| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Technical Developments |
1 From the Department of Medical Imaging, Division of Neuroradiology, Toronto Western Hospital, Fell Pavilion 3-404, 399 Bathurst St, Toronto, Ontario, Canada M5T 2S8 (R.I.F., J.K.K., R.A.W., W.J.M., K.t.B., J.M.C.v.D.); Laboratory of Cardiac Energetics, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md (J.A.D.); Department of Neurosurgery, Leiden University Medical Center, Leiden, the Netherlands (J.M.C.v.D.); and Department of Imaging Research, Sunnybrook and Women’s College Health Science Centre, University of Toronto, Toronto, Ontario, Canada (G.A.W.). Received April 24, 2001; revision requested May 25; revision received August 17; accepted September 7. Supported by grant A3048 from the Heart and Stroke Foundation of Ontario and a 2001 grant from the Canadian Heads of Academic Radiology/Nycomed Research Development Program. Address correspondence to R.I.F. (e-mail: richard.farb@utoronto.ca).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
© RSNA, 2002
Index terms: Arteriovenous malformations, dural, 379.149 • Dura, 379.149 • Dura, MR, 379.121411, 379.12142, 379.12143 • Fistula, arteriovenous, 379.149 • Magnetic resonance (MR), vascular studies, 379.121411, 379.12142, 379.12143 • Spinal cord, abnormalities, 379.149
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
The entity of spinal dural AVF has historically represented a diagnostic challenge in neuroimaging. Although a spinal dural AVF may be suspected after conventional MR imaging, its precise localization has not been consistently demonstrated with MR angiographic techniques, to our knowledge. In most centers, once a spinal dural AVF is suspected clinically and supported by findings at conventional MR imaging, conventional spinal angiography is then required to confirm the diagnosis and localize the site of the fistula.
The purpose of this study was to evaluate real-time monitored autotriggered elliptic centric ordered three-dimensional gadolinium-enhanced MR angiography (hereafter, this MR angiographic technique) for the noninvasive diagnostic confirmation, visualization, and pretreatment localization of spinal dural AVF.
|
Materials and Methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
MR Imaging Examination
In keeping with institutional review board requirements, each MR imaging examination was clinically indicated and requested by the patient’s referring physician as part of clinical care. Informed consent was obtained from all patients. All patients underwent imaging in the supine position with a superconducting 1.5-T system (Signa cv/i, version 8.2.5–8.4 software; GE Medical Systems, Milwaukee, Wis) with a posterior standard phased-array spinal coil. Conventional MR imaging was performed with sagittal T1- and T2-weighted sequences. A T1-weighted spin-echo sequence was performed with the following parameters: repetition time msec/echo time msec of 450/8.0, fractional echo acquisition, field of view of 36 x 36 cm, matrix of 512 x 192, three signals acquired, bandwidth of 32.0 kHz, section thickness of 3 mm, section spacing of 0.5 mm, approximately 14 sections acquired. This sequence was also repeated after the MR angiographic sequence was completed. A T2-weighted fast spin-echo sequence was also performed (before the angiographic sequence) with the following parameters: 3,500/81.4, echo train length of 14, field of view of 36 x 27 cm, matrix of 512 x 192, two signals acquired, bandwidth of 32.0 kHz, section thickness of 3 mm, section spacing of 0.5 mm, approximately 14 sections acquired. A T2-weighted fast spin-echo transverse scout sequence was performed. The three-dimensional centrically encoded MR angiographic sequence was oriented in the sagittal plane over the spinal canal. In the first seven patients, MR angiograms in the thoracolumbar region were evaluated initially and included images from approximately T8 to S1. If the site of the fistula was not identified during this first examination, as occurred in five of seven patients, the patient returned to have the thoracic or cervical region evaluated on another occasion. For patients 8 and 9, the location of the field of view in the initial MR examination was shifted rostrally to include images from T1 to approximately L3.
This MR angiographic technique consisted of four integral parts: (a) initiation of a special two-dimensional single-section bolus-detection sequence oriented in the transverse plane and located at approximately the level of T10; (b) intravenous power injection of contrast material; (c) automated detection of the arrival of intraarterial (aortic) gadolinium-based contrast material at the level of T10, which resulted in automatic termination of the detection sequence and triggering of (d) MR imaging with a fast three-dimensional gradient-echo MR angiographic sequence with elliptic centric ordered phase encoding. This fully automated detection and triggering system has been previously described in detail for intracranial vascular imaging (6). Modifications to the triggering tool parameters included insertion of a 4-second delay into the triggering paradigm to allow adequate perfusion through and filling of the anatomy of the fistula. Further minor modifications are described later.
Detection and Triggering Sequence
The two-dimensional bolus detection sequence was initiated and controlled in real-time by using an interactive user-friendly computer interface on a separate workstation (Sun Microsystems, Mountain View, Calif). This workstation was connected directly to the MR imager subsystems by using a special bus adapter (SBS Technologies [formerly Bit3], St Paul, Minn). The detection sequence consisted of a continuously refreshed two-dimensional gradient-echo acquisition of a transverse section positioned at the level of T10. The detection sequence parameters were 11.5/9.5, flip angle of 30°, field of view of 20 x 20 cm, matrix of 320 x 128, bandwidth of 62.5 kHz, and section thickness of 10 mm. The resultant image update rate was approximately one every 1.5 seconds.
As a first step in the implementation of the detection sequence, interactive section positioning was performed to obtain a transverse image at the level of T10 to allow placement of a region of interest over the aorta. In all cases, the region of interest was placed interactively by the MR technologist and was drawn as a circular tracing approximately 2–3 cm in diameter to encircle the entire aorta. Inferior and superior saturation bands were then applied sequentially. The saturation bands were left active during the detection sequence, which extinguished signal due to flow-related enhancement within the aorta, and thus optimized visualization of the expectant inflow of gadolinium-enhanced arterial blood. A signal intensity triggering threshold was then automatically determined.
Data from the region of interest were collected and evaluated in an automated fashion so the mean signal intensity in the brightest 20% of pixels in the region of interest was averaged for 10 consecutive images. A triggering threshold was then set at 5 SDs higher than the mean of the data. After the technologist was notified that the triggering threshold had been determined, the technologist then potentiated, or armed, the system. Subsequent intravenous injection of a bolus of gadolinium-based contrast material resulted in a rapid increase in signal intensity within the region of interest that exceeded the set threshold. After an additional preprogrammed delay of 4 seconds, the detection sequence was automatically terminated and the sequence for the three-dimensional centric MR angiographic sequence was simultaneously initiated. The 4-second delay was determined by observing the filling rate for the anatomy of the spinal dural AVF at conventional angiography. Spinal dural AVFs are generally slow-flow small fistulae that require time to fill from the feeding artery to the perimedullary plexus (2).
With imaging and processing delays, triggering occurred from 5 to 6 seconds after the arrival of the leading edge of the bolus of gadolinium-based contrast agent. This triggering tool has been presented in detail previously (6,7).
Contrast Material Injection
As part of preparation for MR imaging, a commercially available two-cylinder MR-compatible injector (Spectris; Medrad, Indianola, Pa) was loaded with 30 mL of gadodiamide (Omniscan; Nycomed Amersham, Buckinghamshire, England) in one syringe (syringe 1) and with 60 mL of normal saline in a second syringe (syringe 2). Intravenous access was obtained by using a 20-gauge intravenous catheter located in the right antecubital vein. A very slow infusion (0.5 mL/min) of saline was initiated at the time of intravenous connection and continued during preliminary imaging until injection of the contrast material bolus. The technologist in the control room manually initiated the injection. Thirty milliliters of the gadolinium-based contrast material was injected at a rate of 3 mL/sec. This was immediately followed with a 30-mL bolus of normal saline, which was also injected at a rate of 3 mL/sec. This injection protocol was followed for all patients.
Real-time Monitored Autotriggered Elliptic Centric Ordered Three-dimensional Gadolinium-enhanced MR Angiography
The view order for this MR angiographic sequence was in ascending order of radial k-space distance from the k-space origin, similar to that previously reported (6,8,9). The sequence parameters included 6.2/1.5; flip angle of 30°; fractional echo acquisition; field of view of 36 x 27 cm (three-fourths of the field of view in the frequency-encoding dimension); matrix of 352 x 352; bandwidth of 62.5 kHz; section thickness of 1.2 mm; 70 sections acquired, which resulted in a 8.4-cm-thick sagittal slab; and a total imaging time of 118 seconds. Resultant voxel dimensions were nearly isotropic at 1.0 x 1.0 x 1.2 mm. This spatial resolution was obtained without zero-filling techniques.
Image Processing and Review
Source images obtained with this MR angiographic technique were transferred to a commercially available three-dimensional workstation (Advantage for Windows, version 3.7; GE Medical Systems) for image manipulation and display. The systematic method for visualization of the fistula in each patient was a combination of review of source images and multiplanar volume reformatted images. Specifically, a 10-mm-thick transversely oriented multiplanar volume reformatted image was produced on a template of a coronal reformatted image of the spine. This transverse slab was then sequentially stepped superiorly on the template of the coronal image. In this fashion the bilateral lumbar or intercostal arteries were systematically visualized, which helped confirm their normal anatomic relations or identify the spinal dural AVF. With use of the multiplanar volume reformatting option, the imaged portions of the spine were systematically evaluated until the area of suspected fistula was well understood in three planes. The workstation image processing and evaluation lasted approximately 10–15 minutes.
Postprocessing and review of all MR images was performed by one neuroradiologist (R.I.F.). Conventional MR images were evaluated for hyperintensity within the spinal cord on T2-weighted images, mild diffuse enhancement of the lower thoracic spinal cord on gadolinium-enhanced T1-weighted images, intradural tortuous vascular flow voids posterior to the spinal cord on T2-weighted images, an abrupt change in the size and number of intradural vascular flow voids at the level of the fistula, an enlarged intercostal or lumbar feeding artery that supplied the fistula, and conspicuous enhancement within the involved neural foramen at the level of the fistula on the gadolinium-enhanced T1-weighted images. All MR angiographic examinations were performed without technical difficulty or complication and provided diagnostic images.
DSA Examination
All patients underwent conventional spinal angiography within 3 weeks after the MR angiographic examination. All DSA examinations were performed with a previously described standard protocol with use of general anesthetic and multiple selective arterial injections to help identify the anterior spinal artery and the site of the spinal dural AVF (10). All angiographers were informed of the MR angiographic results and were encouraged to select the specified arteries early in the procedure. Eight of the DSA examinations were performed with a dedicated biplane neuroangiographic system (LCN+, GE Medical Systems, Buc, France [n = 7]; BV-5000, Philips Medical Systems, Best, the Netherlands [n = 1]). One DSA examination was performed with a uniplane system (Integris V-3000; Philips Medical Systems).
Patient 6 experienced transient worsening of his paraparesis after DSA that improved after surgical ligation of the spinal dural AVF. Of the nine spinal dural AVFs, one was treated with endovascular embolization, and the remaining eight were treated with surgical ligation of the draining medullary vein.
|
Results |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
In the majority of cases, one macroscopic feeding artery supplies a spinal dural AVF. Occasionally, additional macroscopic feeding arteries that arise from adjacent vertebral segments are seen. These are in contrast to the microscopic additional feeding arteries that have been shown to be commonly present in spinal dural AVFs (15). The feeding arteries converge on the fistula, which is usually located within the dura of a proximal root sleeve. From this one fistula, one medullary vein (radiculomedullary vein) drains intradurally. The fistula results in retrograde flow within the medullary vein and filling of the perimedullary coronal venous plexus of the spinal cord. There is no nidus associated with these lesions (15). Complete characterization of the AVF with MR angiography or DSA is commonly not possible since the fistula itself and many of the additional feeding arteries are microscopic (15).
The failure of earlier time-of-flight and phase-contrast MR angiographic techniques to localize the sites of spinal dural AVFs relates to the small caliber of the vessels, decreased flow velocity, and tortuosity of the vessels. An early technique in gadolinium-enhanced spinal MR angiography consisted of intravenous injection of gadolinium-based contrast material and subsequent performance of a conventional time-of-flight MR angiographic technique in a nontimed nontriggered fashion (16). That technique produced images that clearly showed multiple abnormal intradural veins. However, without the benefit of temporal resolution, the conspicuity of these veins and other normally enhancing structures obscured the underlying spinal dural AVF, which precluded its visualization. Mascalchi et al (17) recently described gadolinium-enhanced and phase-contrast MR angiographic techniques that confirmed abnormal intradural veins in 28 patients with spinal dural AVF; however, in none of these patients was the location of the fistula identified before DSA. By using susceptibility-induced signal changes associated with bolus injection of gadolinium-based contrast agent, Thorpe et al (18) were able to estimate the probable location of the spinal dural AVF to within several levels in four of five patients.
Recently, authors have described techniques for temporally resolved gadolinium-enhanced MR angiography with coordination of injection and image acquisition that demonstrated the level of the spinal dural AVF in small numbers of patients. Binkert et al (19) described the correct prospective localization of spinal dural AVFs in two of three patients examined with a rapidly repeated three-dimensional MR angiographic sequence performed immediately after injection of gadolinium-based contrast agent. With use of a similar technique with the addition of a test bolus injection of gadolinium-based contrast material to estimate optimal injection-imaging coordination, Shigematsu et al (20) prospectively identified the probable feeding arteries of spinal dural AVFs in three patients. In that study, fistula location was determined by combining data from the spinal MR angiographic and conventional angiographic sequences to help estimate the probable site of the fistula. In their review of spinal MR angiographic techniques, Bowen and Pattany (12) highlighted the fact that, although the level of the fistula may often be determined from the course of the draining vein, only the display of the fistulous communication would provide added confidence in diagnosis and localization. The challenge for contrast material–enhanced spinal MR angiography is to clearly demonstrate the fistula free from the obscuring overlying venous engorgement.
The MR angiographic technique used in the current study benefitted from elliptic centric ordered imaging, rapid bolus injection of contrast material, a triggering mechanism to optimize capture of arterial contrast, and a high-spatial-resolution acquisition to provide direct visualization of the fistula.
T1-weighted short repetition time imaging with elliptic centric ordering of k space yields a final image on which the signal intensities of the structures are determined on the basis of the state of gadolinium enhancement that exists within the first several seconds of imaging (6,8). Optimally timed performance of this type of imaging provides maximum subject-to-background contrast, where subject refers to the fistula with its feeding artery and early draining vein, and background refers to the surrounding tissues, spinal cord, and vascular structures, including an abnormally engorged perimedullary venous plexus that enhances with gadolinium-based contrast material later (in the venous phase). This optimization is made possible by using the automated triggering tool. This tool initiates the elliptic centric encoded angiographic sequence following a preprogrammed 4-second delay after the arrival of the gadolinium-based contrast material within the aorta at the level of T10.
The majority of spinal dural AVFs are found from the midthoracic to the upper lumbar regions of the spinal canal; less commonly, however, a spinal dural AVF can reside within the upper thoracic, lower lumbar, sacral, and intracranial regions (1–3,5,21). Spinal dural AVFs that occur in the cervical region have been reported (22). Therefore, complete evaluation of the spinal canal from cranium to sacrum may be required to find the spinal dural AVF in these patients. With use of a systematic approach, we opted to initially evaluate the lower spine from T8 to S1 in the first seven patients. We hoped to include within the initial field of view (one imaging session) approximately 70% of spinal dural AVFs that would be predicted on the basis of the known reported frequency and location of these lesions (5). In retrospect, in our patient population, a field of view from the midthoracic region to the upper lumbar region would have led to less need for repeat evaluation. This shifted field of view benefitted the last two patients. In this preliminary evaluation, we strove for a maximal degree of background suppression; therefore, we waited at least 24 hours before repeating gadolinium-enhanced MR angiography in another region. A protocol may be contemplated for performing imaging in the upper and lower cervical and thoracic regions and the lumbar and sacral regions at one sitting by using multiple injections of gadolinium-based contrast material.
The search for a spinal dural AVF at conventional angiography is often tedious and requires selective injections into multiple bilateral thoracic intercostal and lumbar arteries. If no fistula is found, then sacral, vertebral, cervical, and intracranial regions are sequentially explored. An exhaustive search for a spinal dural AVF may include as many as 40 selective injections, which necessitates three separate MR angiographic sessions owing to high doses of iodinated contrast material. This MR angiographic technique is not intended to obviate conventional angiography, but the technique can localize the fistula and thus greatly reduce the amount of time and iodinated contrast material required for conventional angiography. State-of-the-art selective spinal conventional angiography is still mandatory before embolization to allow extensive characterization of the spinal dural AVF and identification of macroscopic additional feeding arteries and to determine whether an anterior spinal artery arises from the same pedicle that supplies the dural fistula (23). This information, although not required for surgical management, is required to guide endovascular therapy. Consistent identification of a normal anterior spinal artery at MR angiography has not yet been described, to our knowledge, and is beyond the spatial resolution of current MR systems, including the one used in the current study.
Spatial resolution limitations of this MR angiographic technique do not permit the in-depth characterization of a fistula that is possible with spinal DSA. More complete characterization of a fistula with MR angiography may be possible in the future with refinement of this MR angiographic technique, as well as the introduction of 3.0-T MR imaging.
At our institution, this MR angiographic technique has substantially changed the management of disease in patients suspected of having a spinal dural AVF. We now combine the confirmatory conventional angiographic examination and possible embolization into one procedure performed with one application of general anesthesia. At conventional angiography, the site of the spinal dural AVF targeted at MR angiography is interrogated first, and a decision is made immediately regarding the appropriateness of embolization.
In conclusion, this MR angiographic technique is capable of localizing a spinal dural AVF with its small feeding artery and draining veins. The centric encoding of k space and the use of a triggering tool ensure a high optimal concentration of gadolinium-based contrast material in the fistulous arteries and veins. Findings in this initial assessment have shown that this MR angiographic technique allows noninvasive and reliable confirmation of the diagnosis of spinal dural AVF and precise identification of its location in eight of nine patients.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, R.I.F.; study concepts and design, R.I.F.; literature research, R.I.F., R.A.W.; clinical studies, R.I.F., W.J.M., R.A.W., K.t.B.; data acquisition, R.I.F., J.K.K., J.A.D., G.A.W.; data analysis/interpretation, R.I.F.; manuscript preparation, R.I.F., R.A.W., J.M.C.v.D.; manuscript definition of intellectual content, R.I.F., J.K.K., G.A.W.; manuscript editing, R.I.F., R.A.W., K.t.B., G.A.W.; manuscript revision/review, R.I.F., J.K.K., J.A.D., R.A.W., W.J.M., K.t.B., G.A.W.; manuscript final version approval, R.I.F., R.A.W.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Krings and S. Geibprasert Spinal Dural Arteriovenous Fistulas AJNR Am. J. Neuroradiol., April 1, 2009; 30(4): 639 - 648. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W.H. Backes and R.J. Nijenhuis Advances in Spinal Cord MR Angiography AJNR Am. J. Neuroradiol., April 1, 2008; 29(4): 619 - 631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ali, T.A. Cashen, T.J. Carroll, E. McComb, M. Muzaffar, A. Shaibani, and M.T. Walker Time-Resolved Spinal MR Angiography: Initial Clinical Experience in the Evaluation of Spinal Arteriovenous Shunts AJNR Am. J. Neuroradiol., October 1, 2007; 28(9): 1806 - 1810. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mull, R.J. Nijenhuis, W.H. Backes, T. Krings, J.T. Wilmink, and A. Thron Value and Limitations of Contrast-Enhanced MR Angiography in Spinal Arteriovenous Malformations and Dural Arteriovenous Fistulas AJNR Am. J. Neuroradiol., August 1, 2007; 28(7): 1249 - 1258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Jellema, C. C. Tijssen, and J. v. Gijn Spinal dural arteriovenous fistulas: a congestive myelopathy that initially mimics a peripheral nerve disorder Brain, December 1, 2006; 129(12): 3150 - 3164. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.J. Nijenhuis, M. Mull, J.T. Wilmink, A.K. Thron, and W.H. Backes MR Angiography of the Great Anterior Radiculomedullary Artery (Adamkiewicz Artery) Validated by Digital Subtraction Angiography AJNR Am. J. Neuroradiol., August 1, 2006; 27(7): 1565 - 1572. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.J. Nijenhuis, M.J. Jacobs, J.M.A. van Engelshoven, and W.H. Backes MR Angiography of the Adamkiewicz Artery and Anterior Radiculomedullary Vein: Postmortem Validation AJNR Am. J. Neuroradiol., August 1, 2006; 27(7): 1573 - 1575. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P.H. Lai, M.J. Weng, K.W. Lee, and H.B. Pan Multidetector CT angiography in diagnosing type I and type IVA spinal vascular malformations. AJNR Am. J. Neuroradiol., April 1, 2006; 27(4): 813 - 817. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Horie, M. Morikawa, N. Kitigawa, K. Tsutsumi, M. Kaminogo, and I. Nagata 2D Thick-Section MR Digital Subtraction Angiography for the Assessment of Dural Arteriovenous Fistulas. AJNR Am. J. Neuroradiol., February 1, 2006; 27(2): 264 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. P. Sheehy, G. E. Boyle, and J. F. M. Meaney Normal Anterior Spinal Arteries within the Cervical Region: High-Spatial-Resolution Contrast-enhanced Three-dimensional MR Angiography Radiology, August 1, 2005; 236(2): 637 - 641. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P.-H. Lai, H.-B. Pan, C.-F. Yang, L.-R. Yeh, S.-S. Hsu, K.-W. Lee, M.-J. Weng, M.-T. Wu, H.-L. Liang, and C.-K. Chen Multi-Detector Row Computed Tomography Angiography in Diagnosing Spinal Dural Arteriovenous Fistula: Initial Experience Stroke, July 1, 2005; 36(7): 1562 - 1564. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. H. Luetmer, J. I. Lane, J. R. Gilbertson, M. A. Bernstein, J. Huston III, and J. L. D. Atkinson Preangiographic Evaluation of Spinal Dural Arteriovenous Fistulas with Elliptic Centric Contrast-Enhanced MR Angiography and Effect on Radiation Dose and Volume of Iodinated Contrast Material AJNR Am. J. Neuroradiol., April 1, 2005; 26(4): 711 - 718. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. A. Chuang, M. M. Shroff, R. A. Willinsky, J. M. Drake, P. B. Dirks, and D. C. Armstrong Slow-Flow Spinal Epidural AVF with Venous Ectasias: Two Pediatric Case Reports AJNR Am. J. Neuroradiol., October 1, 2003; 24(9): 1901 - 1905. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. I. Farb, J. N. Scott, R. A. Willinsky, W. J. Montanera, G. A. Wright, and K. G. terBrugge Intracranial Venous System: Gadolinium-enhanced Three-dimensional MR Venography with Auto-triggered Elliptic Centric-ordered Sequence--Initial Experience Radiology, January 1, 2003; 226(1): 203 - 209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. C. van Dijk, K. G. TerBrugge, R. A. Willinsky, R. I. Farb, and M. C. Wallace Multidisciplinary Management of Spinal Dural Arteriovenous Fistulas: Clinical Presentation and Long-Term Follow-Up in 49 Patients Stroke, June 1, 2002; 33(6): 1578 - 1583. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
