 19 seconds) because fast spin-echo sequences are superior to gradient-echo sequences for accurate depiction of tip position (22).



View larger version (69K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. MR-compatible RF electrode and bone biopsy system. A = exposed tip of 17-gauge MR-compatible RF electrode. B = 11-gauge MR-compatible bone biopsy needle. After successful introduction into the bone, the trocar (C) is removed, and the canula (B) is used as a sheath for introduction of the RF electrode (A). Once the tip of the electrode is placed within the bone marrow, the sheath is withdrawn far enough to provide clearance for the 2-cm-long exposed tip (arrows) of the RF electrode. Ruler indicates centimeters.

 
Thermal ablation was then performed with the electrode tip temperature maintained at 90°C ± 2 by applying RF for 10 minutes with a 150-W RF generator operating at 500 kHz (modified model CC1; Radionics). Baseline impedance, RF current, and impedance during ablation were recorded from the generator display each minute. Then, the electrode was withdrawn.

Postablation transverse and sagittal MR images were obtained immediately after the intervention and at days 7 and 14 both without and with gadopentetate dimeglumine (Magnevist [0.1 mmol/kg]; Berlex Laboratories, Wayne, NJ). Postablation imaging was performed with T1-weighted spin-echo sequences (480/26, three signals acquired, section thickness of 5 mm, field of view of 256 mm2, matrix size of 196 x 256, acquisition time of 4 minutes 39 seconds), T2-weighted fast spin-echo sequences (2,405/96, four signals acquired, echo train length of seven, section thickness of 5 mm, field of view of 256 mm2, matrix size of 196 x 256, acquisition time of 4 minutes 35 seconds), fast short inversion time inversion-recovery (STIR) sequences (2,700/48/110 [inversion time msec], four signals acquired, echo train length of seven, section thickness of 5 mm, field of view of 219–300 x 350–400 mm, matrix of 123–192 x 256, acquisition time of 3 minutes 42 seconds). These sequences were chosen on the basis of findings in prior studies that showed the usefulness of T1-weighted, T2-weighted, and STIR MR imaging for visualization of thermally induced lesions (1,5,21,23,24).

After the last acquisition in the imaging sessions on days 0 and 7, the animals were returned to the animal resource center and allowed to recover with veterinary supervision. The pigs were allowed to be freely active during the entire experimental period. No antibiotics were administered.

After the MR examinations on day 14, the animals were sacrificed with an overdose (0.1 mL per pound of body weight) of intravenous thiopental sodium, and the femurs were harvested for gross pathologic and histologic examinations. For gross pathologic examination, the femurs were cut in sagittal sections that were approximately 3 mm wide by using a wet high-speed band saw (SharpTech, Atlanta, Ga). These sections were then photographed and fixed in 10% formalin. Subsequently, they were decalcified (Decalcifier II; Surgipath Medical, Richmond, Ill) and stained with hematoxylin-eosin for histologic examination. For all lesions, the maximum lesion diameter perpendicular to the electrode was determined at gross pathologic examination on the basis of the central section that contained the visible electrode tract.

Technical success (successful MR imaging-guided bone biopsy, electrode placement, and thermal ablation) was obtained in all animals. Mean procedure time for ablation in one femur (including preablation imaging, RF electrode placement, RF ablation, and postablation imaging) was 1 hour 43 minutes. No clinically important complications were noted. All puncture wounds had healed completely, and no signs of wound infections were present. No postablation analgesia had to be administered as judged by the veterinarians, and all pigs were moving freely without favoring a leg on the day after ablation.

Image Analysis
The relative visibility of the thermal lesions in each of the postablation images was evaluated as follows by one of the authors (A.J.A.). Each image was displayed on a freestanding workstation (MagicView; Siemens Medical Systems). For each lesion, two regions of interest were identified interactively by one investigator (A.J.A.) by using the "Evaluate Statistics" routine in the workstation software (Numaris 3). The two regions of interest were (a) the core of the thermal lesion (ablated tissue) at the center of the affected area and (b) the uninvolved bone marrow near the lesion. Regions of interest were chosen in homogeneous artifact-free areas and were made at the same anatomic level for each image sequence. Mean signal intensity amplitudes from each region of interest were recorded three times, and the results were averaged. Regions of interest for signal intensity amplitudes of tissues included at least 20 pixels (ranges: core, 20–47 pixels; uninvolved marrow, 24–84 pixels). Regions of interest for signal intensity amplitudes of noise included at least 500 pixels (range, 511–1,095 pixels). Mean background noise was measured on each image ventral to the femurs, outside the body, and along the phase-encoding axis.

The contrast between the regions of interest in normal bone marrow and those in the core of the thermal lesion was expressed in terms of the contrast-to-noise ratio (CNR). CNR was calculated as the difference between the mean signal intensity amplitudes from the two regions of interest divided by the SD of the noise. The direction of the difference calculation for each CNR was selected to be that which would yield a positive value for the group mean.

In addition, all images were evaluated for signs of osteomyelitis or pathologic fractures.

Statistical Analysis
Measurements of lesion diameter obtained at gross pathologic examination were compared with the diameter measurements obtained at MR imaging (A.J.A.) for the three imaging time points (immediately after ablation and at 7- and 14-day follow-up) in terms of mean, minimum, and maximum differences; SD; and over- or underestimation.

CNRs among images obtained with the various MR sequences were compared for all time points in terms of minimum and maximum values, means, and SDs.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Postablation MR images revealed no signs of osteomyelitis or pathologic fractures. RF current during ablation was a mean of 234 mA ± 37.8. The mean current applied in the 1st minute of ablation was 284 mA ± 41.8 (Fig 2). The baseline impedance was a mean of 96 {Omega} ± 13.5. The impedance decreased by a mean of 5.5 {Omega} ± 6.2 in the 1st minute to a mean of 91 {Omega} ± 13.3. The impedance then leveled out to a mean of 89 {Omega} ± 17.6 (Fig 3). The mean lesion diameter perpendicular to the electrode was 15.4 mm ± 2.7 (range, 11–18 mm).



View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Line graph shows mean RF current versus time during RF ablation in 10 pig femurs. Error bars represent the SD of the mean. Note the continuous decrease in RF current during interstitial thermal therapy.

 


View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Line graph shows mean impedance versus time during RF ablation in 10 pig femurs. Error bars represent the SD of the mean. Note the decrease in impedance in the 1st minute of ablation.

 
With all sequences, the thermal lesions displayed low signal intensity with a sharp rim of high signal intensity (Figs 4, 5). Two lesions were not detectable in the contrast material–enhanced T1-weighted MR images obtained immediately after ablation. T2-weighted images depicted the highest mean CNRs at all three time points, followed by contrast-enhanced T1-weighted images. The lowest CNRs were found for STIR images (Table 1).



View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Findings at 2 weeks after RF bone ablation in one pig femur. A, Sagittal contrast-enhanced T1-weighted MR image. B, Sagittal T2-weighted MR image. C, Sagittal STIR MR image (matrix, 123 x 256; field of view, 219 x 350 mm). D, Gross pathologic specimen. Arrows in A-D indicate necrosis, which is clearly visible in the MR images and corresponds to a sharply delineated thermal lesion on the gross specimen. The ruler on the left indicates inches. The ruler on the right indicates centimeters.

 


View larger version (168K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Bilateral RF bone ablation. MR images of bilateral femurs were obtained immediately after ablation. Top row: Transverse sections through both lesions. Middle row: Sagittal views of right femur. Bottom row: Sagittal views of left femur. From left to right: contrast-enhanced T1-weighted, T2-weighted, and STIR (matrix of 123 x 256 and field of view of 219 x 350 mm) MR images. In the right femur, necrosis (arrows) is visible but not as sharply as in images obtained 2 weeks after ablation (Fig 4).

 

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

  		TABLE 1. CNRs for Different MR Pulse Sequences and Time Points

 
Contrast-enhanced T1-weighted MR images showed an increase in the mean CNR from day 0 (CNR of 3.4) to the later time points (day 7 CNR, 6.5; day 14 CNR, 5.9). The ranges of mean CNRs for T2-weighted and STIR images, however, remained the same throughout the experiment (Table 1).

The errors in maximum lesion diameter perpendicular to the electrode measured on MR images compared with those measured at gross pathologic examination are listed in Table 2 and illustrated in Figure 6. T2-weighted images demonstrated the lowest error immediately after ablation (2.7 mm ± 1.3). The best overall accuracy was obtained with contrast-enhanced T1-weighted images at day 7 (2.1 mm ± 1.5) and day 14 (1.0 mm ± 1.0). The only case of overestimation of lesion size (by 1 mm) was found at day 7 with T2-weighted images.


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

  		TABLE 2. Error with MR Images for Predicting Actual RF Lesion Size Compared with Gross Pathologic Findings

 


View larger version (36K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Bar graph shows mean error of MR images in predicting the actual RF lesion size compared with gross pathologic findings in 10 pig femurs. White bars = day 0, gray bars = day 7, black bars = day 14. Error bars represent the SD of the mean. CE-T1 = contrast-enhanced T1 weighted, T2 = T2 weighted.

 
At gross pathologic examination, slices of distal femur that were obtained through and immediately adjacent to the needle tracts exhibited grossly evident hemorrhage and devitalization of marrow surrounding the tract. Typically, lesions showed three zones: a central hemorrhagic tract core, a middle zone of yellow and mottled mummified tissue (coagulation necrosis not yet organized), and a peripheral red rim of organization. Beyond the lesions, the marrow appeared normal in all respects. Marrow proximal to or beyond the metaphyseal plates appeared normal even when marrow equidistant to the puncture tract appeared necrotic. Moreover, a zone of viable marrow followed the contour of the metaphysis at such sites (Fig 7).



View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7. Gross pathologic and histologic appearances of RF lesion in bone in one pig femur. (a) Gross pathologic specimen was obtained at 14 days after thermal coagulation. Note the sharp delineation of the lesion (arrows), which is surrounded by normal-appearing bone marrow. (b) Low-power photomicrograph. The electrode tract is to the right of this slice. The interface (arrows) between lesional and normal marrow is clearly demarcated. Metaphysis is on the lower left. (Hematoxylin-eosin stain; original magnification, x5.) (c) Photomicrograph of the interface between lesional and normal marrow. Thermally coagulated regions are sharply demarcated from the normal-appearing marrow. Postablation changes include focal hemorrhage and/or residual fat necrosis, dilation of blood vessels, and enlargement of endothelial nuclei. The marrow is severely pancytopenic, and there is replacement of the hematopoietic progenitors with a myxoid edematous stroma. Bone spicules also show reactive and involutional changes. These features are better visualized with higher magnification. (Hematoxylin-eosin stain; original magnification, x20.) (d) Photomicrograph at higher magnification clearly shows viable hematopoietic elements in the nonlesional marrow, including dilated postcapillary venules with reactive endothelial nuclei (dotted arrow), focal resorption of trabecular bone (solid arrow), and new bone formation (arrowheads). Again, note replacement of the normal marrow cells (major portion of the image) with myxoid edematous stroma. (Hematoxylin-eosin stain; original magnification, x80.)

 
Findings at histologic examination confirmed central hemorrhage in the puncture tract and organized or organizing fat necrosis. Scattered small foci of hemorrhage were also present in necrotic marrow at sites remote from the puncture tract. Organization (granulation tissue) was evident with accumulation of edematous collagen and extracellular matrix in the middle zone and with proliferating and dilated blood vessels in the outer zones. In addition, bone changes were conspicuous in the necrotic areas and the adjacent marrow. Numerous spicules exhibited broadening with new bone formation, especially with polarization. Other foci show resorption and/or narrowing of the cancellous bone with prominent osteoclasts. Many empty lacunae were evident within trabecular bone in the central areas. This feature suggests necrosis of bone and was not seen in the adjacent areas of organization. Beyond the borders of the lesion, viable normal marrow cells, with all three hematopoietic lineages represented, were detected among intact bone spicules (Fig 7).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
RF ablation has been used for more than 30 years in stereotactic neurosurgical procedures (25). With the development of reliable image guidance methods over the past decade, RF-induced tumor ablation has been used in clinical trials for treatment of hepatic (1–4), cerebral (5), and osseous (15) metastases, hepatocellular carcinomas (6), and lung malignancies (7), as well as in the treatment of benign bone lesions (8–14).

Typically with cross-sectional imaging guidance, a thin (usually 14–21-gauge) electrode is percutaneously placed directly into the tumor. Coagulative tissue necrosis is caused by the transfer of electrical energy and by resistive heating in tissue as a result of the passage of rapidly alternating current (25). The size of the resulting necrosis is determined by a variety of factors, including probe size, needle gauge, temperature, and duration of ablation, that have been studied extensively (26).

Lundskog (27), in 1972, reported extensive findings about the susceptibility of bone to heat. RF thermal ablation of bone was described by Tillotson et al (28) in 1989 in a dog model with fluoroscopic guidance. Application with CT guidance in patients was published by Rosenthal et al (8) in 1992. The main effects of RF on bone, as well as the corresponding physical RF parameters, are well understood and were not the main target of the current study.

The mean RF current necessary to reach the range of 90°C ± 2 was 284 mA during the 1st minute of ablation and then continuously decreased during interstitial thermal therapy (Fig 2). The impedance after RF application for 10 minutes decreased to a mean of 7.6 {Omega} (Fig 3). Previous reports (8,28) did not include RF current or impedance data; therefore, no direct comparison is possible. Current and impedance showed the same tendency that was described in other organs both ex vivo and in vivo (21,29).

At pathologic analysis, coagulation varied from 11 to 18 mm (mean, 15.4 mm) perpendicular to the RF electrode tract. Tillotson et al (28) reported comparable lesion diameters (range, 9–13 mm) with a tip temperature of 80°C and electrodes with smaller exposed tips. It is reasonable to expect that larger lesions could be treated by using technically modified RF systems (eg, with a perfused tip to allow external cooling of the electrode during ablation). Such systems were developed to increase energy deposition into the tissue (30,31).

Potential applications for RF treatment in bone include osteoid osteomas and painful osseous metastatic disease. Osteoid osteomas are benign tumors with a predilection for the cortex of the shafts of long bones. The spine is involved in about 7%–10% of patients, and the majority (50%–60%) involve the femur and tibia (32,33). Although benign, they cause substantial morbidity because of their painfulness (8). Treatment includes medical (34) and surgical (32) options. In addition, several authors (8–14) reported the successful treatment of osteoid osteomas with CT-guided RF ablation. Advantages of percutaneous thermal coagulation over surgical procedures include the short hospital stay with the patients able to resume normal activity on the day after the procedure (11). Rosenthal et al (10) compared the outcomes in 101 patients with osteoid osteoma, 33 of whom were treated with RF ablation and 68, with open techniques. In their study, the mean hospital stay was 0.18 day for the patients treated with RF ablation compared with 4.7 days for the patients treated with open procedures. There was no statistically significant difference in recurrence rates.

In addition, preliminary clinical experience exists for palliative RF ablation of painful osseous metastatic disease in patients who may not be candidates for surgery to remove tumors and in whom external-beam radiation therapy has been ineffective for tumor control (15,16).

A modern imaging and guidance modality for minimally ablative tumor treatment should fulfill two conditions: First, insertion of the RF electrode into the tumor should be possible in a safe and rapid manner. This condition has been shown with a variety of cross-sectional imaging modalities, including ultrasonography (US) (2,3,35), CT (36), and MR imaging (37). MR imaging, with high soft-tissue contrast, multiplanar imaging capabilities, high spatial resolution, and the ability to depict vessels without administration of contrast medium, seems to be well suited for target definition and trajectory selection (37). Development of open MR imagers with wide patient access allows RF electrode insertion and tracking with MR image guidance (38,39).

Second, and of even more importance, is the necessity for control of destructive RF energy deposition. Jolesz (40) postulated a twofold role for MR monitoring of thermal ablation: Real-time MR imaging with temperature-sensitive pulse sequences can depict transient temperature elevations (41,42), and MR images can depict irreversible tissue necrosis.

The role of real-time MR imaging with temperature-sensitive sequences remains to be fully explored. Real-time visualization of developing tissue necrosis has immediate clinical value, and it allows repositioning of the RF electrode or application of additional RF energy, if necessary, in an interactive manner (1). The ability to monitor thermal necrosis is a definite advantage of MR imaging. To our knowledge, no data exist in the recent literature for the use of either US or CT for monitoring of RF ablations in bone. Data do show, however, that US monitoring of RF ablations in the liver cannot help predict the extent of coagulation necrosis (2,3,35). To our knowledge, helical CT was used for online monitoring of thermal therapy in only the in vitro study by Bahner et al (43) in 1996. To our knowledge, this method could not be replicated in vivo in either animal models or human patients.

Although RF ablation of osteoid osteomas is considered to be safe and successful with CT guidance (8–14), recurrences or therapeutic failures are seen. These are most likely a result of incomplete nidus coagulation, as is stated in the surgery literature; several authors link incomplete nidus resection with recurrence (44,45). Barei et al (14) report one recurrence in 11 patients, and Rosenthal et al (10) report four recurrences in 33 patients who had been treated for a primary lesion. Both authors speculate that these failures were caused by suboptimal electrode positioning or subtherapeutic duration of electrode heating. These speculations are supported by our results, which demonstrate that the effective diameter of a thermal lesion varies between 11 and 18 mm with use of controlled treatment. According to the findings of Lundskog (27), the transmission of heat within the bone is sharply limited by blood flow. Physiologic differences in perfusion will therefore lead to differences in lesion size.

On the basis of our results, MR imaging may be used successfully to monitor RF ablation of bone. T2-weighted images allowed prediction of the size of developing thermal necrosis, with a mean accuracy of less than 3 mm (2.7 mm ± 1.3), without the risk of overestimation of thermal lesion size. STIR MR images can also be used to accomplish this task, although the mean accuracy was less than that with T2-weighted images (Table 2, Fig 6). Contrast-enhanced T1-weighted images are not suitable to assess lesion size in bone immediately after ablation. Two (28%) of seven lesions could not be visualized at all. A higher accuracy was achieved when the thermal lesions were reassessed after 7 or 14 days. In these cases, the mean error in predicting the lesion diameter compared with gross pathologic findings can be as low as 1.0 mm ± 1.0 (mean error for contrast-enhanced T1-weighted images at day 14).

With the exception of some day 0 contrast-enhanced T1-weighted images, MR images obtained with all sequences displayed the characteristic appearance of RF lesions, with a sharp rim of high signal intensity and a central zone of low signal intensity in a targetlike configuration. There was no noteworthy difference in the CNRs of images obtained with any sequence between days 7 and 14. This finding and the fact that no major difference in accuracy was found between images obtained with any of the three sequences on days 7 or 14 suggest that it is possible to reassess thermal lesions in bone with MR imaging at 7 days after ablation.

CT is the modality of choice for the diagnosis of osteoid osteoma because of its ability to show both the nidus and the surrounding sclerotic response (46). MR images tend to depict a misleading aggressive appearance. Still, in a comparison between CT and low-spatial-resolution MR imaging in 19 patients with histologically proven osteoid osteoma, a cortical or cancellous lesion suggestive of an osteoid osteoma nidus was seen in 19 of 19 MR images (46). On the basis of this finding, MR imaging has the potential to serve as an image-guidance modality in the treatment of osteoid osteomas.

In conclusion, we accept the hypothesis that RF interstitial thermal ablation of bone with MR imaging as the sole imaging modality for both guidance and monitoring is feasible and that MR imaging may allow monitoring of thermal ablation to ensure complete tumor treatment.

Practical application: MR imaging–guided and –monitored RF treatment of osteoid osteomas and, to a lesser degree, painful bone metastases seems to be possible in patients and may ultimately reduce therapeutic failures or recurrences of osteoid osteomas compared with those with CT-guided unmonitored RF treatments.


   ACKNOWLEDGMENTS
 
The authors thank Lee Friedman, PhD, and John A. Jesberger, BSEE, for their help with the data analysis; Tami McCourt, AS, for assistance in animal observation and anesthesia; Michael Wendt, PhD, for helpful discussions; Elena DuPont for assistance with manuscript preparation; and all members of the Interventional MR Research group for their outstanding support.


   FOOTNOTES
 
Abbreviations: CNR = contrast-to-noise ratio, RF = radio frequency, STIR = short inversion time inversion-recovery

Author contributions: Guarantors of integrity of entire study, A.J.A., J.S.L.; study concepts, A.J.A., E.M.M., J.S.L., J.L.D., C.A.P.; study design, A.J.A., E.M.M., J.S.L., J.L.D.; literature research, A.J.A., E.M.M.; experimental studies, A.J.A., S.N.E., C.A.P.; data acquisition, A.J.A., S.N.E., J.S.L.; data analysis/interpretation, A.J.A., C.A.P., S.N.E., J.S.L., J.L.D.; statistical analysis, A.J.A.; manuscript preparation, A.J.A.; manuscript definition of intellectual content, A.J.A., E.M.M., J.S.L., J.L.D.; manuscript editing, A.J.A., J.S.L.; manuscript revision/review and final version approval, all authors.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 

  1. Lewin JS, Connell CF, Duerk JL, et al. Interactive MRI-guided radiofrequency interstitial thermal ablation of abdominal tumors: clinical trial for evaluation of safety and feasibility. J Magn Reson Imaging 1998; 8:40-47.[Medline]
  2. Solbiati L, Ierace T, Goldberg SN, et al. Percutaneous US-guided radio-frequency tissue ablation of liver metastases: treatment and follow-up in 16 patients. Radiology 1997; 202:195-203.[Abstract/Free Full Text]
  3. Solbiati L, Goldberg SN, Ierace T, et al. Hepatic metastases: percutaneous radio-frequency ablation with cooled- tip electrodes. Radiology 1997; 205:367-373.[Abstract/Free Full Text]
  4. Livraghi T, Goldberg SN, Monti F, et al. Saline-enhanced radio-frequency tissue ablation in the treatment of liver metastases. Radiology 1997; 202:205-210.[Abstract/Free Full Text]
  5. Anzai Y, Lufkin R, DeSalles A, Hamilton DR, Farahani K, Black KL. Preliminary experience with MR-guided thermal ablation of brain tumors. AJNR Am J Neuroradiol 1995; 16:39-48.[Abstract]
  6. Rossi S, Di SM, Buscarini E, et al. Percutaneous RF interstitial thermal ablation in the treatment of hepatic cancer. AJR Am J Roentgenol 1996; 167:759-768.[Abstract/Free Full Text]
  7. Dupuy DE, Zagoria RJ, Akerley W, Mayo-Smith WW, Kavanagh PV, Safran H. Percutaneous radiofrequency ablation of malignancies in the lung. AJR Am J Roentgenol 2000; 174:57-59.[Free Full Text]
  8. Rosenthal DI, Alexander A, Rosenberg AE, Springfield D. Ablation of osteoid osteomas with a percutaneously placed electrode: a new procedure. Radiology 1992; 183:29-33.[Abstract/Free Full Text]
  9. Rosenthal DI, Springfield DS, Gebhardt MC, Rosenberg AE, Mankin HJ. Osteoid osteoma: percutaneous radio-frequency ablation. Radiology 1995; 197:451-454.[Abstract/Free Full Text]
  10. Rosenthal DI, Hornicek FJ, Wolfe MW, Jennings LC, Gebhardt MC, Mankin HJ. Percutaneous radiofrequency coagulation of osteoid osteoma compared with operative treatment. J Bone Joint Surg Am 1998; 80:815-821.[Abstract/Free Full Text]
  11. de Berg JC, Pattynama PM, Obermann WR, Bode PJ, Vielvoye GJ, Taminiau AH. Percutaneous computed-tomography-guided thermocoagulation for osteoid osteomas. Lancet 1995; 346:350-351.[CrossRef][Medline]
  12. Lindner NJ, Scarborough M, Ciccarelli JM, Enneking WF. CT-controlled thermocoagulation of osteoid osteoma in comparison with traditional methods. Z Orthop Ihre Grenzgeb 1997; 135:522-527[German].[Medline]
  13. Simon MA. Percutaneous radiofrequency coagulation of osteoid osteoma compared with operative treatment. J Bone Joint Surg Am 1999; 81:437-438.[Free Full Text]
  14. Barei DP, Moreau G, Scarborough MT, Neel MD. Percutaneous radiofrequency ablation of osteoid osteoma. Clin Orthop 2000; Apr:115-124.
  15. Dupuy DE. Radiofrequency ablation: an outpatient percutaneous treatment. Med Health R I 1999; 82:213-216.[Medline]
  16. Dupuy DE, Safran H, Mayo-Smith W, Goldberg SN. Percutaneous radiofrequency ablation of painful osseous metastatic disease (abstr). Radiology 1998; 209(P):389.
  17. Tacke J, Adam G, Speetzen R, et al. MR-guided interstitial cryotherapy of the liver with a novel, nitrogen- cooled cryoprobe. Magn Reson Med 1998; 39:354-360.[Medline]
  18. Merkle EM, Boll DT, Boaz TL, et al. MR imaging-guided radiofrequency thermal ablation of implanted VX2 tumors in a rabbit model: feasibility, accuracy, and signal characteristics. Magn Reson Med 1999; 42:141-149.[CrossRef][Medline]
  19. Merkle EM, Shonk JR, Zheng L, Duerk JL, Lewin JS. MR imaging-guided radiofrequency thermal ablation in the porcine brain at 0.2 T. Eur Radiol 2001; 11:884-892.[CrossRef][Medline]
  20. Merkle EM, Haaga JR, Duerk JL, Jacobs GH, Brambs HJ, Lewin JS. MR imaging-guided radiofrequency thermal ablation of the pancreas in a porcine model with a modified clinical C-arm system. Radiology 1999; 213:461-467.[Abstract/Free Full Text]
  21. Merkle EM, Shonk JR, Duerk JL, Jacobs GH, Lewin JS. MR-guided RF thermal ablation of the kidney in a porcine model. AJR Am J Roentgenol 1999; 173:645-651.[Abstract/Free Full Text]
  22. Lewin JS, Duerk JL, Jain VR, Petersilge CA, Chao CP, Haaga JR. Needle localization in MR-guided biopsy and aspiration: effects of field strength, sequence design, and magnetic field orientation. AJR Am J Roentgenol 1996; 166:1337-1345.[Abstract/Free Full Text]
  23. Boaz TL, Lewin JS, Chung YC, Duerk JL, Clampitt ME, Haaga JR. MR monitoring of MR-guided radiofrequency thermal ablation of normal liver in an animal model. J Magn Reson Imaging 1998; 8:64-69.[Medline]
  24. Aschoff AJ, Rafie N, Jesburger JA, Duerk JL, Lewin JS. Thermal lesion conspicuity following interstitial radiofrequency thermal tumor ablation in humans: a comparison of STIR, turbo spin-echo T2-weighted, and contrast-enhanced T1-weighted MR images at 0.2 T. J Magn Reson Imaging 2000; 12:584-589.[CrossRef][Medline]
  25. Aronow S. The use of radio-frequency power in making lesions in the brain. J Neurosurg 1960; 17:431-438.
  26. Goldberg SN, Gazelle GS, Dawson SL, Rittman WJ, Mueller PR, Rosenthal DI. Tissue ablation with radiofrequency: effect of probe size, gauge, duration, and temperature on lesion volume. Acad Radiol 1995; 2:399-404.[CrossRef][Medline]
  27. Lundskog J. Heat and bone tissue: an experimental investigation of the thermal properties of bone and threshold levels for thermal injury. Scand J Plast Reconstr Surg 1972; 9:1-80.[Medline]
  28. Tillotson CL, Rosenberg AE, Rosenthal DI. Controlled thermal injury of bone: report of a percutaneous technique using radiofrequency electrode and generator. Invest Radiol 1989; 24:888-892.[CrossRef][Medline]
  29. Chung YC, Duerk JL, Lewin JS. Generation and observation of radio frequency thermal lesion ablation for interventional magnetic resonance imaging. Invest Radiol 1997; 32:466-474.[CrossRef][Medline]
  30. Goldberg SN, Gazelle GS, Solbiati L, Rittman WJ, Mueller PR. Radiofrequency tissue ablation: increased lesion diameter with a perfusion electrode. Acad Radiol 1996; 3:636-644.[CrossRef][Medline]
  31. Goldberg SN, Solbiati L, Mueller PR, Conrad JE, Hahn PF, Gazelle GS. Radiofrequency ablation: larger volume tissue necrosis using a pulsed, high current technique (abstr). Radiology 1997; 205(P):258.
  32. Ward WG, Eckardt JJ, Shayestehfar S, Mirra J, Grogan T, Oppenheim W. Osteoid osteoma diagnosis and management with low morbidity. Clin Orthop 1993; 291:229-235.
  33. Pennes HH. Analysis of tissue and arterial blood temperatures in the resting human forearm. J Appl Physiol 1948; 1:93-122.[Free Full Text]
  34. Kneisl JS, Simon MA. Medical management compared with operative treatment for osteoid-osteoma. J Bone Joint Surg Am 1992; 74:179-185.[Abstract/Free Full Text]
  35. Amin Z, Donald JJ, Masters A, et al. Hepatic metastases: interstitial laser photocoagulation with real-time US monitoring and dynamic CT evaluation of treatment. Radiology 1993; 187:339-347.[Abstract/Free Full Text]
  36. Gazelle GS, Haaga JR. Guided percutaneous biopsy of intraabdominal lesions. AJR Am J Roentgenol 1989; 153:929-935.[Free Full Text]
  37. Schwabe B, Kahn T, Harth T, Ulrich F, Schwarzmaier HJ. Laser-induced thermal lesions in the human brain: short- and long-term appearance on MRI. J Comput Assist Tomogr 1997; 21:818-825.[CrossRef][Medline]
  38. Silverman SG, Collick BD, Figueira MR, et al. Interactive MR-guided biopsy in an open-configuration MR imaging system. Radiology 1995; 197:175-181.[Abstract/Free Full Text]
  39. Silverman SG, Jolesz FA, Newman RW, et al. Design and implementation of an interventional MR imaging suite. AJR Am J Roentgenol 1997; 168:1465-1471.[Free Full Text]
  40. Jolesz FA. MR-guided thermal ablation of brain tumors. AJNR Am J Neuroradiol 1995; 16:49-52.
  41. Steiner P, Botnar R, Dubno B, Zimmermann GG, Gazelle GS, Debatin JF. Radio-frequency-induced thermoablation: monitoring with T1-weighted and proton-frequency-shift MR imaging in an interventional 0.5-T environment. Radiology 1998; 206:803-810.[Abstract/Free Full Text]
  42. Kahn T, Harth T, Kiwit JC, Schwarzmaier HJ, Wald C, Modder U. In vivo MRI thermometry using a phase-sensitive sequence: preliminary experience during MRI-guided laser-induced interstitial thermotherapy of brain tumors. J Magn Reson Imaging 1998; 8:160-164.[Medline]
  43. Bahner ML, Jenne J, Joechle K, et al. CT als methode zur on-line-temperaturmessung bei minimal-invasiver thermotherapie (abstr). Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1996; 165:314.
  44. Norman A. Persistence or recurrence of pain: a sign of surgical failure is osteoid-osteoma. Clin Orthop 1978; 130:263-266.
  45. Regan MW, Galey JP, Oakeshott RD. Recurrent osteoid osteoma: case report with a ten-year asymptomatic interval. Clin Orthop 1990; 253:221-224.
  46. Assoun J, Richardi G, Railhac JJ, et al. Osteoid osteoma: MR imaging versus CT. Radiology 1994; 191:217-223.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
C. P. Cantwell, J. Kerr, J. O'Byrne, and S. Eustace
MRI Features After Radiofrequency Ablation of Osteoid Osteoma with Cooled Probes and Impedance-Control Energy Delivery.
Am. J. Roentgenol., May 1, 2006; 186(5): 1220 - 1227.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
C. P. Cantwell, J. O'Byrne, and S. Eustace
Radiofrequency ablation of osteoid osteoma with cooled probes and impedance-control energy delivery.
Am. J. Roentgenol., May 1, 2006; 186(5 Suppl): S244 - S248.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
F. K. Wacker, S. G. Nour, R. Eisenberg, J. L. Duerk, and J. S. Lewin
MRI-Guided Radiofrequency Thermal Ablation of Normal Lung Tissue: In Vivo Study in a Rabbit Model
Am. J. Roentgenol., September 1, 2004; 183(3): 599 - 603.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2252001500v1
225/2/471    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Aschoff, A. J.
Right arrow Articles by Lewin, J. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Aschoff, A. J.
Right arrow Articles by Lewin, J. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE