| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Toyama Medical and Pharmaceutical University, 2630
Sugitani, Toyama, 930-0194, Japan.
2 Department of Radiology, Neuroradiology Section, University of Pennsylvania
Medical Center, Philadelphia, PA 19104.
3 Department of Neurosurgery, Toyama Medical and Pharmaceutical University,
Toyama, 930-0194, Japan.
Received December 2, 2002;
accepted after revision July 31, 2003.
Address correspondence to K. Noguchi
(kyo{at}ms.toyama-mpu.ac.jp).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Two neuroradiologists, unaware of patients' histories and angiographic findings, retrospectively reviewed 17 MR angiograms with 3D TOF MR angiography and MR digital subtraction angiography in 15 patients with dural arteriovenous fistula and also reviewed 35 MR angiograms in control patients without findings of dural arteriovenous fistula on angiography. Disagreements were resolved by consensus.
RESULTS. In patients with dural arteriovenous fistula, source images of 3D TOF MR angiography showed two abnormal findings: multiple high-intensity curvilinear or nodular structures adjacent to the sinus wall and high-intensity areas in the venous sinus. Findings of multiple high-intensity structures adjacent to the sinus wall were observed in all cases of dural arteriovenous fistula. Findings of high-intensity areas in the venous sinus were observed in 13 of 17 cases of dural arteriovenous fistula. Findings of multiple high-intensity structures adjacent to the sinus wall were not observed in any control subjects. Findings of high-intensity areas within the venous sinus were observed in five of 35 control subjects. Findings of MR digital subtraction angiography showed early filling of the venous sinus, suggestive of dural arteriovenous fistula, in 13 of 15 patients with dural arteriovenous fistula. Sensitivity and specificity of multiple high-intensity structures adjacent to the sinus wall, high-intensity areas in the venous sinus, and early filling of the venous sinus were 100% and 100%, 76% and 86%, and 87% and 100%, respectively. Although 3D TOF MR angiography failed to show the findings of retrograde cortical venous drainage and venous sinus occlusion, MR digital subtraction angiography clearly showed both findings in all five subjects.
CONCLUSION. A protocol including both 3D TOF MR angiography (source images) and MR digital subtraction angiography allowed the diagnosis of moderate- to high-flow dural arteriovenous fistula. In addition, cortical venous drainage was reliably noted in a small subset of patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Intraarterial catheter angiography remains the gold standard and should be performed in most cases when the diagnosis of dural arteriovenous fistula is being considered. However, it is an invasive procedure that involves ionizing radiation and a risk of permanent neurologic damage [3]. Therefore, reliable clues to the diagnosis on a noninvasive examination can play an important role in the treatment of patients with moderate- to high-flow dural arteriovenous fistula who are at potentially high risk for hemorrhage or infarction.
Conventional MRI is less successful than digital subtraction angiography in showing the exact fistula site [4, 5]. Three-dimensional time-of-flight (3D TOF) MR angiography has been useful in showing an arteriovenous shunt of a dural arteriovenous fistula in a variety of intracranial locations [5]. However, 3D TOF MR angiography is relatively insensitive to dilatation of cortical veins or sinus occlusion.
Recently, contrast-enhanced MR digital subtraction angiography has been used for imaging the head and neck and intracranial regions [3, 6–8]. Like intraarterial digital subtraction angiography, MR digital subtraction angiography allows observation of the passage of the bolus in intracranial vessels with a temporal resolution significantly below a few seconds. The purpose of this study was to examine the results of combining 3D TOF MR angiography and MR digital subtraction angiography with a temporal resolution of less than 5 sec in the diagnosis of dural arteriovenous fistula with moderate- to high-flow arteriovenous shunt.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MRI
MRI was performed on a 1.5-T superconducting unit with a standard head
coil. Routine T1-weighted spin-echo and T2-weighted turbo spin-echo imaging
was performed using the following parameters: for T1-weighted imaging, TR/TE,
550/15; and for T2-weighted imaging, 5,000/90. All images were obtained with a
field of view of 22 cm and a matrix of 192 x 256. Section thickness was
7 mm.
Three-dimensional TOF MR angiography was performed with 3D fast imaging with steady-state precession sequence with the following parameters: 40/6.5; flip angle of 20°, partition of 64 or 80; field of view of 22 cm; and matrix of 192 x 512. The actual thickness of the partitions was 0.8 or 1.0 mm. The saturation pulse was placed above the volume slab in control subjects and all patients with dural arteriovenous fistula.
MR digital subtraction angiography was performed with the 3D fast low-angle shot (FLASH) sequence using zero-fill interpolation (applied in both phase and slice directions asymmetrically) and the following parameters: 4.6/1.8; flip angle of 25°; 8 partitions; slab thickness of 8 cm; field of view of 22 cm; and matrix of 130 x 256 in the sagittal plane. We localized the slab for MR digital subtraction angiography on the basis of 3D TOF MR angiography (source images), choosing a unilateral hemisphere including the superior sagittal sinus and the transverse and sigmoid sinuses. MR digital subtraction angiography was performed every 4 sec after the initiation of a bolus injection of 0.1 mmol/kg of gadolinium-based contrast agent at 1.5 mL/sec for a duration of up to 40 sec by power injector. The last image set (mask images set) before contrast arrival was selected visually and subtracted from the subsequent image sets. Subtraction image sets were generated with commercially available bundled software. Each subtracted image set was presented using the maximum-intensity-projection algorithm in the lateral view.
Transit-Time Difference Between Artery and Sinus
We also performed a preliminary study to evaluate the transit-time
difference between the intracranial artery (internal carotid artery) and the
venous sinus (torcular herophili) in another 10 control subjects (four men and
six women; age range, 21–74 years; mean age, 55.4 years). No dural
arteriovenous fistula was found on digital subtraction angiography in these
control patients. Two-dimensional turbo FLASH sequence was performed with the
following parameters: 5.8/2.4; flip angle of 40°; field of view of 22 cm;
matrix of 188 x 256 in the axial plane. This sequence with a single
slice thickness of 10 mm was performed every second after the initiation of a
bolus injection of 0.1 mmol/kg of gadolinium-based contrast agent at 1.5
mL/sec for a duration of up to 40 sec.
Digital Subtraction Angiography
All patients with dural arteriovenous fistula and all control subjects
underwent angiography with catheters placed in the common carotid arteries,
internal carotid arteries, and external carotid arteries. Angiograms of the
dominant vertebral arteries were obtained in all patients with dural
arteriovenous fistula and in all control subjects. At our hospital,
angiography examinations in five or six vessels are routinely performed for
nonemergent evaluation in patients suspected of having this condition. In
patients with dural arteriovenous fistula, superselective catheterization of
the external branches of the carotid artery, including a delayed venous phase,
was performed. All patients in the group with dural arteriovenous fistula had
moderate- to high-flow dural arteriovenous fistula verified on digital
subtraction angiography by the appearance of venous structures during either
the early arterial or midarterial phases. The presence or absence of dural
arteriovenous fistula, of retrograde cortical venous drainage, and of venous
sinus occlusion was evaluated on the basis of findings of digital subtraction
angiography by the consensus of two neuroradiologists who did not participate
in the interpretation of the blinded MR angiography study.
Image Interpretation
Three-dimensional TOF MR angiograms and MR digital subtraction angiograms
of all patients with dural arteriovenous fistula and of the control group were
reviewed by two neuroradiologists who were unaware of patients' histories and
findings of physical examination and angiography. Disagreements were resolved
by consensus. The presence or absence of dural arteriovenous fistula, of
retrograde cortical venous drainage, and of venous sinus occlusion was
evaluated. The criteria for a diagnosis of dural arteriovenous fistula on
source images of 3D TOF included the presence of multiple high-intensity
curvilinear or nodular structures adjacent to the sinus wall and
high-intensity areas in the venous sinus. The criterion for a diagnosis of
dural arteriovenous fistula on MR digital subtraction angiography was early
filling of the venous sinus. The early filling of the venous sinus was defined
as venous sinus enhancement on the first intracranial arterial phase of MR
digital subtraction angiography, in which the hyperintensity in the venous
sinus exceeded that of the intracranial artery. We evaluated the sensitivity
and specificity of individual MR angiography findings such as multiple
high-intensity curvilinear or nodular structures adjacent to the sinus wall,
high-intensity areas in the venous sinus, and early filling of the venous
sinus of dural arteriovenous fistula, using catheter angiography studies as
the gold standard. Ninety-five percent confidence intervals (CIs) were
calculated for sensitivity and specificity using SPSS statistical software
(Statistical Package for the Social Sciences, Tokyo, Japan).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The findings of multiple high-intensity curvilinear or nodular structures adjacent to the sinus wall and high-intensity areas in the venous sinus were observed on the source images of 3D TOF MR angiography in patients with dural arteriovenous fistula. Multiple high-intensity curvilinear or nodular structures adjacent to the sinus wall findings were observed in all cases of dural arteriovenous fistula (Figs. 1A, 1D, and 2A). High intensity within the venous sinus was observed in 13 of 17 cases of dural arteriovenous fistula (Figs. 1A and 2A). Multiple high-intensity curvilinear or nodular structures adjacent to the sinus wall findings were not observed in any control cases. High-intensity areas within the venous sinus findings were observed in five of 35 control cases. The sensitivity and specificity (with 95% CI) of multiple high-intensity curvilinear or nodular structures adjacent to the sinus wall findings were 100% (77–100%) and 100% (88–100%), respectively (Table 1). The sensitivity and specificity (with 95% CI) of high-intensity areas in the venous sinus were 76% (50–92%) and 86% (69–95%), respectively. All false-positive findings of high-intensity areas within the venous sinus were observed in control sbjects at the transverse sinus, sigmoid sinus, and internal jugular vein on the smaller (hypoplastic) side of venous sinuses.
|
|
|
|
In a preliminary study, the average transit-time difference between the internal carotid artery and the torcular herophili was 5.5 sec (range [± SD], 4–7 ± 1.6 sec). These preliminary results suggested that MR digital subtraction angiography with 4-sec time resolution used in this study could show the separation of arterial and venous phases.
The findings of early filling of the venous sinus were not observed on MR digital subtraction angiography in any of the control subjects. MR digital subtraction angiography showed only normal flow of contrast agent, initially in arteries and later in venous structures (Fig. 3A). The findings of early filling of the venous sinus were observed on MR digital subtraction angiography in 13 of 15 patients with dural arteriovenous fistula (Figs. 1B and 4A). MR digital subtraction angiography could not show two cases of dural arteriovenous fistula because they were small (Fig. 1E). The sensitivity and specificity (with 95% CI) of early filling of the venous sinus findings were 87% (58–98%) and 100% (88–100%), respectively (Table 1).
|
|
|
|
Although 3D TOF MR angiography failed to show the findings of retrograde cortical venous drainage and venous sinus occlusion, MR digital subtraction angiography clearly showed both findings in all five cases (Figs. 4B, 4C, 4D).
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Awad et al. [11] reported that cortical venous drainage and variceal or aneurysmal venous dilatation strongly correlated with progressive neurologic deficits and intracranial hemorrhage. Cognard et al. [12] reported that the finding of retrograde cortical venous drainage is identified as a major risk factor for aggressive behavior of dural arteriovenous fistula, including intracranial hemorrhage. Therefore, dural arteriovenous fistulas were previously grouped into benign or aggressive categories on the basis of the presence or absence of retrograde cortical venous drainage.
The goal of treatment is obliteration of cortical venous drainage and venous hypertension. Hurst et al. [13] reported a series of patients with dementia related to dural arteriovenous fistula with venous hypertension, and after the arterial embolization, cognitive improvement occurred in all five patients. However, many patients do not come to medical attention because of the frequently benign clinical course. In addition, spontaneous regression of the lesions without intervention has rarely been reported [14] (Fig. 2A, 2B, 2C).
|
|
Although intraarterial catheter angiography is an invasive procedure, it is
often performed on patients in whom the clinical suspicion of dural
arteriovenous fistula is high to document the detailed anatomy of the arterial
pedicles and the exact relationship of normal cortical veins to the site of
the fistula. However, many patients present with symptoms or signs that might
be attributable to a dural arteriovenous fistula, but few actually show a
vascular malformation. A noninvasive examination that could accurately triage
patients with dural arteriovenous fistula into groups with high and low risk
of complications would be
helpful.
,
|
|
The protocol in our study could give reliable information about patients who should undergo invasive catheter angiography. For example, this protocol may be helpful for the diagnosis of dural arteriovenous fistula in patients with a history of allergy to iodinated contrast material or in patients with high risk factors of neurologic complications related to catheter angiography. In these patients, negative findings on 3D TOF MR angiography and MR digital subtraction angiography may be sufficient to omit catheter angiography from the workup. If a group of patients at relatively high risk of harboring a dural arteriovenous fistula could be identified, the use of these MR angiography techniques could lead to a more tailored catheter angiography, possibly allowing the relevant vessels to be catheterized earlier and less contrast agent to be used. In addition, these MR angiography techniques might distinguish high-risk from low-risk patients with dural arteriovenous fistula and help select patients for different treatments. Moreover, these MR angiography techniques may be helpful for follow-up of patients with dural arteriovenous fistula under close observation or treatment (Figs. 1A, 1B, 1C, 1D, 1E, 1F and 2A, 2B, 2C).
|
|
In patients with dural arteriovenous fistula, multiple high-intensity curvilinear or nodular structures adjacent to the sinus wall and high-intensity areas within the venous sinus findings were depicted on the source images of 3D TOF MR angiography. The multiple high-intensity curvilinear or nodular structures adjacent to sinus wall findings may correspond to meningeal branches of feeding arteries, representing the feeding pedicles of the dural arteriovenous fistula; and the high-intensity areas in the venous sinus finding may correspond to secondary increased unsaturated flow via an arteriovenous shunt. High-intensity areas in the venous sinus findings were observed in control subjects (14%) without arteriovenous shunt. False-positive findings of high-intensity areas in the venous sinus findings were detected at the smaller (hypoplastic) side of venous sinuses. Although the exact mechanism of false-positive high-intensity areas within the venous sinus findings is not clear, it could be the result of direct unsaturated flow from the temporal to occipital lobes, which bypasses the superiorly placed saturation pulse.
The source images of 3D TOF MR angiography are more useful than those of maximum-intensity-projection images in showing dural arteriovenous fistula. The source images of 3D TOF MR angiography contain both flow and anatomic information. Small intracranial structures can be seen by virtue of the high resolution provided by the acquisition of submillimeter sections and flow-related enhancement. Although the maximum-intensity-projection algorithm provides angiogram-like images, the lower-signal-intensity features of the vessels may be lost, and this loss may result in a lack of visibility of small or slow-flowing vessels.
MR digital subtraction angiography with 4-sec temporal resolution was performed in our study. The preliminary results of the control group showed that the average transit-time difference between the intracranial internal carotid artery and the venous sinus was 5.5 sec (range, 4–7 sec) with gadolinium-based contrast agent. Similar results were reported by Coley et al. [3], who found a 5-sec transit-time difference between the internal carotid artery and the veins of Galen with a gadolinium-based contrast agent. MR digital subtraction angiography with 4-sec temporal resolution could show the separation of the arterial and venous phases in our study. However, this acquisition time is only marginally adequate and needs to be improved using a new MRI technique such as sensitivity encoding [15].
Although 3D TOF MR angiography was not sensitive for the evaluation of cortical venous drainage in our study, probably because of saturation of retrograde flow, MR digital subtraction angiography clearly showed the retrograde cortical venous drainage in sequential images of the arterial to venous phase. The occlusion of the venous sinus was also correctly depicted on MR digital subtraction angiography, which may be useful in the evaluation of patients at risk for an aggressive course.
MR digital subtraction angiography was performed with a bolus injection of 0.1 mmol/kg of gadolinium-based contrast agent at 1.5 mL/sec in our study. Mitsuzaki et al. [16] reported that increased contrast enhancement in an artery without venous enhancement and better temporal separation between arterial and venous enhancement can be obtained with a higher injection rate for gadolinium-based contrast agents. Therefore, the inability to show early venous filling in two cases of small dural arteriovenous fistulas might have been due to the relatively low injection rate. For example, the inability to show early venous filling after embolization may suggest the decreased sensitivity of the MR digital subtraction angiography due to decreased flow and size of lesions (Fig. 1E). In future studies, using higher injection rates may reduce the false-negative rate of this finding.
Our study has some important limitations. First, it did not include cases of dural arteriovenous fistula with low-flow arteriovenous shunt. Dural arteriovenous fistula with very slow-flow arteriovenous shunt may not be shown on source images of both 3D TOF MR angiography and MR digital subtraction angiography. Further studies with a larger number of patients with dural arteriovenous fistulas, including dural arteriovenous fistula with slow-flow arteriovenous shunt, are necessary. Second, compared with catheter angiography, 3D TOF MR angiography also has some limitations, such as lower resolution and a restricted field of coverage, and the latter point may be somewhat ameliorated by the MR digital subtraction angiography technique. Third, MR digital subtraction angiography is limited in the full evaluation of dural arteriovenous fistula because of lower spatial and temporal resolution, compared with catheter angiography.
In conclusion, the findings of multiple high-intensity curvilinear or nodular structures adjacent to the sinus wall finding on source images of 3D TOF MR angiography were reliable for the detection of dural arteriovenous fistula with moderate- to high-flow arteriovenous shunt, and MR digital subtraction angiography clearly showed retrograde cortical venous drainage, a major risk factor for poor clinical outcome in patients with dural arteriovenous fistula. The combination of 3D TOF MR angiography (source images) and MR digital subtraction angiography is reliable in the screening for dural arteriovenous fistula with moderate- to high-flow arteriovenous shunt and can also permit a follow-up of treated lesions and those that are managed conservatively in some cases. As a result, we may be able to decrease the number of invasive intraarterial catheter angiographies for the screening of patients with, or suspected of having, dural arteriovenous fistula.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Meckel, M. Maier, D. S. M. Ruiz, H. Yilmaz, K. Scheffler, E.-W. Radue, and S.G. Wetzel MR Angiography of Dural Arteriovenous Fistulas: Diagnosis and Follow-Up after Treatment Using a Time-Resolved 3D Contrast-Enhanced Technique AJNR Am. J. Neuroradiol., May 1, 2007; 28(5): 877 - 884. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. -P. Chung, H. -Y. Hsu, A. -C. Chao, F. -C. Chang, W. -Y. Sheng, and H. -H. Hu Detection of intracranial venous reflux in patients of transient global amnesia Neurology, June 27, 2006; 66(12): 1873 - 1877. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Noguchi, M. Kubo, N. Kuwayama, Y. Kamisaki, G. Tomizawa, K. Kameda, H. Kawabe, S. Ogawa, N. Watanabe, S. Endo, et al. Intracranial Dural Arteriovenous Fistulas with Retrograde Cortical Venous Drainage: Assessment with Cerebral Blood Volume by Dynamic Susceptibility Contrast Magnetic Resonance Imaging AJNR Am. J. Neuroradiol., June 1, 2006; 27(6): 1252 - 1256. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Alatakis, G. Koulouris, and S. Stuckey CT-Demonstrated Transcalvarial Channels Diagnostic of Dural Arteriovenous Fistula AJNR Am. J. Neuroradiol., October 1, 2005; 26(9): 2393 - 2396. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-Y. Gauvrit, X. Leclerc, C. Oppenheim, T. Munier, D. Trystram, H. Rachdi, F. Nataf, J.-P. Pruvo, and J.-F. Meder Three-Dimensional Dynamic MR Digital Subtraction Angiography Using Sensitivity Encoding for the Evaluation of Intracranial Arteriovenous Malformations: A Preliminary Study AJNR Am. J. Neuroradiol., June 1, 2005; 26(6): 1525 - 1531. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. U. Harrer, O. Popescu, H. H. Henkes, and C. Klotzsch Assessment of Dural Arteriovenous Fistulae by Transcranial Color-Coded Duplex Sonography Stroke, May 1, 2005; 36(5): 976 - 979. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ziyeh, R. Strecker, A. Berlis, J. Weber, J. Klisch, and I. Mader Dynamic 3D MR Angiography of Intra- and Extracranial Vascular Malformations at 3T: A Technical Note AJNR Am. J. Neuroradiol., March 1, 2005; 26(3): 630 - 634. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
