DURAL FISTULAS DURAL FISTULAS

TOPIC

Advances in Surgical Approaches to Dural Fistulas Patrick P. Youssef, MD Albert Jess Schuette, MD C. Michael Cawley, MD Daniel L. Barrow, MD Department of Neurosurgery, Emory University School of Medicine, Atlanta, Georgia Correspondence: Daniel L. Barrow, MD, MBNA Bowman Professor and Chairman, Department of Neurosurgery, Director, Emory Stroke Center, Emory University School of Medicine, The Emory Clinic, 1365 B Clifton Road, NE, Atlanta, GA 30322. E-mail: [email protected] Received, August 28, 2013. Accepted, October 11, 2013. Copyright © 2014 by the Congress of Neurological Surgeons

Dural arteriovenous fistulas are abnormal connections of dural arteries to dural veins or venous sinuses originating from within the dural leaflets. They are usually located near or within the wall of a dural venous sinus that is frequently obstructed or stenosed. The dural fistula sac is contained within the dural leaflets, and drainage can be via a dural sinus or retrograde through cortical veins (leptomeningeal drainage). Dural arteriovenous fistulas can occur at any dural sinus but are found most frequently at the cavernous or transverse sinus. Leptomeningeal venous drainage can lead to venous hypertension and intracranial hemorrhage. The various treatment options include transarterial and transvenous embolization, stereotactic radiosurgery, and open surgery. Although many of the advances in dural arteriovenous fistula treatment have occurred in the endovascular arena, open microsurgical advances in the past decade have primarily been in the tools available to the surgeon. Improvements in microsurgical and skull base approaches have allowed surgeons to approach and obliterate fistulas with little or no retraction of the brain. Image-guided systems have also allowed better localization and more efficient approaches. A better understanding of the need to simply obliterate the venous drainage at the site of the fistula has eliminated the riskier resections of the past. Finally, the use of intraoperative angiography or indocyanine green videoangiography confirms the complete disconnection of fistula while the patient is still on the operating room table, preventing reoperation for residual fistulas. KEY WORDS: AVF, Borden classification, Cortical venous drainage, DAVF, Dural arteriovenous fistula, Surgery Neurosurgery 74:S32–S41, 2014

DOI: 10.1227/NEU.0000000000000228

D

ural arteriovenous fistulas (DAVFs) are abnormal connections of dural arteries to dural veins or venous sinuses originating from within the dural leaflets. They are usually located near or within the wall of a dural venous sinus that is frequently obstructed or stenosed. The dural fistula sac is contained within the dural leaflets, and drainage can be via a dural sinus or retrograde through cortical veins (leptomeningeal drainage). DAVFs can occur at any dural sinus but are most frequently found at the cavernous or transverse sinus.1 Symptoms depend on the pattern of venous drainage.2 Fistulas with drainage solely into venous dural sinuses can be associated with annoying but rarely life-threatening symptoms. Leptomeningeal venous drainage can lead to venous hypertension and intracranial hemorrhage. They can occur at any age but most

ABBREVIATIONS: CVD, cortical venous drainage; DAVF, dural arteriovenous fistula; ECA, external carotid artery; ICA, internal carotid artery

S32 | VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT

www.neurosurgery-online.com

commonly occur between the sixth and seventh decades of life.

CLASSIFICATION Several different classification schemes have been proposed to describe DAVFs. The most commonly used is the Borden classification system, which is based on the pattern of venous drainage of the fistula.3 Type I has drainage into the dural venous sinus only; type II has drainage into the dural venous sinus or meningeal vein with cortical venous reflux; and type III has cortical venous reflux only (Figure 1). The classification is important with respect to clinical presentation and prognosis. Intracranial hemorrhage, focal neurological deficit, or death was found in 2% of Borden type I, 39% of type II, and 79% of type III cranial DAVFs. An alternative classification system that is less commonly used is the Cognard classification system.4 The Cognard classification is based on the direction of dural sinus drainage, the presence or absence of cortical venous drainage

www.neurosurgery-online.com

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

SURGICAL APPROACHES TO DURAL FISTULAS

FIGURE 1. Catheter angiogram, anteroposterior projection, showing a Borden type III dural arteriovenous fistula of the lateral sinus with engorged cortical draining veins and cortical venous reflux (black arrow).

(CVD), and venous outflow architecture (nonectatic cortical veins, ectatic cortical veins, or spinal perimedullary veins). Type I lesions drain into the dural sinus, have an antegrade flow direction, and lack CVD. Type II lesions are subdivided into 3 subcategories: Type IIa lesions drain retrograde into a dural sinus without CVD; type IIb lesions drain antegrade into a dural sinus with CVD; and type IIa 1 b lesions drain retrograde into a dural sinus with CVD. Types III, IV, and V lesions all have CVD, absent dural venous drainage, and various cortical venous outflow architecture.5 Spontaneous carotid cavernous fistulas can be classified with the Barrow classification.6 Type A is a direct high-flow fistula from the cavernous internal carotid artery (ICA) to the cavernous sinus, commonly occurring from a ruptured cavernous aneurysm. Type B is an indirect fistula from meningeal branches of the ICA, commonly the meningohypophyseal or inferolateral trunks. Type C is an indirect fistula from meningeal branches of the external carotid artery (ECA). Type D fistulas are fed by both ICA and ECA branches. Treatment is dictated primarily by whether the fistula is direct or indirect.

NATURAL HISTORY The natural history of DAVFs has largely been determined from retrospective studies because of limitations in the detection of these lesions with noninvasive imaging. Davies et al7 followed up 98

NEUROSURGERY

patients with 102 intracranial DAVFs for an average of 33 months, with complete follow-up in 91%. In this series, 55 patients had DAVFs without leptomeningeal and cortical venous reflux, and of these, 32 received no treatment. Among the patients who received no treatment, 81% had symptom improvement or complete resolution compared with 86% of the treated patients with the same angioarchitecture. In the same series, there were 46 patients who had DAVFs with leptomeningeal or cortical venous reflux, and 14 of these declined further treatment. At follow-up, this group had an 11% nonhemorrhagic neurological deficit rate per year and a 20% intracerebral hemorrhage rate per year. In an updated review of 118 patients with DAVFs and leptomeningeal reflux, the Toronto team demonstrated an annual risk for nonhemorrhagic neurological deficit of 6.9% and a risk of 8.1% for hemorrhage, with an annual mortality rate of 10.4%.8,9 In a recent series from Brigham and Women’s Hospital and Harvard Medical School, 70 patients were combined with 395 historic DAVFs previously documented in 6 studies.10 No hemorrhages occurred during 409 lesion-years of Borden type I DAVFs, although CVD developed in 1.4%. Borden type II DAVFs presented with hemorrhage in 18% and had an annual hemorrhage rate of 6%. The lesions were most frequently transverse-sigmoid or cavernous in location. Borden type III DAVFs presented with hemorrhage in 34% and had an annual hemorrhage rate of 10%, which increased to 21% for those with venous ectasia. The lesions were most frequently tentorial or petrosal in location. Of note, the hemorrhage rate decreased to 2% for asymptomatic or minimally symptomatic type II or III DAVFs and increased to 10% for those presenting with nonhemorrhagic neurological deficits and 46% for those presenting with previous hemorrhage.10 Many DAVFs remain stable and do not change on follow-up. Some involute spontaneously, and some have been known to thrombose after angiography. Some DAVFs undergo progressive recruitment of pachymeningeal feeders with worsening of symptoms. The actual progress of lesions once symptoms develop is unknown. By location, an aggressive course is seen in 75% of anterior fossa lesions, 79% of tentorial lesions, 60% of foramen magnum lesions, and 29% of transverse sinus lesions.2 Factors predisposing to an aggressive course include leptomeningeal (cortical) venous drainage, galenic drainage (deep veins), variceal or aneurysmal venous dilatations, and location at the tentorial incisura. This risk of aggressive course parallels that of the leptomeningeal retrograde venous drainage at each of these locations.

PRESENTATION Clinical presentation from DAVFs is widely variable, and patients may be completely asymptomatic. In symptomatic patients, the symptoms may range from benign, subjective bruit or headache to fatal hemorrhage or nonhemorrhagic progressive neurological deterioration. The symptoms depend on the location

VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT | S33

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

YOUSSEF ET AL

and venous drainage pattern of the DAVF. A petrous region DAVF draining into the transverse or sigmoid sinus commonly produces headaches or pulsatile tinnitus, with or without an associated bruit. Carotid cavernous DAVFs often present with the classic clinical findings of proptosis, chemosis, and bruit. Without treatment, a carotid cavernous DAVF may cause ocular and visual complications, including a decrease in visual acuity or blindness.6 Benign symptoms include pulsatile tinnitus, which is caused by turbulent flow into the affected sinus. Bruit can be heard in 40% of patients with symptomatic tinnitus.11 Other benign symptoms can include facial pain, papilledema, glaucoma, ophthalmoplegia, or glaucoma. Aggressive symptoms such as intracranial hemorrhage, neurological deficit, and death are usually associated with CVD. Intracranial hemorrhage occurs from rupture of an arterialized leptomeningeal vein and commonly causes intraparenchymal hemorrhage, although all kinds of hemorrhage may be seen, including subdural, subarachnoid, and intraventricular hemorrhage. Neurological deficits are caused by ischemia or venous hypertension secondary to venous congestion.

IMAGING Magnetic resonance imaging (MRI) and computed tomography (CT) are frequently unremarkable in cases of DAVF without cortical venous reflux. In some instances, however, MRI and MR angiography can show stenosis or occlusion of the dural venous sinuses. Hydrocephalus may be seen on imaging if the DAVF causes hypertension in the superior sagittal sinus. CT angiography can identify some DAVFs with enlarged intracranial veins. In cases of DAVF with cortical venous reflux, CT and MRI can demonstrate hemorrhage, engorged pial vessels, and white matter edema (indicative of venous hypertension). Dilated pial vessels and white matter edema are more likely to be visualized on MRI (specifically T2 imaging as flow voids and increased white matter signal) compared with CT. Given the limitations of MRI and CT, a negative study cannot exclude a DAVF, and catheter cerebral angiography is indicated. Intra-arterial catheter angiography remains the gold standard for imaging DAVF. Angiography will show early venous drainage and shunting of arterial blood into the venous system through the fistula (Figure 2). It is imperative that imaging be started in the early arterial phase and continue into the late venous phase to fully evaluate the fistula. A 6-vessel angiogram, including dedicated ICA and ECA imaging, is required to fully assess the DAVF. Eight percent of patients with a DAVF have been found to have multiple DAVFs on angiography.11 A dual arterial injection (competitive angiogram) may be required to study the venous drainage of the fistula and normal brain. In this technique, selective injection of the ICA is followed by injection of the ECA. With the use of resubtraction and changing of the mask runs, drainage of the fistula and normal brain parenchyma can be superimposed.12 The nidus of the fistula is the area of convergence of all the feeding arteries and the origin of the draining veins. The nidus is

S34 | VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT

FIGURE 2. Catheter angiography, lateral projection, of the right external carotid artery injection showing the middle meningeal artery arterial feeders (black arrow) and early venous drainage (white arrow) with the fistula sac in between.

often best visualized with superselective injections of distal feeding arteries (Figure 3). It is paramount that the venous drainage pattern be thoroughly evaluated to determine the natural history of the fistula and appropriate treatment planning. Factors to evaluate include the presence of cortical venous reflux, venous sinus circulation or occlusion, flow pattern in the venous sinuses, and any collateral venous drainage pattern. The presence of tortuous, engorged veins during the venous phase is called a pseudophlebitic pattern and is a sign of venous congestion. A pseudophlebitic pattern was seen in 81% of DAVFs with cortical venous reflux and only 8% of DAVFs with drainage into the dural sinus only.13 Other subtle findings during the venous phase can include drainage via the pial or medullary collateral veins, focal areas of delayed circulation, and venous shunting to the orbit or transosseous veins.

FIGURE 3. Superselective catheter angiography of the right middle meningeal artery, lateral view, showing arterial feeder into the fistula sac.

www.neurosurgery-online.com

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

SURGICAL APPROACHES TO DURAL FISTULAS

TREATMENT OPTIONS Borden type I DAVFs are associated with a benign natural history and usually do not require treatment unless the associated symptoms such as headache, bruit or ocular manifestations are intolerable. Treatment options include endovascular, surgical, and radiosurgical options. The goal of all therapeutic options is complete disconnection of the fistula from the venous drainage. With some very specific exceptions, this can be accomplished for many DAVFs by endovascular therapy alone, transarterial, transvenous, or both. Surgical interruption of the fistula is an excellent option for fistulas

FIGURE 4. A, catheter angiography, lateral view, of a dural arteriovenous fistula (DAVF) showing the middle meningeal artery feeder into the fistula sac (black arrow). B, catheter angiography, anteroposterior view, of the DAVF showing the middle meningeal artery feeder into the fistula sac (black arrow).

NEUROSURGERY

FIGURE 5. A, catheter angiography, native image, lateral view, showing Onyx embolization (black arrow) of a dural arteriovenous fistula (DAVF) through microcatheterization of the middle meningeal artery feeding branch (white arrow). B, catheter angiography lateral view showing full cure of the DAVF with transarterial Onyx cast at the DAVF sac (black arrow) and no residual filling.

VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT | S35

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

YOUSSEF ET AL

FIGURE 6. A, superselective catheter angiography, lateral view, showing the arterial feeding branches into the fistula sac of the superior sagittal sinus (black arrow) with early venous drainage. B, catheter angiography lateral view showing embolization of the dural arteriovenous fistula with Onyx (black arrow). C, catheter angiography, lateral view, showing partial Onyx embolization (white arrow) with residual filling of the fistula sac (black arrow).

with complex venous drainage not readily navigated by transvenous catheterization. Radiosurgery is rarely required for treatment. Observation Borden type I DAVFs (those without cortical venous reflux) are rarely associated with aggressive symptoms such as hemorrhage or neurological deficit and may be observed. Observation is especially appropriate if the patients are asymptomatic or the benign symptoms are tolerated well. Borden type I DAVFs have a 2% to 3% chance of progressing and developing cortical venous

FIGURE 7. Intraoperative catheter angiography, lateral view, showing Onyx cast (white arrow) as previously shown in Figure 5 and no residual filling of fistula sac after surgical resection.

S36 | VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT

reflux11 and should be followed up clinically and radiologically. Angiography should be performed if the patient has a change of symptoms. Even an improvement in symptoms such as spontaneous resolution of a subjective bruit should be investigated because it may be due to progressive sinus occlusion and development of leptomeningeal venous drainage.14 In patients with stable symptoms, we perform a baseline and periodic follow-up time-of-flight/ time-resolved MR angiography. Transarterial Embolization To use endovascular techniques to cure a DAVF, the fistula sac itself must be occluded. Feeding artery embolization alone will rarely cure a DAVF. A microcatheter must be navigated and wedged to infiltrate the fistula sac with superselective embolization performed to successfully achieve cure on the arterial side. Ideally, the cyanoacrylate glue is pushed through the microcatheter and into the fistula sac and the most proximal portion of the venous drainage system. A liquid embolic agent with adjustable viscosity is best to control the rate of flow of the glue to prevent venous embolization (Figures 4 and 5). Particulate embolization of ECA feeders is relatively safe and easy. However, it is typically not curative and is generally used for symptomatic relief and in combination with other procedures such as radiosurgery, surgery, or transvenous embolization. N-butyl-2-cyanoacrylate or Onyx (eV3, Irvine, Ca) embolization allows the deposition of glue deep into the fistula but requires good positioning of the microcatheter and slow, controlled injection. Onyx is nonadherent to the vessel wall and reduces the risk of gluing the microcatheter to the vessel. It is available in 2 concentrations with variable viscosity to allow greater control and penetration of the fistula as required. Another useful property of Onyx is its black color, which allows easier visualization of the fistula intraoperatively if only partial cure is accomplished through the endovascular route.

www.neurosurgery-online.com

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

SURGICAL APPROACHES TO DURAL FISTULAS

FIGURE 8. A, noncontrast computed tomographic head axial imaging showing intracranial hemorrhage later found to be associated with a dural arteriovenous fistula (DAVF). B, catheter angiography, lateral view, showing the DAVF sac (black arrow) with multiple small feeding arteries arising from the ophthalmic artery (white arrow) and a single large draining vein into the superior sagittal sinus. This DAVF is not amenable to endovascular embolization because of the proximity of feeding arteries to critical neural structures. C, catheter angiography, oblique view, showing the DAVF sac (black arrow) with multiple small feeding arteries arising from the ophthalmic artery and a single large draining vein into superior sagittal sinus (white arrow). D, intraoperative catheter angiography, lateral view, showing surgical cure of the DAVF with no residual filling.

Frequently, transarterial embolization is unsuccessful as a result of limitations in the properties of liquid embolic agents and the multiple small feeding arteries that supply the fistula. In such cases, the partial embolization may decrease vascularity for later surgical resection (Figures 6 and 7). If left uncured, the DAVF will continue to recruit arterial feeders over time, leading to a recurrence. There is also a risk that a partial embolization will convert a benign DAVF into a more aggressive DAVF as its flow dynamics are altered over time. When used as an adjunct to surgical treatment, transarterial embolization (even partial) can be of great help by decreasing blood flow to the fistula and allowing easier identification

NEUROSURGERY

intraoperatively.15 Preoperative embolization of large feeding arteries carries a relatively low risk (especially in the ECA feeders).11 However, there are risks to consider such as the inadvertent embolization of the important anastomoses between the ICA and ECA systems and the vertebrobasilar circulation. There is a risk of intra-arterial embolization of the agent that can lead to stroke. If the agent were to embolize through the venous system, there would be a risk of pulmonary embolus. Transvenous Embolization Transvenous embolization with liquid embolic agents or microcoils requires a detailed knowledge of the flow dynamics of

VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT | S37

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

YOUSSEF ET AL

the patient’s venous sinuses. Oftentimes, embolization of the fistula alone is not possible, and preparation should be made to ensure that sacrifice of the involved sinus will not create greater complications for the patient. This is not advisable if the patient does not have alternative drainage pathways for the brain.16 Complications of transvenous embolization may arise from venous sinus sacrifice. These can include venous hypertension and infarction. In cases of a carotid cavernous fistula, the ocular symptoms may actually worsen transiently after treatment. Radiosurgery Stereotactic radiosurgery has been used primarily as an adjunct in the treatment of DAVFs.17 Primary treatment of all aggressive

FIGURE 9. A, 3-dimensional computer-aided reconstruction of catheter angiography, lateral view, showing a complex dural arteriovenous fistula (DAVF) of the lateral sinus not amenable to transarterial embolization because of multiple small feeding arteries. White arrow depicts surgical disconnection site. B, intraoperative catheter angiography, lateral view, showing complete cure with no residual filling of the DVAF sac after surgical resection.

S38 | VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT

FIGURE 10. Lateral left carotid early (A) and late (B) angiogram demonstrating lateral sinus dural arteriovenous fistula with leptomeningeal venous drainage. Black arrow points to the arterial feeder at the fistula sac.

www.neurosurgery-online.com

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

SURGICAL APPROACHES TO DURAL FISTULAS

FIGURE 11. A, intraoperative photograph shows arterialized leptomeningeal draining vein (black arrow) as seen on angiogram. B, indocyanine green videoangiogram shows the arterialized vein (white arrow) under fluorescence. C, intraoperative photograph after disconnection of all draining veins from the fistula.

DAVF types should begin with endovascular or microsurgical treatment. The aggressive natural history of Borden type II and III lesions makes them unsuitable for a treatment that has the lengthy period to therapeutic benefit associated with stereotactic radiosurgery. Stereotactic radiosurgery is most often used in nonaggressive (Borden type I) DAVFs to aid in refractory symptomatology such as pulsatile tinnitus. Surgery Early surgical treatment strategies involved complete surgical excision of the fistula, which has been largely abandoned with the recognition that simply disconnecting the venous drainage from the fistula is curative.18 Earlier approaches were quite extensive and involved coagulation of the arterial feeders, excision of the involved dural leaflet, resection of arterialized leptomeningeal veins, and possibly sacrifice of the involved dural venous sinus. These procedures tend to be extensive and can be associated with significant morbidity and mortality.19 Complications can include significant blood loss at any point in the procedure, including from engorged vessels of the scalp or transosseous and dural vessels on exposure. Transosseous bleeding can be controlled with bone wax, and dural bleeding can be controlled with coagulation or hemostatic clips. Surgery is occasionally used to obtain venous access to the fistula for packing and embolization.20 This has been performed in patients with anatomy unfavorable for catheter-based transvenous embolization as a result of thrombosis or stenosis of the involved dural venous sinus. In this case, the involved sinus can be packed with Gelfoam or sacrificed with coagulation and suture. This strategy has largely fallen out of favor because of the potential complications associated with dural venous sacrifice previously discussed. The current surgical strategy involves disconnecting the venous drainage from the fistula. This strategy is effective in reducing the most dangerous risks of the fistula, including intracerebral hemorrhage and neurological deficit.18,21-24 Disconnection of the cortical venous drainage is a simpler procedure than full

NEUROSURGERY

excision of the fistula and carries a much lower risk of morbidity. The procedure involves exposure of the fistula and coagulation of any arterial feeders encountered. The arterialized veins are exposed and coagulated as close to the fistula as possible (Figure 8). To prevent massive blood loss, careful coagulation must ensure the vein is fully coagulated. Nonarterialized veins should be preserved (Figure 9). Intraoperative angiography and, more recently, indocyanine green videoangiography have been shown to be highly beneficial in documenting complete obliteration of the fistula before closure.25 This surgical strategy has been found to be quite effective in the treatment of selected fistulas.18,26 In a recent series from Brigham and Women’s and Children’s Hospital and Harvard Medical School, 35 of 70 DAVFs (50%) presenting to the institution were Borden type III.27 Twenty-four (69%) were treated surgically, with an angiographic obliteration rate of 96%. The combined permanent morbidity and mortality rate was 17%, and after a follow-up of 2.1 years, 13 (54%) improved postoperatively, 7 (29%) were the same, and 4 (17%) were worse. Of note, 13 (54%) were asymptomatic and 18 were independent with a modified Rankin Scale score of 0 to 2. The conclusion was that surgical treatment of DAVF with venous reflux is associated with a high angiographic cure rate and acceptable morbidity and mortality given the natural history of high-grade DAVFs.27 Classic methods for treating other cerebrovascular lesions are being repurposed as tools for DAVF. One such tool is the directional intraoperative Doppler ultrasound, which has long been used in surgery for aneurysms and cerebral arteriovenous malformations.28 In the study, 12 patients with DAVFs underwent microsurgical treatment, 12 were treated for acute hemorrhage and 8 for nonhemorrhagic symptoms. Three advantages of the directional intraoperative Doppler ultrasound were unequivocal identification of veins with cortical/deep venous reflux from the fistula, verification of completeness of occlusion of the fistula, and identification of dural arterial feeders not visualized under the microscope. The directional intraoperative Doppler ultrasound was found to be safe, effective, and reliable in microsurgical resection of DAVFs.28 Another tool that has been used rarely in

VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT | S39

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

YOUSSEF ET AL

surgeon.18,24 Combined endovascular treatment followed by microsurgical resection has improved the safety and efficacy of treatment, especially in relation to previously difficult-to-access and -treat DAVFs.15,30,31 Improvements in microsurgical and skull base approaches have allowed surgeons to approach and obliterate fistulas with little or no retraction on the brain.1 Image-guided systems have also allowed for better localization and smaller, more efficient approaches. Intraoperative CT angiography for lesion localization and craniotomy planning has been successfully used.32 Finally, the use of intraoperative angiography or indocyanine green videoangiography33 confirms the complete disconnection of fistula while patients are still on the operating room table, preventing reoperation for residual fistulas25 (Figures 10-12). Disclosure The authors have no personal financial or institutional interest in any of the drugs, materials, or devices described in this article.

REFERENCES

FIGURE 12. Early (A) and late (B) intraoperative angiograms documenting complete obliteration of the fistula.

DAVF microsurgery is deep hypothermia and circulatory arrest.29 This method has increased anesthesia risk vs classic surgical resection but was found to aid in resection of deep, difficult-toaccess lesions, including deep tentorial lesions. We have never found it necessary to use this for the management of a DAVF. Although many of the advances in DAVF treatment have occurred in the endovascular arena, open microsurgical advances in the past decade have primarily been in the tools available to the

S40 | VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT

1. Awad IA, Barrow DL. Conceptual overview and management strategies for dural arteriovenous malformations. In: Awad IA, Barrow DL, eds. Dural Arteriovenous Malformations. Park Ridge: American Association of Neurological Surgeons; 1993: 231-241. 2. Awad IA, Little JR, Akarawi WP, Ahl J. Intracranial dural arteriovenous malformations: factors predisposing to an aggressive neurological course. J Neurosurg. 1990;72(6):839-850. 3. Borden JA, Wu JK, Shucart WA. A proposed classification for spinal and cranial dural arteriovenous fistulous malformations and implications for treatment. J Neurosurg. 1995;82(2):166-179. 4. Gandhi D, Chen J, Pearl M, Huang J, Gemmete JJ, Kathuria S. Intracranial dural arteriovenous fistulas: classification, imaging findings, and treatment. AJNR Am J Neuroradiol. 2012;33(6):1007-1013. 5. Cognard C, Gobin YP, Pierot L, et al. Cerebral dural arteriovenous fistulas: clinical and angiographic correlation with a revised classification of venous drainage. Radiology. 1995;194(3):671-680. 6. Barrow DL, Spector RH, Braun IF, Landman JA, Tindall SC, Tindall GT. Classification and treatment of spontaneous carotid-cavernous sinus fistulas. J Neurosurg. 1985;62(2):248-256. 7. Davies MA, TerBrugge K, Willinsky R, Coyne T, Saleh J, Wallace MC. The validity of classification for the clinical presentation of intracranial dural arteriovenous fistulas. J Neurosurg. 1996;85(5):830-837. 8. Davies MA, Ter Brugge K, Willinsky R, Wallace MC. The natural history and management of intracranial dural arteriovenous fistulae, part 2: aggressive lesions. Interv Neuroradiol. 1997;3(4):303-311. 9. Davies MA, Saleh J, Ter Brugge K, Willinsky R, Wallace MC. The natural history and management of intracranial dural arteriovenous fistulae, part 1: benign lesions. Interv Neuroradiol. 1997;3(4):295-302. 10. Gross BA, Du R. The natural history of cerebral dural arteriovenous fistulae. Neurosurgery. 2012;71(3):594-602. 11. Schmidek HH, Roberts David W. Schmidek & Sweet surgical management of cranial dural arteriovenous fistulas. In: Javadpour M, Christopher Wallace M, eds. Operative Neurosurgical Techniques. Vol 2. Philadelphia, PA: Saunders Elsevier; 2006:1287-1305. 12. Gupta A, Periakaruppan A. Intracranial dural arteriovenous fistulas: a review. Indian J Radiol Imaging. 2009;19(1):43-48. 13. Willinsky R, Goyal M, terBrugge K, Montanera W. Tortuous, engorged pial veins in intracranial dural arteriovenous fistulas: correlations with presentation, location, and MR findings in 122 patients. AJNR Am J Neuroradiol. 1999;20(6):1031-1036. 14. Oh JT, Chung SY, Lanzino G, et al. Intracranial dural arteriovenous fistulas: clinical characteristics and management based on location and hemodynamics. J Cerebrovasc Endovasc Neurosurg. 2012;14(3):192-202. 15. Murayama Y, Arakawa H, Ishibashi T, et al. Combined surgical and endovascular treatment of complex cerebrovascular diseases in the hybrid operating room. J Neurointerv Surg. 2013;5(5):489-493.

www.neurosurgery-online.com

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

SURGICAL APPROACHES TO DURAL FISTULAS

16. Lasjaunias P, BA. Dural arteriovenous shunts. In: Surgical Neuroangiography. Vol 2. Berlin, Germany: Springer-Verlag; 2004:565-607. 17. Yang HC, Kano H, Kondziolka D, et al. Stereotactic radiosurgery with or without embolization for intracranial dural arteriovenous fistulas. Neurosurgery. 2010;67 (5):1276-1283; discussion 1284-1285. 18. Tomak PR, Cloft HJ, Kaga A, Cawley CM, Dion J, Barrow DL. Evolution of the management of tentorial dural arteriovenous malformations. Neurosurgery. 2003; 52(4):750-760; discussion 760-762. 19. Sundt TM Jr, Piepgras DG. The surgical approach to arteriovenous malformations of the lateral and sigmoid dural sinuses. J Neurosurg. 1983;59(1):32-39. 20. Wachter D, Hans F, Psychogios MN, Knauth M, Rohde V. Microsurgery can cure most intracranial dural arteriovenous fistulae of the sinus and non-sinus type. Neurosurg Rev. 2011;34(3):337-345; discussion 345. 21. Collice M, D’Aliberti G, Arena O, Solaini C, Fontana RA, Talamonti G. Surgical treatment of intracranial dural arteriovenous fistulae: role of venous drainage. Neurosurgery. 2000;47(1):56-66; discussion 66-67. 22. Thompson BG, Doppman JL, Oldfield EH. Treatment of cranial dural arteriovenous fistulae by interruption of leptomeningeal venous drainage. J Neurosurg. 1994;80(4):617-623. 23. van Dijk JM, Willinsky RA. Venous congestive encephalopathy related to cranial dural arteriovenous fistulas. Neuroimaging Clin N Am. 2003;13(1):55-72. 24. Lawton MT, Sanchez-Mejia RO, Pham D, Tan J, Halbach VV. Tentorial dural arteriovenous fistulae: operative strategies and microsurgical results for six types. Neurosurgery. 2008;62(3 suppl 1):110-124; discussion 124-125. 25. Schuette AJ, Cawley CM, Barrow DL. Indocyanine green videoangiography in the management of dural arteriovenous fistulae. Neurosurgery. 2010;67(3):658-662; discussion 662.

NEUROSURGERY

26. Pradilla G, Coon AL, Huang J, Tamargo RJ. Surgical treatment of cranial arteriovenous malformations and dural arteriovenous fistulas. Neurosurg Clin N Am. 2012;23(1):105-122. 27. Gross BA, Du R. Surgical treatment of high grade dural arteriovenous fistulae. J Clin Neurosci. 2013;20(11):1527-1532. 28. Eide PK, Sorteberg AG, Meling TR, Sorteberg W. Directional intraoperative Doppler ultrasonography during surgery on cranial dural arteriovenous fistulas [published online ahead of print July 8, 2013]. Neurosurgery. doi:10.1227/NEU. 0000000000000061. 29. Feng WF, Wang G, Zhang GZ, Li MZ, Qi ST. Surgical management of a complex intracranial dural ateriovenous fistula with deep hypothermia circulatory arrest: a case report and literature review [in Chinese]. Nan Fang Yi Ke Da Xue Xue Bao. 2011;31(10):1784-1788. 30. Spiotta AM, Sivapatham T, Hussain MS, Hui FK, Moskowitz SI, Gupta R. Combined surgical and endovascular approach to a complex dural arteriovenous fistula involving the superior sagittal sinus and torcula. J Stroke Cerebrovasc Dis. 2012;21(4):283-288. 31. Hallaert GG, De Keukeleire KM, Vanhauwaert DJ, Defreyne L, Van Roost D. Intracranial dural arteriovenous fistula successfully treated by combined openendovascular procedure. J Neurol Neurosurg Psychiatry. 2010;81(6):685-689. 32. Raza SM, Papadimitriou K, Gandhi D, Radvany M, Olivi A, Huang J. Intraarterial intraoperative computed tomography angiography guided navigation: a new technique for localization of vascular pathology. Neurosurgery. 2012;71 (2 suppl operative):ons240-ons252; discussion ons252. 33. Kato N, Tanaka T, Suzuki Y, et al. Multistage indocyanine green videoangiography for the convexity dural arteriovenous fistula with angiographically occult pial fistula. J Stroke Cerebrovasc Dis. 2012;21(8):918.e1-918.e5.

VOLUME 74 | NUMBER 2 | FEBRUARY 2014 SUPPLEMENT | S41

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.

Advances in surgical approaches to dural fistulas.

Dural arteriovenous fistulas are abnormal connections of dural arteries to dural veins or venous sinuses originating from within the dural leaflets. T...
1MB Sizes 0 Downloads 0 Views