Neurosurg Focus / Volume 32 / May 2012

Neurosurg Focus 32 (5):E15, 2012

1

Although spinal DAVF is the most common varia-tion of spinal vascular malformations, account-ing for 60%–80% of cases,9,13,23 they are still rare

entities that affect only an estimated 5–10 patients per million per year.17,21 They are slow-flow arteriovenous shunts characterized by an abnormal connection between a radicular artery and vein that is found most often in, or near, the dura of a nerve root sleeve.8,13,16,17,23 The re-sulting venous congestion and hypertension, first sug-gested by Aminoff et al. in 1974,4 eventually results in a progressive congestive myelopathy from the decrease in the arteriovenous pressure gradient, cord ischemia, and eventually infarction with neuronal loss.8,13,15,17,21 Hassler and Thron12 confirmed this hypothesis by demonstrating a venous pressure in a DAVF of approximately 70% of the patient’s systemic arterial pressure. The description by Marie et al.27 of a subacute necrotizing myelopathy is now widely accepted to represent the same pathological process.

Spinal DAVF is considered an acquired disease whose underlying cause remains unknown.13,17,20,23 Pa-tients most often present in the 5th or 6th decade of life, with a strong male predominance of approximately 5:1 in reported series.8,16,17 Although the majority of spinal DAVFs are found in the mid- to lower thoracic spine,9 ve-nous congestion and subsequent myelopathy often begins in the caudal aspect of the spinal cord irrespective of the actual fistula level.17,23 Patients consequently most often present with nonspecific symptoms involving changes in sensation and strength in the lower extremities, lower-extremity pain sometimes exacerbated by exercise, and disturbances in gait.6,8,13,16,29 These symptoms overlap significantly with those of more common diseases such as spinal canal stenosis and polyneuropathy.9,16,17 As the myelopathy progresses, patient often will experience con-tinued loss of lower-extremity strength leading to para-plegia, as well as develop abnormalities in micturition, defecation, and sexual function.17

Conventional MRI is frequently the initial diagnostic examination in patients with a spinal DAVF, suspected or otherwise, and this represents a critical decision point for

Absence of abnormal vessels in the subarachnoid space on conventional magnetic resonance imaging in patients with spinal dural arteriovenous fistulas

TimoThy R. milleR, m.D.,1 CliffoRD J. eskey, m.D., Ph.D.,2 anD alexanDeR C. mamouRian, m.D.1

1Division of Neuroradiology, Department of Radiology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania; and 2Neuroradiology, Department of Radiology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire

Spinal dural arteriovenous fistula (DAVF) is an uncommon condition that can be difficult to diagnose. This often results in misdiagnosis and treatment delay. Although conventional MRI plays an important role in the initial screening for the disease, the typical MRI findings may be absent. In this article, the authors present a series of 4 cases involving patients with angiographically proven spinal DAVFs who demonstrated cord T2 prolongation on conventional MRI but without abnormal subarachnoid flow voids or enhancement. These cases suggest that spinal DAVF cannot be excluded in symptomatic patients with cord edema based on conventional MRI findings alone. Dy-namic Gd-enhanced MR angiography (MRA) was successful in demonstrating abnormal spinal vasculature in all 4 cases. This limited experience provides support for the role of spinal MRA in patients with abnormal cord signal and symptoms suggestive of DAVF even when typical MRI findings of a DAVF are absent.(http://thejns.org/doi/abs/10.3171/2012.2.FOCUS1214)

key WoRDs      •      spinal dural arteriovenous fistula      •      magnetic resonance angiography      •    perimedullary flow voids

1

Abbreviations used in this paper: AVF = arteriovenous fistula; DAVF = dural AVF; MRA = MR angiography.

Unauthenticated | Downloaded 07/28/20 03:11 AM UTC

T. R. Miller, C. J. Eskey, and A. C. Mamourian

2 Neurosurg Focus / Volume 32 / May 2012

the diagnosis.10,13,17,24 The MRI findings of spinal cord T2 prolongation, intramedullary enhancement, and promi-nent vessels in the subarachnoid space as demonstrated by flow voids on T2-weighted imaging and enhancement on postcontrast T1-weighted imaging are considered characteristic for this disease.11,14,17,22,24 Although this combination of findings on conventional MRI is thought to be both fairly sensitive and specific for the diagnosis of a spinal vascular malformation, these findings may not all be present in a significant proportion of patients.11,13,14,22,24

We present a series of 4 cases involving patients with proven spinal DAVF who demonstrated spinal cord edema and enhancement without abnormal flow voids or enhancement in the subarachnoid space on conventional MRI. Three patients presented at our academic refer-ral center, and the fourth presented at another academic practice. Three patients were male and one was female, and their ages ranged from 57 to 72 years. All patients were evaluated by conventional MRI, consisting of sagit-tal T1-weighted, sagittal and axial T2-weighted, and Gd-enhanced sagittal and axial T1-weighted imaging. The diagnosis of spinal DAVF was subsequently confirmed by conventional spinal angiography, and all patients went on to be treated with surgical, endovascular, or multidis-ciplinary approaches. Dynamic Gd-enhanced MRA was performed prior to catheter angiography in 3 patients and was successful at suggesting the diagnosis of a spinal DAVF by demonstrating abnormal vessels on the surface of the spinal cord; MRA also helped to target subsequent catheter angiography in these patients by correctly iden-tifying the fistula level. In a fourth patient, dynamic Gd-enhanced MRA was performed to evaluate for residual flow in a cervical DAVF after an attempted endovascular embolization; MRA correctly identified the residual fis-tula, which was subsequently closed.

Case ReportsCase 1

This 70-year-old man with a history of diabetes, pe-ripheral vascular disease, and arthritis presented to our institution with a 4-year history of progressive lower-extremity pain and weakness. The patient’s initial symp-toms consisted of bilateral lower-extremity pain on exer-tion, and he was subsequently diagnosed with claudication secondary to peripheral vascular disease. He underwent lower-extremity arterial angioplasty on 2 separate occa-sions for these symptoms 4 and 2 years prior to presenting to our institution. The patient had also received regular Synvisc knee injections for osteoarthritis with temporary relief of his lower-extremity pain. At presentation, the pa-tient was using a walker to ambulate and had bilateral hip flexor weakness of (4+/5), as well as slight weakness of left knee extension (4+/5). A Romberg sign was present, and the patient had a wide-based, unsteady gait. He also had markedly decreased vibration and proprioception sensation to the midshins bilaterally. Both temperature and light touch sensation in the lower extremities were intact. Deep tendon reflexes were 1+ at the patella; there was no right ankle jerk, and the left ankle jerk was 1+.

Conventional MRI demonstrated a long segment of

increased intramedullary T2 prolongation in the thoracic cord beginning at the T5–6 level and extending to the conus. There was no abnormal flow void or enhance-ment noted in the subarachnoid space. The presence of a DAVF, however, was considered based on the patient’s symptoms, and he therefore underwent dynamic contrast-enhanced MRA. That examination demonstrated abnor-mal vessels dorsal to the inferior thoracic cord and conus medullaris. A prominent vessel was evident extending to the left L-1 neural foramen, suggesting the level of the fistula. Follow-up spinal angiography confirmed the pres-ence of a spinal DAVF originating from the left T12–L1 level with enlargement of the coronal venous plexus (Fig. 1). More precise localization was not possible due to tor-tuosity and atherosclerosis involving the aortic segmental arteries. The patient subsequently underwent a success-ful microsurgical resection of the fistula. On follow-up examinations his lower-extremity weakness and sensory changes remained stable, likely due to the long-standing duration of his myelopathy prior to treatment, but no pro-gression was evident.

Case 2This 72-year-old man presented to an outside insti-

tution with a 2-month history of progressive upper- and lower-extremity myelopathic symptoms of transient par-esthesias, as well as mild gait difficulty. Conventional MRI of the cervical spine performed at the outside hos-pital at 2 time points demonstrated abnormal T2 prolon-gation throughout the cervical and upper thoracic spinal cord, as well as anterior intramedullary enhancement in the cervical cord. The patient was referred to our institu-tion for a spinal cord biopsy because of concern for an un-derlying spinal cord tumor. A review of the outside MRI studies revealed no evidence of abnormal flow void or en-hancement in the cervical or thoracic subarachnoid space. Cervical spinal canal stenosis secondary to degenerative disease was present, however.

Based on our previous experience and the patient’s symptoms, spinal angiography was performed to evaluate the possibility of a spinal vascular malformation. A cervi-cal DAVF was found—supplied by an intradural branch of the right vertebral artery and with venous drainage both anterior and posterior to the cervical cord. An attempt was subsequently made to embolize the fistula. However, following the intervention there was clinical concern for continued patency of the DAVF, which was confirmed by follow-up dynamic contrast-enhanced MRA demonstrat-ing enlarged, tortuous vessels along the surface of the cervical cord (Fig. 2).

This case was subsequently managed with a multidis-ciplinary approach. After the failed endovascular embo-lization, microsurgical ligation of an arterialized vein at the C-1 level was performed, but the patient’s symptoms did not improve, and a follow-up conventional angiogram demonstrated an additional, previously unrecognized, ar-terial feeder arising from the intradural right vertebral ar-tery. This second arterial feeder was successfully embo-lized with Onyx, and a final postintervention angiogram demonstrated no evidence of residual DAVF. The patient has subsequently undergone extensive physical therapy

Unauthenticated | Downloaded 07/28/20 03:11 AM UTC

Neurosurg Focus / Volume 32 / May 2012

Absence of typical MRI findings in patients with spinal DAVF

3

with improvement in his gait, to the point where he can ambulate with a cane or walker, and resolution of his par-esthesias.

Case 3This 57-year-old man with a medical history signifi-

cant for Lyme disease presented to our institution with a 4-month history of progressive lower-extremity paresthe-sias, extending to the perineum, and gait instability. Con-ventional MRI of the spine demonstrated a long segment of abnormal, intramedullary, T2 prolongation involving the thoracic spinal cord with associated faint enhance-ment on Gd-enhanced T1-weighted imaging. No abnor-

mal flow voids or abnormal enhancement were evident in the subarachnoid space.

Dynamic contrast-enhanced MRA, however, demon-strated prominent, tortuous vessels anterior to the midtho-racic cord suggestive of a vascular malformation. In addi-tion, MRA suggested potential fistula levels on the right at T7–8 and on the left at T6–7. Conventional spinal angiog-raphy demonstrated a DAVF that was supplied by both a radicular branch of the right T-8 intercostal artery, with the artery of Adamkiewicz arising from the left T-7 intercostal artery (Fig. 3). The fistula was subsequently successfully embolized using a combination of 30% n-butyl 2-cyanoac-rylate glue and 70% Lipiodol. Four-year clinical follow-up

Fig. 1. Case 1. A and B: Sagittal thoracic T2-weighted (A) and Gd-enhanced T1-weighted (B) images demonstrating a long segment of T2 hyperintensity and intramedullary enhancement in the thoracic spinal cord extending to the conus medullaris in a 70-year-old man with gradual onset of bilateral lower-extremity weakness and alteration of sensation. There is no evidence of abnormal flow voids or enhancement in the subarachnoid space to suggest abnormal spinal vasculature. C: Dynamic contrast-enhanced MR angiogram demonstrating abnormal, dilated, and tortuous vessels extending along the dorsal aspect of the tho-racic spinal cord raising the possibility of a spinal DAVF. Magnetic resonance angiography was also suggestive of a possible fistula location on the left at the L-1 level. D: Image subsequently obtained by catheter angiography, confirming the diagnosis and demonstrating the feeding radicular artery arising from the left T12–L1 level.

Fig. 2. Case 2. A and B: Sagittal cervical-thoracic T2-weighted (A) and Gd-enhanced T1-weighted (B) images demonstrat-ing a long segment of T2 hyperintensity throughout the cervical and superior thoracic spinal cord with anterior intramedullary enhancement in the cervical spine in a 72-year-old man with upper and lower extremity myelopathy symptoms who was referred to our institution for spinal cord biopsy due to concerns for an intramedullary tumor. There is no evidence of abnormal flow voids or enhancement in the subarachnoid space to suggest abnormal spinal vasculature. C: Dynamic contrast-enhanced MR an-giogram obtained after attempted endovascular embolization demonstrating abnormal, dilated, and tortuous vessels extending along the ventral and dorsal aspects of the cervical cord suggestive of residual spinal DAVF. D: Image obtained by catheter angiography confirming the MRA findings and demonstrating arterial supply from a branch of the intradural right vertebral artery.

Unauthenticated | Downloaded 07/28/20 03:11 AM UTC

T. R. Miller, C. J. Eskey, and A. C. Mamourian

4 Neurosurg Focus / Volume 32 / May 2012

demonstrated improvement in the patient’s lower-extremity paresthesias with no residual gait abnormality.

Case 4This 66-year-old woman had progressive lower-

extremity weakness and paresthesias over the course of several months. A conventional MRI examination dem-onstrated a long segment of abnormal T2 prolongation in the thoracic spinal cord that extended to the conus medul-laris along with mild intramedullary enhancement. No ab-normal flow voids or enhancement were identified in the subarachnoid space. However, dynamic contrast-enhanced MRA demonstrated prominent, tortuous vessels along the dorsal aspect of the thoracic spinal cord, which—along with the symptoms—suggested a spinal DAVF with possible lo-calization of the fistula at the right T-12 level. Follow-up conventional spinal angiography confirmed the presence of a spinal DAVF supplied by the right T-12 segmental artery (Fig. 4). The patient was treated successfully with surgical ligation of the fistula.

DiscussionA delay in the diagnosis of spinal DAVF can be at-

tributed to both the rarity of the diagnosis and the non-specific quality of the presenting symptoms.2,9,11,13 Jellema et al.17 found in their review of the literature that the delay between symptom onset and diagnosis ranged between 12 and 44 months. Unfortunately, it is not unusual for the patient’s symptoms to significantly progress during that time interval.17,20,23 These symptoms include worsening lower-extremity weakness as well as the onset of addi-tional symptoms, such as difficulties with micturition, defecation, and sexual function.9,17 Successful treatment requires occlusion of the draining, arterialized, intradural vein, which has been found to provide a lasting and dura-ble treatment of the symptoms of spinal DAVFs.1,8,25 Cur-

rently, surgical ligation, endovascular embolization, and a multidisciplinary approach are all used for the treatment of spinal DAVFs.9,30,32,33

Once the diagnosis of a spinal DAVF has been made and the fistula has been successfully closed, a majority of patients, up to 80%–90% in some series,17 will demon-strate either improvement or stabilization of symptoms, particularly lower-extremity strength.8,13 Aghakhani et al.2 showed that even patients with complete paraplegia at the time of treatment can demonstrate a significant im-provement in function following intervention. This rep-resents a dramatic divergence from the natural history of spinal DAVF, which consists of progressive deterioration, as demonstrated by Aminoff and Logue,5 who showed that 50% of patients will be wheelchair-bound 3 years after presentation. While a few studies have described a possible late deterioration of patient function on longer-term follow-up,7 these findings have served as an impe-tus to increase knowledge of spinal DAVF in the general medical community in the hope of reducing delays in di-agnosis and thereby improving patient outcomes.9,17

Spinal cord edema on conventional MRI has been found to be very sensitive for the diagnosis of spinal DAVF, with reported sensitivity close to 100%.10,18,19,22,29 However it is a nonspecific finding that can be attributed to much more common entities such as demyelination, trauma, and spinal canal stenosis secondary to degen-erative disease of the spine.8,10,17,18 Our experience shows that even using current MR technology, the more specific findings of abnormal vasculature in the subarachnoid space may be absent in some patients with spinal DAVFs on conventional MR images. Thiex et al.34 reported a sim-ilar experience in a patient with a spinal DAVF in whom conventional MRI failed to demonstrate abnormal ves-sels in the subarachnoid space. The results of earlier and larger published series support our assertion, with some reporting only 45%–55% of spinal DAVF patients dem-

Fig. 3. Case 3. A and B: Sagittal thoracic T2-weighted (A) and Gd-enhanced T1-weighted (B) images demonstrate a long segment of T2 hyperintensity with minimal intramedullary enhancement in the thoracic spinal cord in a 57-year-old man with partial lower-extremity paraplegia, radiating leg pain, and new-onset urinary retention. No abnormal flow voids or enhancement are evident in the subarachnoid space to suggest abnormal spinal vasculature. C: Dynamic contrast-enhanced MR angiogram demonstrating some dilated and tortuous vessels extending along the ventral aspect of the thoracic spinal cord raising the pos-sibility of a spinal DAVF. The MRA study was suggestive of a possible fistula level on the right at T7–8 and on the left at T6–7 (not shown). D: Image obtained by catheter angiography confirming the presence of a spinal DAVF supplied by a feeding radicular artery arising at the left T-7 level.

Unauthenticated | Downloaded 07/28/20 03:11 AM UTC

Neurosurg Focus / Volume 32 / May 2012

Absence of typical MRI findings in patients with spinal DAVF

5

onstrating abnormal spinal vasculature on conventional MRI.11,13,14,17,18,22 However, this lower range of reported sensitivities may be attributable, at least in part, to the limitations of less sophisticated MRI units in earlier stud-ies.

The diagnosis of a spinal DAVF therefore cannot be excluded when spinal cord edema is encountered on conventional MRI in the appropriate clinical setting. However, the lack of specificity of spinal cord edema and symptoms alone are frequently insufficient to justify con-ventional spinal angiography, given its small, but signifi-cant, risks and complexity.

Although dynamic Gd-bolus MRA may have a role in patients with typical findings of DAVF because it can reduce both fluoroscopy time and contrast agent volume10,20,26,28,31 necessary on conventional angiography, our small case series suggests that it may have a very im-portant role in establishing the diagnosis of spinal DAVF. Other published reports demonstrating high sensitivity and reasonably high specificity of dynamic Gd-enhanced MRA for detection of spinal DAVFs support this asser-tion.3,10,26

Clinical history and the suspicion of a spinal DAVF should ultimately remain the deciding factors that deter-mine whether one should proceed with spinal angiog-raphy. However, in cases involving patients who do not present with diagnostic MRI findings or typical symp-toms of a spinal DAVF, MRA examination may provide compelling evidence to go on to invasive conventional angiography.

Disclosure

The authors report no conflict of interest concerning the mate-rial or methods used in this study or the findings specified in this paper.

Author contributions to the study and manuscript preparation include the following. Conception and design: all authors. Analysis

and interpretation of data: Miller, Mamourian. Drafting the article: Miller. Critically revising the article: all authors. Reviewed submit-ted version of manuscript: all authors. Study supervision: Eskey, Mamourian.

References

1. Afshar JK, Doppman JL, Oldfield EH: Surgical interruption of intradural draining vein as curative treatment of spinal du-ral arteriovenous fistulas. J Neurosurg 82:196–200, 1995

2. Aghakhani N, Parker F, David P, Lasjaunias P, Tadie M: Cur-able cause of paraplegia: spinal dural arteriovenous fistulae. Stroke 39:2756–2759, 2008

3. Ali S, Cashen TA, Carroll TJ, McComb E, Muzaffar M, Shaiba-ni A, et al: Time-resolved spinal MR angiography: initial clini-cal experience in the evaluation of spinal arteriovenous shunts. AJNR Am J Neuroradiol 28:1806–1810, 2007

4. Aminoff MJ, Barnard RO, Logue V: The pathophysiology of spinal vascular malformations. J Neurol Sci 23:255–263, 1974

5. Aminoff MJ, Logue V: Clinical features of spinal vascular malformations. Brain 97:197–210, 1974

6. Aminoff MJ, Logue V: The prognosis of patients with spinal vascular malformations. Brain 97:211–218, 1974

7. Cecchi PC, Musumeci A, Rizzo P, Faccioli F, Bricolo A: Late deterioration of neurologic function in patients surgically treated for spinal dural arteriovenous fistulas. Surg Neurol 72:257–262, 2009

8. da Costa L, Dehdashti AR, terBrugge KG: Spinal cord vascu-lar shunts: spinal cord vascular malformations and dural arte-riovenous fistulas. Neurosurg Focus 26(1):E6, 2009

9. Dehdashti AR, Da Costa LB, terBrugge KG, Willinsky RA, Tymianski M, Wallace MC: Overview of the current role of endovascular and surgical treatment in spinal dural arteriove-nous fistulas. Neurosurg Focus 26(1):E8, 2009

10. Eddleman CS, Jeong H, Cashen TA, Walker M, Bendok BR, Batjer HH, et al: Advanced noninvasive imaging of spinal vascular malformations. Neurosurg Focus 26(1):E9, 2009

11. Gilbertson JR, Miller GM, Goldman MS, Marsh WR: Spinal dural arteriovenous fistulas: MR and myelographic findings. AJNR Am J Neuroradiol 16:2049–2057, 1995

Fig. 4. Case 4. A and B: Sagittal thoracic T2-weighted (A) and Gd-enhanced T1-weighted (B) images obtained in a 66-year-old woman with progressive lower-extremity myelopathy symptoms demonstrating a long segment of T2 hyperintensity in the thoracic spinal cord extending to the conus with associated intramedullary enhancement. There were no definite abnormal flow voids or enhancement in the subarachnoid space. The patient subsequently underwent an extensive diagnostic workup for numerous possible causes, including transverse myelitis, malignancy, demyelination, and infectious processes. C: Dynamic contrast-enhanced MR angiogram demonstrating abnormal, dilated, and tortuous vessels extending along the dorsal aspects of the thoracic spinal cord raising the possibility of a spinal DAVF. The MRA study was suggestive of a possible fistula level on the right at the T-12 level. D: Image obtained by catheter angiography confirming the presence of a spinal DAVF at this level.

Unauthenticated | Downloaded 07/28/20 03:11 AM UTC

T. R. Miller, C. J. Eskey, and A. C. Mamourian

6 Neurosurg Focus / Volume 32 / May 2012

12. Hassler W, Thron A: Flow velocity and pressure measure-ments in spinal dural arteriovenous fistulas. Neurosurg Rev 17:29–36, 1994

13. Hessler C, Regelsberger J, Grzyska U, Illies T, Zeumer H, West-phal M: Therapeutic clues in spinal dural arteriovenous fistu-las—a 30 year experience of 156 cases. Cen Eur Neu ro surg 71:8–12, 2010

14. Hurst RW, Grossman RI: Peripheral spinal cord hypointen-sity on T2-weighted MR images: a reliable imaging sign of venous hypertensive myelopathy. AJNR Am J Neuroradiol 21:781–786, 2000

15. Hurst RW, Kenyon LC, Lavi E, Raps EC, Marcotte P: Spinal dural arteriovenous fistula: the pathology of venous hyperten-sive myelopathy. Neurology 45:1309–1313, 1995

16. Jellema K, Tijssen CC, Sluzewski M, van Asbeck FWA, Koud-staal PJ, van Gijn J: Spinal dural arteriovenous fistulas—an underdiagnosed disease. A review of patients admitted to the spinal unit of a rehabilitation center. J Neurol 253:159–162, 2006

17. Jellema K, Tijssen CC, van Gijn J: Spinal dural arteriovenous fistulas: a congestive myelopathy that initially mimics a pe-ripheral nerve disorder. Brain 129:3150–3164, 2006

18. Jones BV, Ernst RJ, Tomsick TA, Tew J Jr: Spinal dural ar-teriovenous fistulas: recognizing the spectrum of magnetic resonance imaging findings. J Spinal Cord Med 20:43–48, 1997

19. Kendall BE, Logue V: Spinal epidural angiomatous mal for-mations draining into intrathecal veins. Neuroradiology 13: 181–189, 1977

20. Klopper HB, Surdell DL, Thorell WE: Type I spinal dural arteriovenous fistulas: historical review and illustrative case. Neurosurg Focus 26(1):E3, 2009

21. Koch C: Spinal dural arteriovenous fistula. Curr Opin Neu-rol 19:69–75, 2006

22. Koch C, Kucinski T, Eckert B, Röther J, Zeumer H: [Spinal dural arteriovenous fistula: clinical and radiological findings in 54 patients.] Rofo 175:1071–1078, 2003 (Ger)

23. Krings T, Geibprasert S: Spinal dural arteriovenous fistulas. AJNR Am J Neuroradiol 30:639–648, 2009

24. Krings T, Lasjaunias PL, Hans FJ, Mull M, Nijenhuis RJ, Alva-rez H, et al: Imaging in spinal vascular disease. Neuroimaging Clin N Am 17:57–72, 2007

25. Logue V: Angiomas of the spinal cord: review of the pathogen-esis, clinical features, and results of surgery. J Neurol Neuro-surg Psychiatry 42:1–11, 1979

26. Luetmer PH, Lane JI, Gilbertson JR, Bernstein MA, Huston J III, Atkinson JL: 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 26: 711–718, 2005

27. Marie P, Foix C, Alajouanine T: La myélite necrotique subai-gue. Rev Neurol (Paris) 2:1–42, 1926

28. Mull M, Nijenhuis RJ, Backes WH, Krings T, Wilmink JT, Thron A: Value and limitations of contrast-enhanced MR an-giography in spinal arteriovenous malformations and dural arteriovenous fistulas. AJNR Am J Neuroradiol 28:1249–1258, 2007

29. Muralidharan R, Saladino A, Lanzino G, Atkinson JL, Rabin-stein AA: The clinical and radiological presentation of spinal dural arteriovenous fistula. Spine 36:E1641–E1647, 2011

30. Ruiz-Juretschke F, Perez-Calvo JM, Castro E, García-Leal R, Mateo-Sierra O, Fortea F, et al: A single-center, long-term study of spinal dural arteriovenous fistulas with multidisci-plinary treatment. J Clin Neurosci 18:1662–1666, 2011

31. Saraf-Lavi E, Bowen BC, Quencer RM, Sklar EM, Holz A, Falcone S, et al: Detection of spinal dural arteriovenous fis-tulae with MR imaging and contrast-enhanced MR angiogra-phy: sensitivity, specificity, and prediction of vertebral level. AJNR Am J Neuroradiol 23:858–867, 2002

32. Sharma AK, Westesson PL: Preoperative evaluation of spinal vascular malformation by MR angiography: how reliable is the technique: case report and review of literature. Clin Neu-rol Neurosurg 110:521–524, 2008

33. Sivakumar W, Zada G, Yashar P, Giannotta SL, Teitelbaum G, Larsen DW: Endovascular management of spinal dural ar-teriovenous fistulas. A review. Neurosurg Focus 26(5):E15, 2009

34. Thiex R, Mayfrank L, Krings T, Mull M: Delayed diagnosis of spinal dural arteriovenous fistula in the absence of patho-logical vessels on MRI. Zentralbl Neurochir 67:94–98, 2006

Manuscript submitted January 12, 2012.Accepted February 6, 2012.Portions of this work were presented in poster form at the ASNR

49th Annual Meeting & the Foundation of the ASNR Symposium 2011, June 4–9, 2011, Seattle, Washington.

Please include this information when citing this paper: DOI: 10.3171/2012.2.FOCUS1214.

Address correspondence to: Timothy R. Miller, M.D., Hospital of the University of Pennsylvania, Division of Neuroradiology, 3400 Spruce Street, 2 Dulles Room 219, Philadelphia, Pennsylvania 19104. email: Timothy.Miller@uphs.upenn.edu.

Unauthenticated | Downloaded 07/28/20 03:11 AM UTC