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


     

DOI: 10.1148/radiol.2302021056
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
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 Google Scholar
Google Scholar
Right arrow Articles by Nour, S. G.
Right arrow Articles by Strauss, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nour, S. G.
Right arrow Articles by Strauss, M.
(Radiology 2004;230:359-368.)
© RSNA, 2004

Experimental Studies

Percutaneous MR Imaging– guided Radiofrequency Interstitial Thermal Ablation of Tongue Base in Porcine Models: Implications for Obstructive Sleep Apnea Syndrome1

Sherif Gamal Nour, MD, Jonathan S. Lewin, MD, Michael Gutman, MD, Claudia Hillenbrand, PhD, Frank K. Wacker, MD, John W. Wong, MD, Ian C. Mitchell, BSc, Charles B. Armstrong, BSc, Mufaddal M. Hashim, MD, Jeffrey L. Duerk, PhD and Melvin Strauss, MD

1 From Depts of Radiology (S.G.N., J.S.L., C.H., F.K.W., I.C.M., C.B.A., M.M.H., J.L.D.), Otolaryngology (M.G., M.S.), Pathology (J.W.W.), and Biomedical Engineering (J.S.L., J.L.D.), Univ Hosp of Cleveland/Case Western Reserve Univ School of Med, 11100 Euclid Ave, Cleveland, OH 44106-5056 and Dept of Diagnostic Radiology, Cairo Univ Hosp, Egypt (S.G.N.). Supported in part by Siemens Medical Systems and Radionics and grants from Whitaker Foundation, American Cancer Society, and NIH RO1 CA-81431–01, R01-CA84433, R33 CA/AG 88144–01, and P20 CA91710–01. Received Aug 20, 2002; revision requested Oct 7; revision received Apr 23, 2003; accepted Jun 19. Address correspondence to J.S.L. (e-mail: lewin@uhrad.com).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To test the feasibility and safety of a percutaneous magnetic resonance (MR) imaging–guided technique for radiofrequency (RF) interstitial thermal ablation of the tongue base and to correlate MR appearance of induced thermal lesions with histopathologic findings in pigs in acute and chronic porcine models.

MATERIALS AND METHODS: A 1-cm-tip RF electrode was inserted percutaneously into the tongue in 10 pigs with 0.2-T real-time MR guidance. The RF electrode was advanced up the midline between lingual arteries and stopped short of tongue mucosa. RF interstitial thermal ablation was performed at 90°C ± 2 and lasted 10 minutes. Postablation images were obtained with a 1.5-T MR imager. Five pigs were sacrificed immediately (acute model), while five were followed up for 1 month (chronic model) before they were sacrificed. MR-compatible fiducial coils were inserted into tongues with MR imaging guidance prior to RF ablation in the chronic group. Tongues were harvested for histopathologic analysis. Mean thermal lesion volume was compared with the Student t test on images obtained immediately, 2 weeks, and 1 month after RF ablation. Interclass correlation coefficients of lesion diameters at gross pathologic analysis and corresponding diameters with each pulse sequence were calculated.

RESULTS: Successful MR imaging–guided electrode positioning was achieved in all procedures without intra- or postprocedure complications because there was high vascular conspicuity and tissue contrast. Thermal lesions appeared hypointense with hyperintense surrounding rims with all sequences in both groups. At pathologic analysis, acute lesions appeared as pale necrotic areas surrounded by hyperemic rims, while chronic lesions demonstrated progressive circumferential fibrosis and significant volume shrinkage (P < .01). Thermal lesion diameters measured at gross pathologic analysis best agreed with corresponding diameters measured on short inversion time inversion-recovery images (interclass correlation coefficient = 0.85).

CONCLUSION: The results of this investigation demonstrate MR imaging–guided RF interstitial thermal ablation of the tongue base is feasible and safe and illustrate imaging and pathologic phenomena associated with creation and evolution of the induced thermal lesions.

© RSNA, 2004

Index terms: Magnetic resonance (MR), interventional procedures, 262.12149 • Sleep apnea, 262.827 • Radiofrequency (RF) ablation • Tongue, 262.12149


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Obstructive sleep apnea (OSA) syndrome is a well-recognized risk factor for motor vehicle accidents and is a leading cause of loss of productivity because of associated daytime somnolence. Losses in productivity attributed to OSA have been estimated to reach more than $20 billion per year in the United States (1). Other associated conditions that have been linked to OSA include gastroesophageal reflux, impotence, cardiac arrhythmias, hypertension, and increased risk of stroke and myocardial infarction (2,3). Various conservative and surgical approaches have been developed to treat OSA; however, management of this disorder is often complex, and the success rate of uvulopalatopharyngoplasty—which is the most popular surgical approach—is as low as 50% (1).

Radiofrequency (RF) interstitial thermal ablation of the tongue base has recently been reported as a treatment for OSA. This technique produces scar tissue contraction at the site of the induced thermal lesion, thereby reducing tongue base volume and alleviating airway narrowing (4,5). Current limitations of RF interstitial thermal ablation of the tongue base include its blind application through the open mouth with the potential risk of neurovascular injury, mucosal damage, postoperative pain, and infection; an inability to preselect the exact ablation site; and the difficulty of precise RF electrode navigation through the relatively tight space of the open mouth.

The aims of this investigation were to test the feasibility and safety of percutaneous magnetic resonance (MR) imaging–guided RF interstitial thermal ablation of the tongue base and to correlate the MR imaging appearance of induced thermal lesions with histopathologic findings in pigs in the acute and the chronic group.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
In accordance with a protocol that was approved by our Institutional Animal Care and Use Committee, 10 American farm pigs (25–35 kg) were divided equally into two experimental groups (eg, the acute and chronic groups).

Acute Porcine Model (Nonsurvival Experiments)
The procedures were performed entirely within an interventional suite with a 0.2-T open C-arm MR imaging system (Magnetom Open; Siemens Medical Solutions, Erlangen, Germany). Researchers were able to simultaneously operate the imager and view images at the imager side with an in-room high-resolution 1,024 x 1,280-pixel RF-shielded liquid crystal monitor controlled with an MR-compatible mouse and foot pedal.

A 21-cm-diameter, belt-shaped solenoid surface receiver coil (Siemens Medical Solutions) was used to allow room for RF electrode manipulation.

Induction of anesthesia was achieved with intramuscular injection of 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 Pharmaceutical, St Joseph, Mo). This was followed by intravenous administration of thiopental sodium (15 mg/kg, Pentothal; Abbott Laboratories, North Chicago, Ill) to facilitate tracheal intubation. Inhalation anesthesia was maintained during all procedures by using halothane 1% (Halocarbon Laboratories, River Edge, NJ). No premedications were administered prior to the procedures.

Animals were placed in the right lateral decubitus position with the shoulder regions shaved and covered with two 8 x 12-cm wire mesh grounding pads (Radionics, Burlington, Mass), one on each side, to serve as a return path for the RF ablation current.

A custom-made 1-cm exposed tip, 17-gauge MR-compatible monopolar titanium RF electrode (Radionics) was then inserted percutaneously with near real-time MR imaging guidance by using a rapid gradient-echo sequence, usually fast imaging with steady-state precession (repetition time msec/echo time msec, 17.8/8.1; flip angle, 90°; number of signals acquired, three), with a temporal resolution of three frames per 22 seconds or fast low-angle shot imaging (93.0/12.7; flip angle, 60°; number of signal averages, two) with a temporal resolution of three frames per 17 seconds. The electrode was advanced while aiming at the midline plane between the genioglossus and geniohyoid muscle complexes and avoiding the lingual arteries, as depicted on the coronal image (Fig 1). Once within the tongue musculature, sagittal images centered on the RF electrode were obtained to plan further manipulation of the electrode tip along the anteroposterior dimension of the tongue. The electrode was then advanced in the desired trajectory and stopped short of the tongue mucosa.



View larger version (74K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1.  A, Coronal diagram of the percutaneous approach for tongue base ablation shows the RF electrode (13) inserted percutaneously through the chin and advanced cranially in a strict midline trajectory to reach the tongue base without puncturing the tongue mucosa or risking any of the vital structures at the floor of the mouth. 1 = mucous membrane of tongue, 2 = tongue muscles, 3 = genioglossus muscle, 4 = geniohyoid muscle, 5 = hyoglossus muscle, 6 = sublingual gland, 7 = lingual vessels and nerve, 8 = mylohyoid muscle, 9 = body of mandible, 10 = submandibular gland, 11 = platysma muscle, 12 = skin. B, Coronal MR image from a series of images obtained with fast imaging with steady-state precision techniques (17.8/8.1; flip angle, 90°; number of signals acquired, three) with a temporal resolution of three frames per 22 seconds acquired during MR fluoroscopic guidance of the RF electrode into the base of the tongue. Note the lingual arteries (straight arrows) and surface mucosa (arrowheads). The RF electrode (curved arrow) can be stopped short of the mucosa, and the induced thermal lesion can be planned to lay entirely with the muscular portion of the tongue base.

 
RF interstitial thermal ablation was performed in four of these five animals and lasted 10 minutes. A 200-W RF generator (RFG-3C; Radionics) and a standard noncooled technique with the RF electrode tip temperature kept at 90°C ± 2 were used. Various parameters (ie, RF current, power, tissue impedance, and electrode tip temperature) were recorded at 1-minute intervals throughout each ablation procedure. The first pig in the acute group was used as a control subject to test our ability to use MR imaging guidance to insert a 17-gauge RF-electrode percutaneously into the tongue base and to optimize our imaging parameters and experimental setup.

Immediately after ablation, images were obtained with T2-weighted imaging (2,600/96; echo train length, seven; number of signal averages, seven; field of view, 220 x 220 mm; matrix, 256 x 256 pixels; section thickness, 3 mm), short inversion time inversion-recovery (STIR) imaging (2,700/48; echo train length, seven; number of signal averages, seven; field of view, 220 x 220 mm; matrix, 256 x 256 pixels; section thickness, 3 mm), and pre- and postcontrast (0.2 mL/kg gadodiamide Omniscan; Nycomed, Princeton, NJ) T1-weighted imaging (528/26; flip angle, 90°; number of signal averages, four; field of view, 220 x 220 mm; matrix, 256 x 256 pixels; section thickness, 3 mm).

Animals were then immediately sacrificed with intravenous administration of pentobarbital sodium (0.22 mL/kg Euthasol; Diamond Animal Health; Des Moines, Iowa) with a concentration of 390 mg/mL. Their necks were then dissected, and the tongues were harvested for histopathologic analysis. The formalin-fixed tongues were then sliced sagittally into 3-mm-thick slices with a custom-made slicing apparatus while attempting to cut exactly through the plane of the RF electrode track. Tissue faces were then photographed, and the maximum thermal lesion diameter perpendicular to the RF electrode track was measured on gross pathologic images (by J.W.W.) and on each of the postablation MR images (by S.G.N.). Induced thermal lesion volumes were calculated on the MR images by using the ellipsoid formula: volume = (4{pi}/3)(d1/2)(d2/2)(d3/2), where d1, d2, and d3 are the diameters of the thermal lesion.

Chronic Porcine Model (Survival Experiments)
All procedures in a chronic model were performed (by S.G.N., M.G., C.H., F.K.W., C.B.A.) with general intravenous anesthetic to avoid the unreproducible deforming effect of an endotracheal tube on the tongue base when comparing induced thermal lesion volumes on subsequent follow-up images. Anesthesia was achieved with intramuscular injection of a mixture of tiletamine hydrochloride and zolazepam hydrochloride (4–6 mg/kg, Telazol; Lederle Parenterals, Carolina, Puerto Rico) with a concentration of 100 mg/mL. Anesthesia was maintained with continuous intravenous infusion of xylazine (2 mg/kg Xyla-Ject; Phoenix Pharmaceutical) with a concentration of 20 mg/mL and ketamine hydrochloride (20 mg/kg) with a concentration of 100 mg/mL. Again, no medications were administered prior to the procedures.

Experiments in a chronic model were begun by acquiring preliminary baseline images of the tongue with a 1.5-T MR imager (Sonata; Siemens Medical Solutions). Imaging consisted of T2-weighted imaging (4,000/99; echo train length, 11; number of signals acquired, one; field of view, 220 x 220 mm; matrix, 256 x 256 pixels; section thickness, 3 mm), STIR imaging (5,300/35; echo train length, seven; number of signals acquired, one; field of view, 220 x 220 mm; matrix, 256 x 256 pixels; section thickness, 3 mm), pre- and postcontrast T1-weighted imaging (539/13; flip angle, 90°; number of signals acquired, one; field of view, 220 x 220 mm; matrix, 256 x 256 pixels; section thickness, 3 mm), and three-dimensional construction interference in steady-state (CISS) imaging (10.4/5.2; flip angle, 64°; number of signals acquired, two; field of view, 210 x 210 mm; matrix, 256 x 256 pixels; section thickness, 0.8 mm).

The pigs were then moved to the same interventional MR suite used for the experiments in an acute model. The same experimental setup was used and animals were placed in the same positions, but an aseptic technique was used and experiments in chronic models began with insertion of an MR-compatible fiducial coil (MWCE-18–1.0–3-Hilal; Cook, Bloomington, Ind) into the tongue musculature to mark the approximate junction between the posterior and middle thirds of the tongue for future reference, as the pig would mature during the follow-up period. These coils were deployed through a 22-gauge 10-cm MR-compatible biopsy needle (E-Z-Em, Westbury, NY) inserted percutaneously with direct MR fluoroscopic guidance by using the same gradient-echo sequences described earlier. RF ablation of the tongue base was then performed in all five animals in this group with the same RF electrode, generator, and technique used earlier. After RF ablation, the pigs were transferred back to the 1.5-T MR imager, where imaging was repeated with the same parameters used to obtain preablation control images.

The pigs were then returned to the animal facility, where they remained during a 1-month follow-up period. Tolerance of the procedure and delayed complications were evaluated by a full-time veterinarian who observed the pigs on a daily basis during the 1st week after RF interstitial ablation for signs indicating (1) pain, such as vocalization, lethargy, or both, (2) loss of appetite, or (3) defective defecation. MR imaging was performed at 2-week and 1-month follow-up with the 1.5-T MR imager and the same pulse sequences and imaging parameters used for baseline and immediate postablation imaging.

All five animals in the chronic group were sacrificed 1 month after RF interstitial ablation, and their tongues were harvested and processed in the same way as in the acute group. Again, maximum short-axis thermal lesion diameter was measured on each of the images obtained immediately, 2 weeks, and 1 month after RF interstitial ablation (by S.G.N.) and on the gross pathologic specimens (by J.W.W.). Thermal lesion volumes were calculated on the immediate postablation images and on each of the follow-up MR images by using the ellipsoid formula.

Data and Statistical Analysis
Feasibility and safety.—In both groups, feasibility of the MR imaging–guided technique was determined by assessing the ability to place the RF electrode accurately into the tongue base with the sole guidance of MR fluoroscopy. Electrode placement was considered successful when the 1-cm exposed electrode tip was depicted as centered within the targeted part of the tongue base on confirmatory T1-weighted images (Fig 2).



View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2.  A, Coronal and, B, sagittal spin-echo T1-weighted MR images (528/26; flip angle, 90°; number of signals acquired, four) confirm the position of the RF electrode prior to ablation. Note the smaller artifact associated with the RF electrode (curved arrows) on these spin-echo images compared with the gradient-echo image (Fig 1, B), although both were acquired with the phase-encoding direction perpendicular to the electrode shaft. Note also the delineation of surface mucosa (arrowheads) and lingual arteries (straight arrows).

 
The mean time spent for interactive MR imaging–guided RF electrode insertion starting from preliminary orientation imaging and lasting until the electrode position was deemed satisfactory on confirmation images was calculated (by S.G.N.). The means of RF current and of tissue impedance recorded (by I.C.M. or C.B.A.) during the 10-minute ablation procedures were calculated (by S.G.N.).

Short-term safety of the procedure was evaluated (by S.G.N.) on the basis of the presence or absence of evidence of complications such as vascular injury, hematoma, or mucosal damage in both groups of pigs on images obtained during or immediately after the procedure.

Long-term safety of the procedure was estimated in the chronic group by observing the animals’ tolerance of the procedure and by reviewing follow-up MR images (S.G.N.) and gross pathologic specimens (J.W.W.) for delayed complications such as infection, vascular injury, or mucosal disruption.

MR imaging and histopathologic correlation.—The MR appearance of induced thermal lesions, including lesion shape, margin definition, signal characteristics, and enhancement pattern, was recorded (by S.G.N.) for the acute and chronic groups. The corresponding pathologic specimens were examined both grossly and microscopically (with hematoxylin-eosin and trichrome staining) for evidence of cell death, fibrosis, hemorrhage, and vascular or mucosal damage. Pathologic specimens were examined by a pathologist (J.W.W.) who was blinded to whether the specimen belonged to the acute or chronic study group.

To evaluate the performance of the various implemented pulse sequences in the prediction of actual thermal lesion size, we calculated the interclass correlation coefficients of the maximum short-axis diameters of acute and chronic lesions at gross pathologic analysis and the corresponding diameters, as measured with each pulse sequence, on MR images obtained immediately after the procedure and at 1-month follow-up, respectively.

The mean volume of acute thermal lesions was calculated for each of the immediate postablation images by using the ellipsoid formula, as described previously. For pigs in the chronic group, mean thermal lesion volume on the 1-month follow-up postablation images was compared with the mean thermal lesion volume on the 2-week and immediate postablation images (by S.G.N.).

Thermal lesion conspicuity was also compared for the various pulse sequences by calculating the lesion-to-tongue con-trast-to-noise (CNR) ratio (S.G.N.) with each pulse sequence on the immediate postablation and follow-up images by using the formula CNR = (SAL - SAT)/SDN, where SAL is the signal amplitude of the lesion, SAT is the signal amplitude of the part of the tongue that did not receive ablation, and SDN is the standard deviation of the signal amplitude of the background noise. Background noise was measured along the phase-encoding direction on the same images used to obtain lesion and tongue signal amplitude measurements. The diameters of the regions of interest used to measure these signal amplitudes were chosen to encompass the largest possible artifact-free parts of the areas being evaluated. The areas of regions of interest were approximately 0.2–0.4 cm2 for thermal lesions, 0.9–1.6 cm2 for adjacent tongue muscles, and 1.7–17.0 cm2 for background noise.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Feasibility and Safety
MR imaging–guided positioning of the RF electrode in the desired portion of the tongue base was successful in all procedures in both groups, and there was no MR evidence of short-term complications such as vascular injury, hematoma, or mucosal damage. The high vascular conspicuity offered by fast gradient-echo imaging allowed safe and confident advancement of the RF electrode along the midline plane away from the lingual arteries. Furthermore, soft-tissue contrast was high enough to permit sufficient identification of the mucous membrane covering the tongue. Thus, we were able to plan the developing thermal lesions to avoid mucosal injury (Fig 1, B).

The mean time (± SD) required for interactive MR imaging–guided RF electrode insertion was 24 minutes ± 0.01. The mean RF current applied during the nine 10-minute ablation procedures was 0.18 A ± 0.03, whereas the mean tissue impedance was 110 {Omega} ± 22.8.

Clinical and imaging follow-up of pigs in the chronic group showed that the procedure was well tolerated by all animals. No pigs required analgesia after the procedure, and all began to eat after recovering from anesthesia. All animals resumed normal defecation the next morning. No imaging or pathologic evidence of delayed complications such as infection, vascular injury, or mucosal disruption was noted.

MR Imaging and Histopathologic Correlation
Acute porcine models (nonsurvival experiments).—Induced thermal lesions had well-defined ovoid configurations, with their long axes oriented along the RF electrode tracks. These lesions appeared hypointense with all pulse sequences, with hyperintense surrounding rims of reactive tissue changes on T2-weighted and STIR images and marginal enhancement on postcontrast T1-weighted images (Fig 3, A–C). The RF-electrode tracks had a linear hyperintense appearance bisecting thermal lesions on T2-weighted and STIR images (Fig 4).



View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3.  Acute thermal lesion of the tongue base. A, Sagittal contrast-enhanced spin-echo T1-weighted (528/26; flip angle, 90°; number of signals acquired, four) MR image. B, Fast spin-echo T2-weighted (2,600/96; echo train length, seven; number of signals acquired, seven) MR image. C, Fast spin-echo STIR (2,700/48; echo train length, seven; number of signals acquired, seven) MR image. All images were acquired immediately after RF ablation with a low-field-strength (0.2-T) MR imager. Images show the typical MR appearance of acute thermal lesions of the tongue base, being hypo- or isointense at all pulse sequences and surrounded by hyperintense rims of reactive tissue changes on the T2-weighted and STIR images (arrowheads in B and C), with marginal enhancement on the postcontrast image (arrowheads in A). Note the RF electrode track (arrow in C). D, Corresponding gross pathologic specimen shows pale area of coagulation necrosis surrounded by well-defined dark hyperemic and hemorrhagic rim (arrowheads) and traversed by hemorrhagic puncture wound (arrow). E, Trichrome-stained histologic section of the same lesion shows basophilic staining of the muscle fibers at the site of acute thermal injury (arrowheads) and intact tongue mucosa. Examination of the corresponding hematoxylin-eosin-stained specimen (not shown) yielded less impressive cellular changes and barely defined thermal lesion boundaries. F, Magnified section of E demonstrates contraction banding of the myofibrils (arrowheads), which is a sign of acute muscle injury.

 


View larger version (70K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4.  Chronic thermal lesion of the tongue base. A, Sagittal MR image obtained with high-resolution three-dimensional CISS (10.4/5.2; flip angle, 64°; number of signals acquired, two) and a high-field-strength (1.5-T) MR imager immediately after RF ablation in a chronic porcine model. Thermal lesion appears isointense to the adjacent intact tongue muscles and is surrounded by hyperintense reactive tissue changes (arrowheads). The RF electrode track is well defined as a hyperintense line (straight arrow). Note the MR-compatible fiducial coil (curved arrow) and the hyperintense line extending from the floor of the mouth to the site of the coil, indicating the track of the deployment needle. B, Maximum intensity projection image constructed from three-dimensional CISS data demonstrates topographic location of the thermal lesion (curved arrow) relative to the lingual arteries (straight arrow).

 
At gross pathologic analysis, an acute thermal lesion appeared as a pale necrotic area surrounded by a thin hypointense rim of hyperemia and hemorrhage (Fig 3, D). Microscopic examination showed evidence of acute muscle injury exemplified by aggregation of the cytoplasmic proteins and contraction banding of the myofibrils (Fig 3, F). The blood vessels within the injured area showed mural necrosis with infiltration by polymorphs. Some vessels showed luminal thrombosis. Trichrome staining showed basophilic staining of the injured myocytes with no evidence of healing or organization (Fig 3, E). No evidence of mucosal injury was noticed.

The mean volume of acute thermal lesions was 0.92 mL ± 0.85 on contrast-enhanced T1-weighted images, 1.51 mL ± 0.87 on T2-weighted images, and 0.99 mL ± 0.93 on STIR images.

The mean lesion-to-tongue CNR, as measured on immediate postablation images acquired with the 0.2-T imager, was higher with contrast-enhanced T1-weighted imaging (548 ± 189) than with T2-weighted imaging (235 ± 145) or STIR imaging (21 ± 5).

Chronic porcine models (survival experiments).—Thermal lesions had essentially the same MR signal intensity characteristics that were observed in the acute porcine models. These lesions had hypointense signal intensity with all sequences, with hyperintense surrounding rims on T2-weighted, STIR, and three-dimensional CISS (Fig 4) images and marginal enhancement on contrast-enhanced images.

Induced thermal lesions in chronic porcine models showed significant volume shrinkage during the 1-month follow-up period. The mean thermal lesion volume immediately after ablation was 2.9 mL ± 0.6 on contrast-enhanced T1-weighted images, 3.4 mL ± 0.7 on T2-weighted images, and 3.0 mL ± 0.5 on STIR images. The mean lesion volume at 2-week follow-up was 1.2 mL ± 1.0 on contrast-enhanced T1-weighted images, 1.1 mL ± 1.2 on T2-weighted images, and 1.2 mL ± 1.0 on STIR images. The mean lesion volume at 1-month follow-up was 0.5 mL ± 0.4 on contrast-enhanced T1-weighted images (P < .01), 0.4 mL ± 0.5 on T2-weighted images (P < .01), and 0.4 mL ± 0.3 on STIR images (P < .01) (Fig 5).



View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5.  Chart illustrates mean (± SD) volumes of chronic thermal lesions measured immediately after RF ablation, at 2-week follow-up, and at 1-month follow-up on each of the contrast-enhanced T1-weighted images (CE-T1WI; black bars), T2-weighted images (T2WI; white bars), and STIR images (gray bars). Data show the continued shrinkage of thermal lesions during 1-month follow-up, and the volumes are significantly smaller with all pulse sequences at 1-month follow-up imaging than immediately after RF ablation. Note also the wide error bars in the 2-week and 1-month groups, indicating the variable rates of shrinkage among the individual thermal lesions.

 
Chronic thermal lesions, however, did not show the same rate of shrinkage. On STIR images, which showed the best thermal lesion size correlation with gross pathologic findings (as will be described later), thermal lesion volume was 2.31–3.72 mL immediately after ablation, while the residual lesions ranged between complete replacement with scar tissue (Fig 6, A–C) and 0.82 mL residual necrosis encased by scar tissue (Fig 7, A–C) at 1-month follow-up.



View larger version (95K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6.  MR appearance and histopathologic correlation of a chronic thermal lesion. Sagittal fast spin-echo STIR images (5,300/35; echo train length, seven; number of signals acquired, one) acquired with a high-field-strength (1.5-T) MR imager obtained A, immediately after RF ablation; B, at 2-week follow-up; and C, at 1-month follow-up. Note the rapid rate of thermal lesion shrinkage from 3.7 mL on the image obtained immediately after ablation (A) to 1 mL on the image obtained at 2-week follow-up (B) and ending as a thin band of enhanced scar tissue on the image obtained at 1-month follow-up (arrow in A-C). Corresponding gross pathologic specimen (D) and hematoxylin-eosin-stained (E) trichrome-stained and (F) histologic specimens demonstrate total replacement of the area of necrosis by a grayish dense fibrous tissue (arrowheads in D) that lacks inflammatory cells, denotes a healed scar (arrowheads in E), and stains blue with trichrome stain (arrowheads in F). Note the inward traction of the yet-intact surface mucosa by the contracting scar tissue (curved arrow in D-F). This effect illustrates the basic theory behind use of thermal energy to induce tongue base volume shrinkage as a treatment of OSA syndrome.

 


View larger version (89K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7.  MR appearance and histopathologic correlation of a chronic thermal lesion that underwent a slower, yet still significant, rate of shrinkage than the lesion in Figure 6. Sagittal fast spin-echo STIR images (5,300/35; echo train length, seven; number of signals acquired, one) acquired with a high-field-strength (1.5-T) MR imager obtained A, immediately after RF ablation; B, at 2-week follow-up; and C, at 1-month follow-up. The thermal lesion had a volume of 2.8 mL immediately after RF ablation, 1.7 mL at 2-week follow-up, and 0.8 mL at 1-month follow-up (arrow). Corresponding gross pathologic specimen (D) and low- (E) and (F) high-power trichrome-stained histologic specimens obtained immediately after 1-month follow-up imaging. Rectangle on E indicates the magnified area that is F. These images demonstrate tissue changes associated with the early healing process, as represented by the circumferential encasement of the area of coagulation necrosis by fibrous tissue giving rise to four distinct layers of histopathologic findings as follows: 1, normal muscle tissue of the base of the tongue; 2, mature fibrous tissue; 3, active granulation tissue; and 4, coagulated (mummified) muscle tissue.

 
Gross pathologic and microscopic evaluation of chronic lesions demonstrated the tissue changes associated with this observation. At one end of the spectrum, thermal lesions showed uniform healing of the treated area into a fibrous scar composed of organized granulation tissue. The scar was devoid of residual muscle fibers, except at the periphery, where muscle fibers intermingled with the fibrous scar tissue. The paucity of inflammatory cells within the fibrous area was consistent with that of a healed scar (Fig 6, D, E). The trichrome-stained sample showed an area of dense fibrosis within the scar (Fig 6, F). At the other end of the spectrum, the lesion was characterized by a grossly visible ring of actively organizing granulation tissue (Fig 7, D) with a central core of "mummified" muscle fibers. The granulation tissue was most mature at the periphery of the lesion and was progressively less mature and active toward the center. No evidence of organization was present in the central mummified area. Blood vessels within the lesion appeared necrotic and showed no active blood flow. The trichrome-stained tissue showed a ring-shaped area of fibrosis around the lesion (Fig 7, E, F). Again, no evidence of mucosal injury was noticed in the chronic lesions.

The maximum thermal lesion diameter perpendicular to the RF-electrode track (short axis), as measured at gross pathologic examination in the acute and chronic models, best agreed with the corresponding diameters measured on STIR images (interclass correlation coefficients = 0.85) followed by contrast-enhanced T1-weighted images (interclass correlation coefficients = 0.77) and T2-weighted images (interclass correlation coefficients = 0.63).

As in the acute group, the mean lesion-to-tongue CNR in the chronic group was consistently higher on contrast-enhanced T1-weighted images than on T2-weighted, STIR, or CISS images. On images obtained immediately after RF interstitial thermal ablation with a 1.5-T MR imager, the mean CNR was 424 ± 96 on contrast-enhanced T1-weighted images, 241 ± 29 on T2-weighted images, 244 ± 64 on STIR images, and 154 ± 46 on CISS images. On images obtained at 2-week follow-up, the mean CNR was 546 ± 173 on contrast-enhanced T1-weighted images, 214 ± 96 on T2-weighted images, 332 ± 190 on STIR images, and 149 ± 91 on CISS images. On images obtained at 1-month follow-up, the mean CNR was 618 ± 157 on contrast-enhanced T1-weighted images, 231 ± 100 on T2-weighted images, 234 ± 180 on STIR images, and 237 ± 113 on CISS images.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
REFERENCES
 
OSA is an episodic upper-airway obstruction that occurs most commonly at the pharyngeal level during sleep (2). This condition, which affects 1%–4% of all adults (1), may be considered a neuromuscular disorder in part. Normally, neural output to the pharyngeal muscles decreases with the onset of sleep, thus reducing their tone. At inspiration, the negative pressure created in the upper airway has the potential to collapse the hypotonic pharyngeal wall if the pharyngeal airway is originally smaller or more compliant than normal (6).

A number of treatment options exist, including conservative and surgical options (1,7–15), with patient response depending on the severity and level of obstruction and on whether a single or, more commonly, multiple levels are involved. Tracheostomy to bypass the entire upper airway is an effective last treatment option for patients with severe OSA.

Given the complexity of the treatment of OSA and the frequent need to combine various approaches to achieve satisfactory therapeutic results, a great deal of interest has developed around the potential use of minimally invasive procedures as either alternative or complementary measures to existing techniques.

Laser energy has been used with laser-assisted uvulopalatoplasty mainly to treat simple snoring rather than sleep apnea (7,16). Laser midline glossectomy has also been reported to be useful in some patients with OSA (17).

Much of the excitement about the use of minimally invasive procedures to treat OSA has been centered on RF thermal ablation of the tongue base to achieve scar tissue contraction and subsequent tongue base volume reduction during the healing phase of induced thermal lesions. Although the data are still sparse, a few studies in animals (4) and in humans (5,18–21) have documented the feasibility and efficacy of such treatment. For example, tongue volume reductions of 26.3% and 17% were achieved in animal models (4) and human subjects (5), respectively. Additional data in humans show reduction of the mean respiratory disturbance index by 55% and reduction of the apnea index by 77.3% after RF tongue base reduction treatment (5,22).

The current transoral approach for RF treatment of OSA may be suitable for soft palate ablation (23,24). Nevertheless, it is considered suboptimal to blindly create a thermal lesion within the tongue base by inserting the RF-electrode transmucosally through the limited space of the open mouth while attempting to minimize the risk of infection and spare the integrity of the neurovascular bundles of the tongue and surface mucosa. Reported complications associated with transoral tongue base RF ablation have included mucosal erosion, pain (19), temporary tongue base neuralgias, tongue base abscesses, and edema on the floor of the mouth with airway compromise (5,25). To our knowledge, change in functional parameters, such as taste or swallowing, has not been reported.

In an attempt to refine the technique of RF interstitial thermal ablation of the tongue base and to introduce a safer procedure that takes advantage of the recent developments in the growing field of interventional MR imaging (26–28), we report the use of interactive near-real-time MR imaging guidance to achieve safe and controlled thermal ablation of the tongue base with a percutaneous approach.

Compared with the transoral route, the percutaneous approach allows tongue base ablation to be accomplished without the need to puncture the tongue mucosa, thereby reducing the risk of mucosal injury and/or ulceration, postprocedure pain, and infection caused by the oral bacterial flora, a risk that is increased by the multiple ablation sessions often required (5,22) to produce therapeutic tongue base reduction. In addition, the percutaneous approach provides more room for the interventionist to direct the electrode tip toward the desired part of the tongue base without being restricted by the limited capacity of the mouth.

Percutaneous tongue base ablation with direct imaging guidance, particularly MR imaging guidance, ensures additional patient safety and procedure success by permitting accurate interactive electrode navigation and allowing immediate monitoring of the induced thermal lesion as it forms. Thus, injury of vital structures such as the neurovascular bundles and tongue mucosa can be avoided, and more efficient planning of subsequent ablations can be achieved with respect to the three-dimensional geometry of existing lesions.

Our experiments in a porcine model demonstrated the feasibility and simplicity of this technique. Advancement of the RF electrode between the lingual arteries may sound challenging; however, our experience shows that this technique requires no more skill than is required for other interventional MR imaging procedures (26–29), owing in particular to the high vascular conspicuity and the high temporal resolution offered by rapid gradient-echo sequences.

MR images of acute thermal lesions obtained during and immediately after the procedure, in combination with findings of gross and histopathologic examinations, showed that no vascular injury or hematoma complicated any of the procedures that were performed with the new technique. Observation of pigs in the chronic group further documented the safety of this procedure. No mucosal injury or ulceration, infection, nerve injury, or airway compromise was noticed in any of the animals in the chronic group. Although pain and taste sensation cannot be assessed as accurately in porcine models as in human subjects, our veterinary staff did not notice any sign that would indicate moderate or severe pain or suffering after the pigs recovered from anesthesia or on the subsequent days. The mild to moderate pain reported in the literature (18,19) after transoral tongue base RF interstitial thermal ablation was associated with puncture of the tongue mucosa and/or subsequent mucosal injury or ulceration. The percutaneous approach does not require mucosal puncture, and during the procedure the surface mucosa can always be identified as a darker band relative to the intrinsic muscles of the tongue. Thus, because the developing thermal lesion can be planned to avoid mucosal injury, we do not expect even mild pain—except at the site of skin puncture—to result when the procedure is performed in future clinical trials. Likewise, we do not expect taste sensation to be affected, as it is carried on the chorda tympani fibers distributed along the lingual nerves, which have been consistently spared in our experiments.

In addition to testing the feasibility and safety of the MR imaging–guided technique for tongue base ablation, a parallel investigation was being conducted to monitor the MR appearance of acute and chronic thermal lesions of the tongue and to correlate MR findings with associated gross and histopathologic findings. In this context, it should be noted that this study was not intended to reproduce the pioneering efforts of previous investigators (4,5,18–22) who have proved the efficacy of RF tongue base ablation for producing therapeutic volume shrinkage with more aggressive and multiple treatment sessions. Rather, we intended to evaluate the unit thermal lesion and its temporal evolution to provide insight into the tissue basis for this kind of therapy.

In our study, the MR, gross, and histopathologic appearances of acute and chronic thermal lesions of the tongue base corresponded to those described in earlier studies of RF-mediated thermal lesions in other organs (30–34). In the current study, however, the pattern of temporal evolution of thermal lesions showed consistent reduction in thermal lesion volume at 2-week follow-up, with continued shrinkage at 1-month follow-up. This pattern differs from that observed after kidney ablation (29), when thermal lesions appeared to have grown at 2-week follow-up before they eventually began to shrink.

Although all thermal lesions were significantly smaller at the end of the relatively short 1-month follow-up period than at immediate postablation imaging, the observed variability in their shrinkage rates remains unexplained.

In conclusion, we introduce a technique of percutaneous RF interstitial thermal ablation of the tongue base with direct MR fluoroscopic guidance as an alternative to the current practice of blind transoral RF ablation puncture used to treat OSA syndrome. Our investigation demonstrates that the procedure is both feasible and safe and illustrates the imaging and pathologic phenomena associated with the creation and evolution of thermal lesions in the tongue base.

Practical applications: The choice of the most appropriate intervention for a given patient in whom conservative treatment of OSA syndrome failed is still a largely subjective process without actual guidelines; however, the crucial initial step is to accurately define the level (or often, multiple levels) of airway obstruction. Classically, surgeons use the Fujita classification system (8) to classify pharyngeal obstruction as velopharyngeal, retroglossal, or combined. According to this classification, MR imaging–guided RF ablation of the tongue base may be used as the sole treatment for patients with retroglossal-only type obstruction, but this procedure will need to be combined with others—including RF ablation of the palate (23,24)—when associated with pathologic conditions at a higher level in the pharynx. Recently, Moore and Phillips (35) have further subdivided tongue-base narrowing into four basic structural patterns and have suggested that RF ablation be used in patients with a high tongue base rather than in those with retroepiglottic narrowing.

From a technical perspective, multiple ablation sessions are required to achieve therapeutic improvement of the physiologic parameters associated with airway narrowing. Although the use of aggressive therapy to reduce the number of treatment sessions without an increase in postoperative morbidity has been reported by a few investigators (18,22), this option should be practiced with care. The immediate postablation edema and possibly serious compromise of the already narrowed airway should always be weighed against spanning the treatment sessions over a longer period of time.


   ACKNOWLEDGMENTS
 
The authors acknowledge the members of the Interventional MRI Research Program at University Hospitals of Cleveland and Case Western Reserve University for their ongoing commitment to the development of interventional MR imaging techniques. The authors thank Nanette Kleinman, DVM, Tami McCourt, AAS, Jean Janesz, BA, AAS, and Markeya Owens, BS, AAS, for their assistance with animal observation and anesthesia; Bonnie Hami, MA, for her invaluable editorial assistance; and Elena Du Pont for help with manuscript preparation.


   FOOTNOTES
 
Abbreviations: CISS = construction interference in steady state, OSA = obstructive sleep apnea, RF = radiofrequency, STIR = short inversion time inversion recovery

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


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 

  1. Roman AF, Jr, Goldsmith A. Sleep disorders. In: Lucente FE, Har-El G, eds. Essentials of otolaryngology. 4th ed. Philadelphia, Pa: Lippincott Williams & Wilkins, 1999; 181-190.
  2. Maniglia AJ, Davis JA, Maniglia JV. Obstructive sleep apnea syndrome. In: Lee KJ, eds. Essential otolaryngology: head and neck surgery. 7th ed. Stamford, Conn: Appleton & Lange, 1999; 859-874.
  3. Jager L, Gunther E, Gauger J, Reiser M. Fluoroscopic MR of the pharynx in patients with obstructive sleep apnea. AJNR Am J Neuroradiol 1998; 19:1205-1214.[Abstract]
  4. Powell NB, Riley RW, Troell RJ, Blumen MB, Guilleminault C. Radiofrequency volumetric reduction of the tongue: a porcine pilot study for the treatment of obstructive sleep apnea syndrome. Chest 1997; 111:1348-1355.[Abstract/Free Full Text]
  5. Powell NB, Riley RW, Guilleminault C. Radiofrequency tongue base reduction in sleep-disordered breathing: a pilot study. Otolaryngol Head Neck Surg 1999; 120:656-664.[CrossRef][Medline]
  6. Galvin JR, Rooholamini SA, Stanford W. Obstructive sleep apnea: diagnosis with ultrafast CT. Radiology 1989; 171:775- 778.[Abstract/Free Full Text]
  7. Johnson JT. Obstructive sleep apnea. In: Gates GA, eds. Current therapy in otolaryngology: Head and neck surgery. 6th ed. St Louis, Mo: Mosby, 1998; 422-424.
  8. Fujita S, Conway W, Zorick F, Roth T. Surgical correction of anatomic abnormalities in obstructive sleep apnea syndrome: uvulopalatopharyngoplasty. Otolaryngol Head Neck Surg 1981; 89:923-934.[Medline]
  9. Hochban W, Brandenburg U, Peter JH. Surgical treatment of obstructive sleep apnea by maxillomandibular advancement. Sleep 1994; 17:624-629.[Medline]
  10. Riley RW, Powell NB, Guilleminault C. Obstructive sleep apnea syndrome: a surgical protocol for dynamic upper airway reconstruction. J Oral Maxillofac Surg 1993; 51:742-747.[Medline]
  11. Afzelius LE, Elmqvist D, Laurin S, Risberg AM, Aberg M. Sleep apnea syndrome caused by acromegalia and the treatment with a reduction plasty of the tongue: case report. ORL J Otorhinolaryngol Relat Spec 1982; 44:142-145.[Medline]
  12. Djupesland G, Lyberg T, Krogstad O. Cephalometric analysis and surgical treatment of patients with obstructive sleep apnea syndrome: a preliminary report. Acta Otolaryngol 1987; 103:551- 557.[Medline]
  13. Woodson BT, Fujita S. Clinical experience with lingualplasty as part of the treatment of severe obstructive sleep apnea. Otolaryngol Head Neck Surg 1992; 107:40-48.[Medline]
  14. Faye-Lund H, Djupesland G, Lyberg T. Glossopexia: evaluation of a new surgical method for treating obstructive sleep apnea syndrome. Acta Otolaryngol Suppl 1992; 492:46-49.[Medline]
  15. Chabolle F, Wagner I, Blumen MB, Sequert C, Fleury B, De Dieuleveult T. Tongue base reduction with hyoepiglottoplasty: a treatment for severe obstructive sleep apnea. Laryngoscope 1999; 109:1273-1280.[CrossRef][Medline]
  16. Littner M, Kushida CA, Hartse K, et al. Practice parameters for the use of laser-assisted uvulopalatoplasty: an update for 2000. Sleep 2001; 24:603-619.[Medline]
  17. Fujita S, Woodson BT, Clark JL, Wittig R. Laser midline glossectomy as a treatment for obstructive sleep apnea. Laryngoscope 1991; 101:805-809.[Medline]
  18. Stuck BA, Maurer JT, Hormann K. Tongue base reduction with radiofrequency energy in sleep apnea. HNO 2001; 49:530-537. [German].[CrossRef][Medline]
  19. Woodson BT, Nelson L, Mickelson S, Huntley T, Sher A. A multi-institutional study of radiofrequency volumetric tissue reduction for OSAS. Otolaryngol Head Neck Surg 2001; 125:303-311.[CrossRef][Medline]
  20. Nelson LM. Combined temperature-controlled radiofrequency tongue reduction and UPPP in apnea surgery. Ear Nose Throat J 2001; 80:640-644.[Medline]
  21. Guilleminault C, Chervin R, Palombini L, Powell N. Radiofrequency (pacing and thermic effects) in the treatment of sleep-disordered breathing. Sleep 2000; 23(suppl 4):S182-S186.
  22. Stuck BA, Maurer JT, Hormann K. Tongue base reduction with radiofrequency tissue ablation: preliminary results after two treatment sessions. Sleep Breath 2000; 4:155-162.[CrossRef][Medline]
  23. Brown DJ, Kerr P, Kryger M. Radiofrequency tissue reduction of the palate in patients with moderate sleep-disordered breathing. J Otolaryngol 2001; 30:193-198.[Medline]
  24. Boudewyns A, Van De Heyning P. Temperature-controlled radiofrequency tissue volume reduction of the soft palate (somnoplasty) in the treatment of habitual snoring: results of a European multicenter trial. Acta Otolaryngol 2000; 120:981-985.[CrossRef][Medline]
  25. Pazos G, Mair EA. Complications of radiofrequency ablation in the treatment of sleep-disordered breathing. Otolaryngol Head Neck Surg 2001; 125:462-467.[Medline]
  26. Lewin JS, Petersilge CA, Hatem SF, et al. Interactive MR imaging-guided biopsy and aspiration with a modified clinical C-arm system. AJR Am J Roentgenol 1998; 170:1593-1601.[Abstract/Free Full Text]
  27. Lewin JS. Interventional MR imaging: concepts, systems, and applications in neuroradiology. AJNR Am J Neuroradiol 1999; 20:735-748.[Free Full Text]
  28. Lewin JS, Nour SG, Duerk JL. Magnetic resonance image-guided biopsy and aspiration. Top Magn Reson Imaging 2000; 11:173-183.[CrossRef][Medline]
  29. Lewin JS, Nour SG, Connell CF, Sulman A, Duerk JL, Resnick MI. Followup findings of a phase II clinical trial of interactive MR-guided interstitial radiofrequency thermal ablation of primary kidney tumors (abstr) Proceedings of the Tenth Meeting of the International Society for Magnetic Resonance in Medicine. Berkeley, Calif: International Society for Magnetic Resonance in Medicine, 2002; 245.
  30. 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]
  31. Merkle EM, Haaga JR, Duerk JL, Jacobs GH, Brambs HJ, Lewin JS. MR imaging–guided radio-frequency thermal ablation in the pancreas in a porcine model with a modified clinical C-arm system. Radiology 1999; 213:461-467.[Abstract/Free Full Text]
  32. De Salles AA, Brekhus SD, De Souza EC, et al. Early postoperative appearance of radiofrequency lesions on magnetic resonance imaging. Neurosurgery 1995; 36:932-937.[Medline]
  33. McGahan JP, Brock JM, Tesluk H, Gu WZ, Schneider P, Browning PD. Hepatic ablation with use of radio-frequency electrocautery in the animal model. J Vasc Interv Radiol 1992; 3:291-297.[Medline]
  34. Nour SG, Aschoff AJ, Mitchell IC, Emancipator SN, Duerk JL, Lewin JS. MR imaging–guided radio-frequency thermal ablation of the lumbar vertebrae in porcine models. Radiology 2002; 224:452-462.[Abstract/Free Full Text]
  35. Moore KE, Phillips C. A practical method for describing patterns of tongue-base narrowing (modification of Fujita) in awake adult patients with obstructive sleep apnea. J Oral Maxillofac Surg 2002; 60:252-261.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
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 Google Scholar
Google Scholar
Right arrow Articles by Nour, S. G.
Right arrow Articles by Strauss, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nour, S. G.
Right arrow Articles by Strauss, M.

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