Published online before print November 17, 2005, 10.1148/radiol.2381041500
(Radiology 2005;238:107-112.)
© RSNA, 2005
Osteoid Osteoma in an ex Vivo Animal Model: Temperature Changes in Surrounding Soft Tissue during CT-guided Radiofrequency Ablation1
Rudi G. Bitsch, MD,
Rüdiger Rupp, MSc,
Ludger Bernd, MD, PhD and
Karl Ludwig, MD, PhD
1 From the Department of Radiology, Stiftung Orthopädische Universitätsklinik, Schlierbacher Landstr 200A, 69118 Heidelberg, Germany. Received August 31, 2004; revision requested November 5; revision received January 25, 2005; accepted February 1.
Address correspondence to R.G.B. (e-mail: Rudi_Georg.Bitsch{at}urz.uni-heidelberg.de).
 |
ABSTRACT
|
|---|
Purpose: To assess temperature changes in the soft tissue surrounding bone during radiofrequency (RF) ablation of osteoid osteoma in an ex vivo animal model.
Materials and Methods: Intracortical cavities were created in fresh bovine long bone specimens obtained from a slaughterhouse as models for osteoid osteoma. Three groups of three specimens each were defined according to the thickness (1, 3, and 5 mm) of the cortical bone lamella separating the nidus from the periosteum. Three thermocouples were applied to the soft tissue surrounding the bone in defined distances (0, 5, and 10 mm) from the periosteum. Before RF ablation, the thickness of the cortical bone lamella was documented at computed tomography. Specimens were heated in a 37°C basin. As soon as the measured temperature in the cavity of the specimen reached 35°C, RF ablation was performed for 400 seconds, with a target temperature of 95°C. During RF ablation, continuous measurements were performed simultaneously with digital thermometers. No simulation of vessel perfusion was used. The effect of the thickness of residual osseous lamella and the effect of the distance between the thermocouple and the periosteum were tested with an analysis of variance. Post hoc Bonferroni tests were performed.
Results: Mean maximum temperatures of 69.1°, 51.3°, and 42.5°C for 1-mm lamella; 59.2°, 46.5°, and 41.1°C for 3-mm lamella; and 50.6°, 44.8°, and 40.0°C for 5-mm lamella were measured 0, 5, and 10 mm, respectively, from the periosteum. Significant temperature differences were shown with analysis of variance and post hoc tests for the three groups of bone lamella thickness and distance (P < .001).
Conclusion: In the model of osteoid osteoma, the surrounding temperature (soft tissue) during RF ablation was shown to depend on the thickness of the cortical bone lamella and the distance from the periosteum.
© RSNA, 2005
|
INTRODUCTION
|
|---|
Osteoid osteomas are small benign bone tumors usually found in the lower extremities of children or young adults. They constitute approximately 10% of benign bone tumors and display a low rate of growth (1). The diagnosis of an osteoid osteoma can be made with a high degree of certainty on the basis of clinical and radiologic findings. Histologically, osteoid osteomas are circumscribed nodules of woven bone and osteoid with prominent osteoblastic rimming (the nidus) surrounded by thickened cortical and trabecular bone and loose fibrovascular tissue (the reactive zone) (2). This formation generally appears as a small radiolucent nidus with or without central calcification and with sclerosis of the surrounding bone on radiographs or computed tomographic (CT) scans (3). The location is diaphyseal or metaphyseal; epiphyseal lesions are very rare. Lesions are common in the femur and tibia (4). Osteoid osteomas typically cause nocturnal pain that responds particularly to treatment with salicylates. The pain is mediated by prostaglandins, which are produced by the tumor nidus (2).
Minimal invasive techniques have been percutaneous resection (5), CT-guided drilling (6), and different thermoablative methods of treatment. Radiofrequency (RF) ablation is the best described and most common thermoablative technique. It has become established in recent years as the most effective standard method owing to its relatively low cost and high safety, along with efficacy proved in numerous clinical studies (7–13).
Success and recurrence rates of CT-guided RF ablation are at least equivalent to those of conventional surgical excision methods, with lower complication rates, a shorter time of hospitalization, and faster reconvalescence (14–16).
Despite the many studies of clinical usage, there have been few publications concerning the thermodynamic processes involved in the soft tissue surrounding the bone during the use of RF ablation techniques. Thus, the aim of our study was to assess the temperature changes in the soft tissue surrounding the bone during RF ablation of osteoid osteomas in an ex vivo animal model.
|
MATERIALS AND METHODS
|
|---|
Ex Vivo Animal Model
Slices of fresh bovine tibial bones obtained from a slaughterhouse (Fleischversorgungszentrum, Mannheim, Germany) were used as ex vivo specimens. In each specimen, a nidus was modeled by using a drill hole and a bone stamp. By using a custom-made mounting clamp, it was possible to drill longitudinally at a defined distance from the periosteum. Three groups of three specimens each were defined with a set distance of 1, 3, and 5 mm from the outer limit of the drilled nidus to the periosteum (Fig 1). The drill holes had a diameter of 6.6 mm and a depth of 28 mm. A bone plug of 20 mm in length was obtained from the same fresh slaughtered bovine tibial bone such that its diameter caused it to firmly block the aforementioned drill hole; this was then centrally drilled with a 2-mm diameter bit. After insertion of the bone plug, a cavity of 8 mm in height and 6.6 mm in diameter resulted. This cavity remained accessible via the central 2-mm drill hole in the bone plug and was filled with 0.8% agarose gel. All specimens were manufactured by one author (R.G.B.). Each specimen was examined by another author (K.L.) with a multisection CT (MX8000; Philips Medical Systems, Eindhoven, Netherlands) scanner to validate the size of the cavity and the distance of the cavity from the cortical bone surface (Figs 2, 3).
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1: Drilling of a 5-mm specimen, with mounting for setting of distance between the drill hole and periosteum (marked with Kirschner wire). The cover plate allows drilling in 1-, 2-, and 5-mm (labels) distances depending on its direction. A second cover plate was used to drill in a 3-mm distance.
|
|
After preparation, all specimen slices were placed in a thermostatically controlled bath of physiologic NaCl solution. With the aid of the specially constructed mounting, four thermocouples (Amarell Electronic Laboratory Devices, Kreuz Wertheim, Germany) were placed in or near the specimen. Each thermocouple was calibrated at 40°C and was accurate to ±0.1 C°. In each specimen, one thermocouple was placed in direct contact with the periosteum at the same depth as the RF electrode, and two were placed more radially at 5 and 10 mm, respectively, from the periosteum. A fourth thermocouple was placed in the NaCl solution to monitor the surrounding bath temperature and to make sure that there was no change in the general conditions during RF ablation (Figs 4, 5).
View larger version (61K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4: Diagram of the position of cortical bone cavity and placement of thermocouples parallel to the periosteum, with varying distances (X) per nidus position group between thermocouples and RF (RFA) electrode.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5: Mounting with 1-mm specimen in NaCl bath. Position of RF electrode (white cable) and thermocouples (black cables) is in direct contact with the periosteum and at 5- and 10-mm distances from the periosteum in the surrounding tissue.
|
|
RF Ablation Procedure
The temperature of the bath was controlled with a thermostatic circulator (Multitemp II; LKB Produkter AB, Bromma, Sweden) and was set to 37°C. As soon as the measured temperature in the cavity of the specimen reached 35°C, RF ablation was initiated and conducted for 400 seconds, with a target temperature of 95°C. Two authors (R.G.B., K.L.) performed all of the ablations together; one (K.L.) had 8 years of experience in the clinical use of RF ablation of osteoid osteomas. An RF generator (Neuro N 50; Stryker Leibinger, Freiburg, Germany) with a conventional monopolar needle electrode (TCM 101; Stryker Leibinger, Freiburg, Germany) and a neutral electrode was used with the same settings as for clinical usage. Temperatures were measured (R.R.) at 10-second intervals.
The method just described was conducted with three specimens for each of the three nidus position groups 1, 3, and 5 mm from the outer limit of the drilled nidus to the periosteum.
Statistical Analysis
The primary outcome measure was the maximum soft-tissue temperature. Statistical analysis was performed in three steps. First we tested the distribution of data. This was done graphically with box-and-whisker plots and also quantitatively by comparison of mean and median temperatures. If the shift of the mean and median was less than twice the standard error, a normal distribution was assumed. In a second step, the effect of the thickness of residual osseous lamella and the effect of the distance between the thermocouple and the periosteum was examined with an analysis of variance. Third, post hoc tests (Bonferroni tests) were performed for each group of nidus position and osseous lamella thickness. All tests were two sided, and P 
.05 was considered to indicate significant difference. Data analysis was performed with SPSS for Windows 12.0 (SPSS, Chicago, Ill).
|
RESULTS
|
|---|
In the group of specimens with a cortical lamella thickness of 1 mm, the maximal temperature measured 400 seconds after start of ablation directly at the periosteum was 69.1°C (68.8°–69.3°C). In the same group, the maximal temperature 5 mm from the periosteum was 51.3°C (48.0°–53.0°C), and that at a distance of 10 mm from the periosteum was 42.5°C (40.7°–44.3°C). The graphs of temperatures versus time for all nidus position groups are shown in Figures 6–8. The three specimens from the nidus position group with osseous lamella thickness of 3 mm revealed the following maximal temperatures: directly at periosteum, 59.2°C (58.0°–60.5°C); at 5 mm, 46.5°C (46.1°–47.4°C); and at 10 mm, 41.1°C (40.2°–41.6°C). The three specimens with osseous lamella thickness of 5 mm had the following maximal temperatures: directly at periosteum, 50.6°C (48.2°–52.3°C); at 5 mm, 44.8°C (44.2°–46.8°C); and at 10 mm, 40.0°C (39.9°–40.0°C). The analysis of variance demonstrated a significant influence of bone lamella thickness (P < .001) and thermocouple distance (P < .001) on the maximal temperatures. Significant maximal temperature differences were also shown with the post hoc tests (Bonferroni tests) for each of the three groups of bone lamella thickness and thermocouple distance (P < .001).
The target temperature of 95°C was attained with the RF electrode within 20–30 seconds. After the target temperature had been reached, it was maintained at 90°–100°C. After the RF electrode was turned off, temperatures decreased at the periosteum in each group of specimens to under 45°C within the first 60 seconds.
|
DISCUSSION
|
|---|
To our knowledge, RF ablation was first described for osteoid osteomas by Rosenthal et al (17) in 1992. Findings of clinical studies with up to 271 ablation procedures have shown that RF ablation is a safe and effective technique (16). There have been few publications, however, concerning the heat distribution and heat transport during RF ablation. Thermoablation of osteoid osteoma in clinical practice causes heat conduction through several heterogeneous tissues; thus, adequate modeling of this process is complex. One ex vivo study on heat conduction was conducted within bone to assess the feasibility of RF ablation in large bone tumors (18). One in vivo-ex vivo study was performed to detect temperature distribution in the spinal canal during RF ablation in a vertebral body (19). In contrast to these studies, we investigated RF ablation and the resulting temperature distributions in the vicinity of long bones. Small bone cavities were used to model the nidus of osteoid osteomas. Temperature was measured in the soft tissue beyond the periosteal border. In vivo temperature measurements during cemented total hip replacements resulted in ambient bone temperatures of 27°–35°C in the acetabulum and of 22°–35°C in the human femur (20). Thus, 35°C was used as the starting point for our measurements.
As described by Froese et al (21), the effects of temperature increases on neural structures in the spinal cords of mice indicate a 50% damage rate for nerve tissue after warming to 43.1°C for 1 hour. The same damage rate was achieved at 45°C within 10.8 minutes; in conclusion, a time reduction factor of 2.25 was found to occur for comparable nerve tissue damage for every degree increase in temperature (21). Damage to nerve tissue at 45°C has also been shown for peripheral nerves (22).
The results show temperatures directly at the periosteum of 50°–70°C for several minutes, which would imply significant damage to nerve tissue. Temperatures only decreased below the 45°C mark at 10 mm from the periosteum in all specimens without perfusion. Only the 5-mm lamella thickness group did not reach this mark at 5 mm from the periosteum.
Our experiment involved a complex heat dissipation pattern through several heterogeneous tissues with varying heat conduction characteristics (eg, agar, bone, periosteum, muscle tissue), which complicates somewhat a direct comparison with similar studies that only dealt with heat dissipation in bone alone. Thermoablation of osteoid osteoma in clinical practice, however, also causes heat conduction from the thrombotic nidus through the surrounding sclerotic bone to the surrounding bone and muscle tissue.
Rachbauer et al (18) also investigated heat conduction and dissipation in cortical bone during RF ablation and found mean temperatures of 64.33°C at 5 mm from the thermocouple and of 41.67°C at 10 mm from the thermocouple. Despite taking into consideration that the thermocouples in the aforementioned study were located in the bone and not soft tissue, the results found were not dissimilar to ours.
Dupuy et al (19) conducted spinal osteoid osteoma ablations in patients, as well as in living pigs and bone-agar phantoms. Temperatures of 48°C and 41°C in porcine vertebra were measured at 5 and 10 mm, respectively, from the RF ablation electrode; further, epidural temperatures of 44°C were found. These results were obtained under intact perfusion conditions of cancellous vertebral bone, along with the cooling effect of cerebrospinal fluid and epidural venous plexus, and can also be regarded as comparable to our results.
Our study had limitations in that some aspects of our model differed from the conditions found during clinical usage. First, the cooling effects of perfusion could not be simulated in the chosen model. Heat losses due to perfusion in large vessels can affect thermoablation lesions significantly (23). There are, however, no large vessels in the cortical bone, and blood flow in the region of a thermoablation lesion is restricted by heat-induced thrombosis in living tissue. Vessels less than 3 mm in diameter show endothelial cell necrosis in 100% of cases and a luminal thrombus in histologic studies (24). A more exact assessment of the effect of cooling by perfusion could not be found in the literature and remains difficult to quantify. The periosteum and muscle tissue with their higher perfusion rates can be regarded as tissues in which heat loss due to perfusion is an important factor compared with that in the bone. The measured temperature values are therefore higher than those that could be expected to be found in clinical use and can be regarded as an upper limit or worst-case scenario.
Second, specimens were obtained from fresh slaughtered cows, which might differ in some physical characteristics from human bone tissue (25). Thermal characteristics of bone are mainly dependent on their content of water, protein, fat, and minerals. Bovine bone is harder and has a lower water content than human bone (26). Heat dissipation in the tissue is promoted by higher water content, which causes greater heat flow to be necessary for an increase in temperature. Surrounding an osteoid osteoma there tends to be a zone of thickened, sclerotic, and trabecular cortical bone rather than normal human bone tissue (1). Thermal characteristics of such sclerotic human bone tissues have not, to our knowledge, been investigated; however, the water content is most probably reduced and, therefore, the heat conduction is worse.
A further limitation was that recording of tissue temperatures was time coupled to our RF generator. We therefore were unable to continuously extend the time axis of Figures 6–8 after the RF electrode was turned off to show how quickly the temperatures returned to baseline.
Despite these limitations, our study was the first of which we are aware to analyze heat flow and conduction phenomena from a bone cavity in long bones to the surrounding soft tissue during RF ablation. It can be concluded that the maximal temperatures attained in the surrounding soft tissues are dependent on the thickness of the cortical bone, the thickness of the osseous lamella between the RF electrode and the periosteum, and the duration of RF ablation. We consider the maximal temperatures found in this study to provide useful parameters for assessing the effects of RF ablation in clinical practice.
Practical application: Although the precise degree of temperature modulation due to perfusion in the periosteum and muscle tissue during thermoablation cannot yet be evaluated, we recommend the maintenance of a 10-mm safety distance between the periosteum and the nearest neural structure in RF ablation procedures in extremities with an osseous lamella thickness of less than 5 mm between the RF electrode and the periosteum.
Further studies that include evaluation of the cooling effects of perfusion are necessary to enable a more precise recommendation for safe distances from neural structures in RF ablation procedures.
|
ACKNOWLEDGMENTS
|
|---|
Special thanks are due to Gerhard Buchmann, Department of Orthopedic Research, University of Heidelberg, Germany, for manufacturing the experimental setting and Sven Schneider, Department of Orthopedic Research, University of Heidelberg, Germany, for the statistical analyses.
|
FOOTNOTES
|
|---|
Abbreviations: RF = radiofrequency
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, L.B., K.L.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, R.G.B., K.L.; experimental studies, R.G.B., R.R.; statistical analysis, R.G.B.; and manuscript editing, R.G.B., L.B., K.L.
|
References
|
|---|
- Greenspan A. Benign bone-forming lesions: osteoma, osteoid osteoma, and osteoblastoma—clinical, imaging, pathologic, and differential considerations. Skeletal Radiol 1993;22:485–500.[Medline]
- O'Connell JX, Nanthakumar SS, Nielsen GP, Rosenberg AE. Osteoid osteoma: the uniquely innervated bone tumor. Mod Pathol 1998;11:175–180.[Medline]
- Cerase A, Priolo F. Skeletal benign bone-forming lesions. Eur J Radiol 1998;27(suppl 1):S91–S97.[CrossRef][Medline]
- Levine SM, Lambiase RE, Petchprapa CN. Cortical lesions of the tibia: characteristic appearances at conventional radiography. RadioGraphics 2003;23:157–177.[Abstract/Free Full Text]
- Sans N, Galy-Fourcade D, Assoun J, et al. Osteoid osteoma: CT-guided percutaneous resection and follow-up in 38 patients. Radiology 1999;212:687–692.[Abstract/Free Full Text]
- Adam G, Neuerburg J, Vorwerk D, Forst J, Gunther RW. Percutaneous treatment of osteoid osteomas: combination of drill biopsy and subsequent ethanol injection. Semin Musculoskelet Radiol 1997;1:281–284.[Medline]
- Barei DP, Moreau G, Scarborough MT, Neel MD. Percutaneous radiofrequency ablation of osteoid osteoma. Clin Orthop Relat Res 2000;373:115–124.[CrossRef][Medline]
- Ghanem I, Collet LM, Kharrat K, et al. Percutaneous radiofrequency coagulation of osteoid osteoma in children and adolescents. J Pediatr Orthop B 2003;12:244–252.[CrossRef][Medline]
- Lindner NJ, Ozaki T, Roedl R, Gosheger G, Winkelmann W, Wortler K. Percutaneous radiofrequency ablation in osteoid osteoma. J Bone Joint Surg Br 2001;83:391–396.[CrossRef][Medline]
- Simon MA. Percutaneous radiofrequency coagulation of osteoid osteoma compared with operative treatment. J Bone Joint Surg Am 1999;81:437–438.[Free Full Text]
- Torriani M, Rosenthal DI. Percutaneous radiofrequency treatment of osteoid osteoma. Pediatr Radiol 2002;32:615–618.[CrossRef][Medline]
- Woertler K, Vestring T, Boettner F, Winkelmann W, Heindel W, Lindner N. Osteoid osteoma: CT-guided percutaneous radiofrequency ablation and follow-up in 47 patients. J Vasc Interv Radiol 2001;12:717–722.[Medline]
- Venbrux AC, Montague BJ, Murphy KP, et al. Image-guided percutaneous radiofrequency ablation for osteoid osteomas. J Vasc Interv Radiol 2003;14:375–380.[Medline]
- 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]
- Vanderschueren GM, Taminiau AH, Obermann WR, Bloem JL. Osteoid osteoma: clinical results with thermocoagulation. Radiology 2002;224:82–86.[Abstract/Free Full Text]
- Rosenthal DI, Hornicek FJ, Torriani M, Gebhardt MC, Mankin HJ. Osteoid osteoma: percutaneous treatment with radiofrequency energy. Radiology 2003;229:171–175.[Abstract/Free Full Text]
- 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]
- Rachbauer F, Mangat J, Bodner G, Eichberger P, Krismer M. Heat distribution and heat transport in bone during radiofrequency catheter ablation. Arch Orthop Trauma Surg 2003;123:86–90.[Medline]
- Dupuy DE, Hong R, Oliver B, Goldberg SN. Radiofrequency ablation of spinal tumors: temperature distribution in the spinal canal. AJR Am J Roentgenol 2000;175:1263–1266.[Free Full Text]
- Toksvig-Larsen S, Franzen H, Ryd L. Cement interface temperature in hip arthroplasty. Acta Orthop Scand 1991;62:102–105.[Medline]
- Froese G, Das RM, Dunscombe PB. The sensitivity of the thoracolumbar spinal cord of the mouse to hyperthermia. Radiat Res 1991;125:173–180.[Medline]
- Letcher FS, Goldring S. The effect of radiofrequency current and heat on peripheral nerve action potential in the cat. J Neurosurg 1968;29:42–47.[Medline]
- Goldberg SN, Hahn PF, Tanabe KK, et al. Percutaneous radiofrequency tissue ablation: does perfusion-mediated tissue cooling limit coagulation necrosis? J Vasc Interv Radiol 1998;9:101–111.
- Lu DS, Raman SS, Vodopich DJ, Wang M, Sayre J, Lassman C. Effect of vessel size on creation of hepatic radiofrequency lesions in pigs: assessment of the "heat sink" effect. AJR Am J Roentgenol 2002;178:47–51.[Abstract/Free Full Text]
- Biyikli S, Modest MF, Tarr R. Measurements of thermal properties for human femora. J Biomed Mater Res 1986;20:1335–1345.[CrossRef][Medline]
- Clattenburg R, Cohen J, Conner S, Cook N. Thermal properties of cancellous bone. J Biomed Mater Res 1975;9:169–182.[CrossRef][Medline]
TOP
ABSTRACT
INTRODUCTION
