Clinical Neurology and Neurosurgery 128 (2015) 123–129
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Diagnosis and management of dural arteriovenous fistulas: A 10 years single-center experience F. Signorelli, G.M. Della Pepa ∗ , G. Sabatino, E. Marchese, G. Maira, A. Puca, A. Albanese Institute of Neurosurgery, Catholic University of Rome, Rome, Italy
a r t i c l e
i n f o
Article history: Received 29 September 2014 Received in revised form 1 November 2014 Accepted 16 November 2014 Available online 24 November 2014 Keywords: Dural arteriovenous fistulas DAVF Intracranial dural arteriovenous fistula Hemorrhage Cortical venous drainage Surgery Embolization
a b s t r a c t Objectives: Dural arteriovenous fistulas (DAVFs) are a challenging condition in vascular neurosurgery. Disease natural history and its management is still debated. In the present paper we report our center series on DAVFs over a period of 10 years. Our data were compared with relevant literature. Patient and methods: Our series includes 45 cases: 14 cavernous sinus, 11 transverse-sigmoid, 8 patients tentorial, 6 anterior cranial fossa, 5 patients spinal, 1 patient foramen magnum. Results and conclusions: DVAFs distribution, clinical presentation and hemorrhagic risk are discussed. Cavernous sinus DAVFs are the most common site in our series. Other locations in order of frequency are transverse-sigmoid sinus, tentorial, anterior cranial fossa, spinal and foramen magnum. The majority of patients presented with non-aggressive symptoms. 18% presented with intracranial hemorrhage: all the hemorrhages occurred in high-grade DAVFs. For most patients, endovascular treatment, transarterial or transvenous, was the first option. Surgery was performed for the anterior cranial fossa DAVFs and other complex lesions draining mostly transverse-sigmoid sinus and tentorium. In 7% of cases a combination of endovascular + surgical treatment was used. Our series has been carefully analyzed in comparison ‘side by side’ with most relevant literature on DVAFs, focusing particularly on management strategies, therapeutic options and risks related to treatment. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Dural arteriovenous fistulas (DAVFs) are pathologic shunts between dural arteries and dural venous sinuses, meningeal veins or cortical veins [1,2]. They account for approximately 10–15% of intracranial vascular malformations and are more frequent among middle-aged and older patients, though children can be affected [1,2]. Venous hypertension, neoangiogenesis and risk factors for venous thrombosis predisposes to the development of DAVFs in adulthood. The more recent and most commonly used classification scheme are those developed by Borden et al. and Cognard et al. The Borden classification system has three subtypes and stratifies DAVFs on the basis of the site of venous drainage, the presence of cortical venous reflux (CVR) and number of fistula (single-hole or multiple-hole fistulas) [3]. The advantage of the Borden classification lies in its simplicity without loss of predictability [4]. Cognard classification is based on the direction of dural sinus drainage,
∗ Corresponding author at: Institute of Neurosurgery, Catholic University of Rome, Largo A. Gemelli 8, 00168 Rome, Italy. Tel.: +39 0630154120; fax: +39 063051343. E-mail address: [email protected] (G.M. Della Pepa). http://dx.doi.org/10.1016/j.clineuro.2014.11.011 0303-8467/© 2014 Elsevier B.V. All rights reserved.
the presence or absence of CVR and the venous outflow architecture (nonectatic cortical veins, ectatic cortical veins, or spinal perimedullary veins) [5]. Cognard type I lesions drain into the dural sinus with an anterograde flow direction and lack CVR. Type IIa lesions drain retrogradely into a dural sinus without CVR; type IIb lesions drain antegradely into a dural sinus with CVR; type IIa + b lesions drain retrogradely into a dural sinus with CVR. Type III drain directly in cortical veins without dural venous drainage. Types IV and V also lack dural venous drainage, have CVR and varying cortical venous outflow architecture (venous ectasias and spinal perimedullary drainage respectively in types IV and V). This classification system permits an accurate comparison of clinical and radiological parameters. Indeed lack of CVR (Cognard types I and IIa) is associated with a benign natural history with an extremely low risk of intracranial hemorrhage. Conversely the presence of CVR (Cognard type IIb-V) is an aggressive feature and is associated with a high risk of hemorrhage. An annual mortality rate of 10.4%, an annual risk of intracranial hemorrhage of 8.1% and annual risk of non-hemorrhagic neurologic deficit (seizures, Parkinsonism, cerebellar symptoms, failure to thrive and cranial nerve deficits) of 6.9% have been reported [5,6]. These lesions could have a dynamic course, so the risk stratification is not rigorous. The persistence of arterialized blood flow in
124
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
the involved dural sinus can lead over time to mechanical obstruction of the sinus and result in retrograde drainage of blood away from the sinus and into the cortical vein, predisposing to intracranial hemorrhage. Thus type I lesions can develop CVR with time. The risk of conversion is low, having only been reported in 2% of low-grade lesions [7]. Conversely, cases of spontaneous thrombosis/resolution of DAVFs have also been reported [8,9]. Any change in patient’s symptoms can reflect a variation of the venous drainage pattern and requires further diagnostic examinations. Initial radiologic investigation includes CT and MR imaging. Noncontrast CT is limited to rule out intracranial hemorrhage and edema due to venous congestion. MR imaging can demonstrate dilated vessels, venous pouches, vascular enhancement and sign of venous hypertension. Conventional angiography remains the gold standard for detection and classification of DAVFs. In addition to diagnosis, it permits an anatomical and hemodynamic assessment of the fistula which has a crucial relevance for planning therapy. Therapeutic armamentarium for management of DAVFs includes conservative treatment, endovascular procedures, surgical treatment and radiation therapy. In the present paper we retrospectively analyze our center series, over 10 years, and discuss the pertinent literature.
2. Materials and methods Between January 2003 and December 2013, 45 patients with intracranial DAVFs were enrolled in this study. These patients included 28 males and 17 females, with a mean age of 58.6 years, ranging from 31 to 82 years. We reviewed hospital records and imaging studies during a follow-up period of 48 months, ranging from 4 to 96 months. Conventional angiography was performed in all patients with intracranial DAVFs. Patients were stratified into seven groups according with the Cognard classification and were categorize into six groups based on anatomic location: transverse-sigmoid sinus, cavernous sinus, tentorial, anterior cranial fossa, foramen magnum and spinal. Fistulas were localized to the tentorium if the draining veins emerged from the superior or inferior tentorial surface, the area of the tentorial incisura, including the galenic system, and the tentorial attachment. Both angiographic and clinical assessment of treatment outcomes was performed. Statistical analyses were performed using Fisher exact test. Statistical significance was set at a probably value (p value) less than 0.05. Endovascular embolization and surgical interruption, alone or in combination, were performed according to clinical setting, lesion location and venous drainage pattern.
Conservative treatment was indicated in patients harboring a low-grade fistula (Cognard types I and IIa) or in some higher-grade cases located in cavernous sinus. 3. Results According to the location of DAVFs, cavernous sinus DAVFs were the most common. 14 patients (31%) had cavernous sinus DAVFs; 11 patients (24%) had transverse-sigmoid sinus DAVFs; 8 patients (18%) had tentorial DAVFs; 6 patients (13%) had anterior cranial fossa DAVFs; 5 patients (11%) had spinal DAVFs and 1 patient (2%) had foramen magnum DAVF. A majority of patients presented with non-aggressive symptoms. 8 patients (18%) had intracranial hemorrhage: 3 patients with tentorial DAVF, 3 with anterior cranial fossa DAVF, 1 with transverse-sigmoid sinus DAVF and one with spinal DAVF (Table 1) All the hemorrhages occurred in high-grade DAVFs (Table 1) but the hemorrhage with respect to the lesion location was not significant for the transverse-sigmoid sinus (p = 0.65), for the tentorium (p = 0.14), anterior cranial fossa (p = 0.063) and spine (p = 0.56). A significative absence of hemorrhage was observed in cavernous sinus DAVFs (p = 0.04). For most patients, endovascular treatment, transarterial or transvenous, was the first option. Surgery was performed for the anterior cranial fossa DAVFs and other complex lesions draining mostly transverse-sigmoid sinus and tentorium. 39 patients (87%) underwent surgical treatment, endovascular embolization or a combination treatment. In 7 cases (15%) only surgical treatment was performed. 29 patients (64%) underwent only endovascular embolization, via transarterious or transvenous route, or both. 3 patients (7%) who had shown incomplete obliteration with a single treatment modality, were treated with combined therapy. Conservative treatment was performed in 5 patients (11%). 2 patients (4%) refused endovascular treatment, 1 with cavernous sinus DAVF (type IIa) and 1 with transverse-sigmoid sinus DAVF (type I). A summary of the anatomic locations and treatment strategies of the DAVFs is shown in Table 2. Complete occlusion demonstrated at follow-up by angiography was achieved in 19 patients (65%) treated with embolization alone, in 2 patients (67%) treated with embolization and surgery, and in 6 patients (87%) treated with surgery alone. In one case of tentorial DAVF, endovascular procedure was complicated by an intracranial hemorrhage. Then surgical intervention warranted hemorrhage removal and fistula exclusion. Another patient with type IV anterior cranial fossa DAVF and hemorrhage developed transient right hemiparesis 4 days after surgical intervention due to vasospasm. She gradually recovered and in fifteenth day after surgery was discharged with no neurological deficit.
Table 1 Symptoms of DAVFs. Symptom
Cavernous sinus
Sigmoid transverse
Tentorial
Anterior cranial fossa
Headache Ocular Cranial nerves Incidental Bruises/tinnitus Aphasia Seizure Dysarhtria Anosmia Syncope Motor deficits Sensitive deficits Sphinteric deficits Hemorrhage
2 (I, III) 13 (IIa, IIb, III) 7 (IIa, IIb, IIa + b)
5 (IIa + b, IV)
4 (III, IV)
3 (IV)
1 (I) 6 (I, IIa, III) 1 (IIa + b)
1 (IV)
1 (IV)
Spinal
Total
1 (V) 2 (V) 1 (V) 1 (V)
14 (I, IIa + b, III, IV) 13 (IIa, IIb, III) 7 (IIa, IIb, IIa + b) 3 (I, IV) 6 (I, IIa, III) 1 (IIa + b) 2 (III) 2 (III) 1 (III) 1 (IV) 1 (V) 2 (V) 1 (V) 8 (IIa + b, III, IV, V)
2 (III) 2 (III) 1 (III) 1 (IV)
1 (IIa + b)
3 (III, IV)
3 (III, IV)
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129 Table 2 Treatment strategies according with anatomic location of DAVFs. Cognard
Conservative
Endovascular
I IIa IIb IIa + b III
3 (2ST, 1FM) 2 (1ST, 1CS)
3 (2ST, 1 CS) 4 (4CS) 6 (5CS, 1T) 1 (1ST) 5 (1ST, 1CS, 2T, 1 ACF) 6 (2ST, 3T, 1ACF) 1 26
IV V Total
2 (2CS)
2 9
Surgery
1 (1ST) 1 (1ACF) 3 (3ACF) 2 7
Endo + surgery
1 (1T) 2 (1ST, 1T) 3
Total 6 6 6 2 9 11 5 45
ST: sigmoid-transverse; FM: foramen magnum; CS: cavernous sinus; T: tentorial; ACF: anterior cranial fossa.
Overall, complication rate of DAVFs was 4%. Patients who had incomplete obliteration underwent only clinical and angiographic follow-up. No hemorrhages were recorded in the group of patients who had incomplete obliteration or those who did not undergo treatment. 4. Discussion Since an acquired etiology for DAVFs was first proposed by Castaigne and Djindjian in the late 1970s [10], several etiologic hypotheses have been formulated over years. Some authors sustained that physiologic arteriovenous shunts between meningeal arteries and dural venous sinuses enlarge as a response to local venous hypertension, resulting in a pathologic shunt [11,12]. Others underlined the role of neoangiogenesis as a response to the decreased cerebral perfusion secondary to venous hypertension [11,13]. Condition associated with DAVFs includes also stenosis or occlusion of the draining dural sinuses, sinus thrombosis, prothrombin gene mutation, resistance to activated protein (APCR), mutation in factor V gene, presence of antiphospholipid antibodies [1,10,14,15]. Probably a predisposing hypercoagulability status leads to venous sinus thrombosis. The subsequent venous hypertension opens up the microvascular shunts within the dura recruiting pial supply from parenchimal vessels [16]. The involved dural sinus receives arterialized blood flow that can lead to mechanical obstruction of the sinus and result in retrograde drainage of blood into the cortical veins. In addition, as a response, dilatation of the cortical veins may occur, predisposing to intracranial hemorrhage [10]. This pathogenetic mechanism explains the above mentioned dynamic features of DAVFs. Patients with DAVFs typically present between the fifth and the sixth decade of life, with symptoms depending on the characteristics of the venous outflow and the anatomic location. Similarly to other series, in our study patients with DAVFs presented at a mean age of 58.6 years, and the majority of lesions were located in the cavernous and the transverse-sigmoid sinuses (Table 3). Several studies showed no gender preponderance [11], some other demonstrated a female preponderance. A female-to-male ratio of 2:1 is reported for cavernous and transverse-sigmoid sinuses [17]. Kirsch reported on a series of 141 patents with cavernous sinus DAVFs, 112 of whom (79%) were women [18]. In our series, conversely, there was a male preponderance (62%). 4.1. Transverse-sigmoid sinus DAVFs Pulsatile tinnitus was the most common symptom of transversesigmoid sinus DAVFs. It is an experiencing auditory sensation in the absence of external stimuli synchronous with the patient’s heartbeat [19] which results from increased blood flow through the dural venous sinuses close to the middle ear [2,5]. Transverse-sigmoid
125
sinus DAVFs are the most common cause of objective pulsatile tinnitus in patients with normal otoscopic exam [19]. In Kirsch series, this symptom was present in 121 of 150 patients harboring transverse-sigmoid sinus (81%), and in 63% as the only symptom [18]. In our series, pulsatile tinnitus was complained by 54% of patients, followed by headache (45%). Other authors reported symptoms are bruit, insomnia, visual decline, seizure and dementia [20,21]. Intracranial hemorrhage is the presenting symptom in 10% of cases [19,20], mainly occurring in those lesions with a retrograde leptomeningeal venous drainage. In our series intracranial hemorrhage was recorded in one case (9%) of high grade DAVF (IIa + b Cognard); the patient underwent endovascular treatment, had no neurological deficits and was discharged in good conditions. An example is shown in Fig. 1. Despite intracranial hemorrhage represents one of the most serious sequelae of intracranial DAVFs, the prognosis of these patients is significantly better than that of other vascular lesions such as ruptured aneurysms [22,23]. This is probably related to the bleeding site being venous (although arterialized) rather than from a direct arterial source [22,24]. Nonetheless there is evidence that rebleeding may occur early after presentation, so treatment should be carried out soon after the diagnosis. Spontaneous regression of transverse-sigmoid sinus DAVFs is rare and usually occurs following a hemorrhage [21]. Brasse et al. documented the repeat spontaneous closure and reopening of a DAVF [25]. We based our therapeutic strategy on the pattern of venous drainage, since it determines the natural history of the DAVF. All cases of high grade transverse-sigmoid sinus DAVFs underwent treatment and three cases of low grade lesions had a conservative management. Low grade DAVFs, which drain directly into a dural venous sinus with normal anterograde flow, underwent treatment when the symptoms, especially tinnitus and bruits, were not well tolerated by the patient. In his series of 95 patients, Oh found a correlation between the transverse sigmoid sinus location and intracranial hemorrhage, besides the venous drainage pattern. Therefore, due to the low rate of spontaneous regression and the relatively high grade of aggressive symptoms, he suggests that all transverse-sigmoid sinus DAVFs require treatment [2]. Nonetheless the treatment-associated risk has to be weighed against the risk of the disease itself [26], which is very low in lowgrade DAVFs. Lv suggests transarterial embolization alone in patients with a low grade DAVF without risk of intracranial hemorrhage because it carries a low complication rate [21]. This was the predominant endovascular procedure in the beginning of the 1990s, when the experience with intracranial coil was limited [18]. Kirsch, in his series of 150 transverse-sigmoid DAVFs managed endovascularly, treated 33 patients with transarterial embolization alone, the majority of whom harboring a Cognard I DAVF and complaining non-aggressive neurological deficits [18]. Transarterial embolization of transverse-sigmoid sinus alone is also effective in palliating disabling symptoms even without completely curing the DAVF and is also indicated in the case of incomplete interruption of the arteriovenous shunt after transvenous coil occlusion or as adjunctive therapy for surgical treatment or radiosurgery in complex DAVFs that are not accessible by percutaneous transvenous catheterization [17]. Frequently, DAVFs involve a multitude of feeders, which often arise as multiple, small tributaries from major cerebral arteries that are not amenable to transarterial embolization alone. DAVFs that are partially treated with transarterial embolization alone may later recur and result in hemorrhage [17]. Moreover transarterial embolization with Onyx carries the risk of migration of the glue from the sinus to the arterialized draining
126
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
Table 3 Literature review. Author
Patients
Sex
Emorrhage
Cavernous Sinus
Transverse sigmoid
ACF
Oh et al. Kim et al. Davies et al. Piippo et al. (ref) Kirsch et al.
95 53 98 237 291
44 M 51 F 20 M 33 F 53 M 45 F 90 M 127 F 100 M 191 F
15 (16%) 15 (28%) 16 (16%) 33 (15%) 0 (CS) 15 (TS 10%) Nr Nr Nr 8 (18%)
15 (16%) 34 (64%) 28 (29%) 23 (10%) 141 (48%)
32 (34%) 11 (22%) 43 (44%) 153 (67%) 150 (52%)
2 (2%)
3 (3%)
4 (2%)
14 (14%) 6 (3%)
Chung et al. Urtasun et al. Klisch et al. Present series
60 24 31 45
Nr Nr Nr 28 M 17 F
34 (57%) 17 (55%) 14 (31%)
Nr 20 (83%) 14 (45%) 11 (24%)
Tentorial
Foramen magnum
SSS 14 (13.5%) 8 (14%)
Other 24 (25%) 13 (13%)
4 (2%)
6 (3%)
4 (17%) 6 (13%)
8 (18%)
1 (2%)
5 (11%)
ACF: anterior cranial fossa; SSS: superior sagittal sinus.
veins, distal venous occlusion and consequent venous infarction or hemorrhage. Growing experience with tranvenous procedures made it the primary therapy and transarterial embolization as an additional option for residual arterial feeders only [18]. The management of high grade DAVFs, where there is reflux through the sinus or reflux straight into a cortical vein, is more complex. Some authors consider arterial embolization followed by transvenous disconnection of the involved dural sinus the most efficient strategy for high grade DAVFs [21].
In his series of 24 patients who underwent percutaneous transvenous embolization, 20 of whom harboring transverse sigmoid DAVF, Urtasun obtained a complete occlusion in 17, while 20 patients were clinically cured [27]. The goals of surgical treatment of DAVFs include ligation of the point of fistula [17]. Because partial excision/obliteration of DAVFs does not offer protection against future hemorrhage risk, but rather increases it, a complete resection is of paramount importance. Recently, intraoperative angiography has been advocated as a useful tool for assessing the adequacy of surgical treatment
Fig. 1. A complex DVAF type 2a + b 8 (according to Cognard classification) of the left transverse-sigmoid sinus. The feeding arteries originate from the following branches: multiple petrosal branches of left middle meningeal artery, dural branches of the occipital artery, the hypertrophic ascending pharyngeal artery, the posterior meningeal artery (from the origin to the left vertebral artery), some pious-dural branches of the left posterior cerebral artery, dural branches from right occipital artery. The early venous drainage in the dural sinuses is associated with retrograde reflux into the superior sagittal sinus and in the deep veins, though the rectus sinus, the ampulla of Galen and basal veins.
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
not only of cerebral aneurysms and AVMs, but also DAVFs. In his series of 29 patients who underwent surgery for intracranial davfs, Pandey found a higher incidence of residual fistula sin patients with cavernous, tenurial, and SSS fistulas and a lower incidence for transverse-sigmoid DAVFs [28]. We performed surgical treatment in two cases of high grade DAVF when transarterial and transvenous embolization failed or were hazardous. 4.2. Cavernous sinus DAVFs In our study, the most common symptoms of cavernous sinus DAVFs were ocular symptoms, exophtalmos and headache. Reversed blood flow in the superior ophthalmic vein results in various ophthalmic symptoms such as proptosis, chemosis, arterialized conjunctival veins, ocular paresis, retroorbital pain, elevated intraocular pressure and diminished visual acuity, which can at least temporarily aggravate by partial or complete spontaneous thrombosis of draining veins. On the other hand, cavernous sinus DAVFs may cause serious morbidity including blindness, seizures, stroke and intracranial hemorrhage, when associated with CVR [18]. An example is shown in Fig. 2. In our series, aggressive neurologic symptoms were rare and there was no occurrence of intracranial hemorrhage. Interestingly, we observed no hemorrhage in this population of patients despite
127
57% of them (8 of 14) harbored high-grade DAVFs (Cognard types IIb–III). Conversely, 26% of all patients harboring a high-grade DAVF had intracranial hemorrhage. Suh et al. already reported that, although a cavernous sinus DAVF has a high-grade angiographic finding, it can have a benign course because it has sufficient venous drainage routes [29]. The only 2 patients with high-grade DAVF who underwent conservative management harbored a cavernous sinus lesion, and both cases displayed no clinical neither angiographic evolutions at 56 months follow-up. In a series of 141 patients harboring cavernous sinus DAVFs [18], of which 34% with CVR, no patient presented with intracranial hemorrhage. Interestingly, a single endovascular procedure was performed in 87% of patients and a complete occlusion of the fistula was achieved in 81% of cases. Current treatment is primarily endovascular and the transvenous route is preferred to the transarterial route, due to higher clinical and anatomical cure rates and a lower incidence of complications. The transvenous approach may be technically challenging but offers a high success rate with a relatively low likelihood of serious complications compared with transarterial embolization [18,30,31]. When petrosal sinus retrograde catheterization proves impossible, the cavernous sinus may still be approached through the superior opthalmic vein, the superior orbital fissure, the pterygoid
Fig. 2. A cavernous sinus fistula fed by multiple thin dural branches of the ascending pharyngeal artery and thin dural meningo-pituitary branches of right carotid artery. Left inferior petrosal sinus is occluded, so the fistula from the cavernous sinus drains exclusively in the omolateral superior ophthalmic vein, which appears ectasic, with reverse flow and drainage into angular and facial veins. No intracranial pial reflux is evident.
128
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
plexus, the facial vein or the superficial temporal vein [19]. As a last resort, a surgical transphenoidal approach may allow access to the cavernous sinus, or similarly a subtemporal approach may allow exposure to the superficial middle cerebral vein [19]. In his series of 17 patients harboring cavernous sinus DAVF, Klisch performed different transvenous approaches obtaining the best results via the superior opthalmic vein approach which was associated with a rate of complete fistula occlusion of 80% [32]. Endovascular treatment seems to be suitable also for highgrade/symptomatic cavernous sinus DAVF even if a complete embolization is not achievable or too risky, with angiographic follow-up of the residual fistula. Considering the extremely low risk of hemorrhage, the treatment purpose of cavernous sinus DAVFs should be considered the clinical improvement rather than the angiographic cure. Nonetheless, for the same reasons, a conservative management should be reasonable not only in asymptomatic cavernous sinus DAVFs but also in low grade mildly symptomatic DAVFs with symptoms not greatly impacting the quality of life [19] and some high grade lesions when the endovascular procedure is too hazardous. 4.3. Tentorial and anterior cranial fossa DAVFs In our series tentorial and anterior cranial fossa were the most common location for DAVFs presenting with hemorrhage (37.5% and 37.5%).
The venous drainage od anterior cranial fossa DVAFs is always via cortical venous drainage through the frontal and olfactory veins. This constant presence of CVR explains the high incidence of aggressive clinical behavior for lesions in this location and aggressive treatment is thus warranted. Anterior cranial fossa DAVFs are more amenable to surgery because are supplied by the ophthalmic arteries, in which catheterization is difficult and dangerous [33,34]. Transarterial embolization is usually limited to ECA branches. Embolization of ophthalmic artery branches is usually avoided as it carries the risk of central retinal artery occlusion. Transvenous embolization has been described but is not possible for most lesions due to the lack of venous access. Surgery is associated with high cure rates and low complication rates and is the treatment of choice for most anterior fossa DAVFs. This is performed through a low unilateral frontal or bifrontal craniotomy in the supine position with minimal retraction of the frontal lobes followed by disconnection of the arterialized leptomeningeal veins which should be coagulated and divided. In this study four patients with an anterior cranial fossa DAVFs underwent successful surgery and achieved complete exclusion. An example is shown in Fig. 3. A patient with type IV anterior cranial fossa DAVF and hemorrhage developed transient right hemiparesis 4 days after surgical intervention due to vasospasm. Probably the vasospasm was a delayed response to the pre-existent hemorrhage, rather than
Fig. 3. Anterior cranial fossa dural fistula type 4 (according to Cognard classification) with “foot” of the fistula localized medially on the cribriform plate. The fistula is fed by multiple ethmoidal branches originating from the distal portion of both the internal maxillary artery and to a lesser extent by the distal branches of both ophthalmic arteries. The fronto-basal pial venous drainage, dilated and tortuous, soon bifurcates into a frontal branch that flows into the superior sagittal sinus and in a left one which has an aneurysmal dilatation. This venous drainage flows into the left cavernous sinus with secondary reflux into the contralateral cavernous sinus. In the right lower box the CT scan at admission shows a deep left frontal hemorrhage.
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
a true procedure complication. She gradually recovered and in 15th day after surgery was discharged with no neurological deficit. Tentorial DAVFs display very often a leptomeningeal venous drainage involving the basal vein of Rosenthal, the lateral mesencephalic veins and the vein of Galen. The presence of CVR explains the high incidence of intracranial hemorrhage and neurologic deficit with these lesions. Transarterial embolization of the arterial feeders is often the preferred strategy to palliatively treat tentorial DAVFs without CVR [35]. Surgical exposure of tentorial DAVFs may be performed using a posterior parasagittal craniotomy and interhemispheric approach followed by CVR disconnection. In one case of tentorial DAVF, endovascular procedure was complicated by an intracranial hemorrhage. Then surgical intervention warranted hemorrhage removal and fistula exclusion. There is evidence that rebleeding may occur early after presentation, so it is reasonable to recommend treatment soon after diagnosis [6]. Moreover, tentorial and anterior cranial fossa DAVFs are among the most frequently presenting with hemorrhage, in our as in other series. Therefore, due to the difficult endovascular access route, surgical treatment should be recommended for recently diagnosed tentorial DAVFs and for residual ones after incomplete endovascular embolization. 5. Conclusions DAVFs encompass a broad spectrum of symptoms and clinical manifestations depending on their location and hemodynamic features. Thus strategies for treatment should be decided according to the characteristics of the DAVFs, subjective symptoms of the patients and after careful analysis of risk rate of each procedure. Disclosure statement The authors declare no interest to disclose, they have any personal or institutional financial interest in drugs, materials, or devices described in this submission and that they did not receive any specific funding. The paper and any of its contents have been presented previously. References [1] Gandhi D, Chen J, Pearl M, Huang J, Gemmete JJ, Kathuria S. Intracranial dural arteriovenous fistulas: classification, imaging findings, and treatment. Am J Neuroradiol 2012;33:1007–13. [2] Oh JT, Chung SY, Lanzino G, Park KS, Kim SM, Park MS, et al. Intracranial dural arteriovenous fistulas: clinical characteristics and management based on location and hemodynamics. J Cerebrovasc Endovasc Neurosurg 2012;14: 192–202. [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;82:166–79. [4] 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;85:830–7. [5] Cognard C, Gobin YP, Pierot L, Bailly AL, Houdart E, Casasco A, et al. Cerebral dural arteriovenous fistulas: clinical and angiographic correlation with a revised classification of venous drainage. Radiology 1995;194:671–80. [6] van Dijk JM, terBrugge KG, Willinsky RA, Wallace MC. Clinical course of cranial dural arteriovenous fistulas with long-term persistent cortical venous reflux. Stroke 2002;33:1233–6. [7] Satomi J, van Dijk JM, terbrugge KG, Willinsky RA, Wallace MC. Benign cranial dural arteriovenous fistulas: outcome of conservative management based on the natural history of the lesion. J Neurosurg 2002;97:767–70.
129
[8] Saito A, Furuno Y, Nishimura S, Kamiyama H, Nishijima M. Spontaneous closure of transverse sinus dural arteriovenous fistula: case report. Neurol Med Chir (Tokyo) 2008;48:564–8. [9] Olutola PS, Eliam M, Molot M, Talalla A. Spontaneous regression of a dural arteriovenous malformation. Neurosurgery 1983;12:687–90. [10] Gupta A, Periakaruppan A. Intracranial dural arteriovenous fistulas: a review. Indian J Radiol Imaging 2009;19:43–8. [11] Chung SJ, Kim JS, Kim JC, Lee SK, Kwon SU, Lee MC, et al. Intracranial dural arteriovenous fistulas: analysis of 60 patients. Cerebrovasc Dis 2002;13:79–88. [12] Kerber CW, Newton TH. The macro and microvasculature of the dura mater. Neuroradiology 1973;6:175–9. [13] Kojima T, Miyachi S, Sahara Y, Nakai K, Okamoto T, Hattori K, et al. The relationship between venous hypertension and expression of vascular endothelial growth factor: hemodynamic and immunohistochemical examinations in a rat venous hypertension model. Surg Neurol 2007;68:277–84, discussion 284. [14] Fujita A, Kuwamura K, Saitoh M, Sakagami Y, Takaishi Y, Suzuki S, et al. Cerebral sinus thrombosis in a patient with protein S deficiency: a case report. No Shinkei Geka 1997;25:467–72. [15] Gerlach R, Yahya H, Rohde S, Bohm M, Berkefeld J, Scharrer I, et al. Increased incidence of thrombophilic abnormalities in patients with cranial dural arteriovenous fistulae. Neurol Res 2003;25:745–8. [16] Houser OW, Campbell JK, Campbell RJ, Sundt Jr TM. Arteriovenous malformation affecting the transverse dural venous sinus–an acquired lesion. Mayo Clin Proc 1979;54:651–61. [17] Ghobrial GM, Marchan E, Nair AK, Dumont AS, Tjoumakaris SI, Gonzalez LF, et al. Dural arteriovenous fistulas: a review of the literature and a presentation of a single institution’s experience. World Neurosurg 2013;80:94–102. [18] Kirsch M, Liebig T, Kuhne D, Henkes H. Endovascular management of dural arteriovenous fistulas of the transverse and sigmoid sinus in 150 patients. Neuroradiology 2009;51:477–83. [19] Hacein-Bey L, Konstas AA, Pile-Spellman J. Natural history, current concepts, classification, factors impacting endovascular therapy, and pathophysiology of cerebral and spinal dural arteriovenous fistulas. Clin Neurol Neurosurg 2014;121:64–75. [20] Bernstein R, Dowd CF, Gress DR. Rapidly reversible dementia. Lancet 2003;361:392. [21] Lv X, Jiang C, Li Y, Liu L, Liu J, Wu Z. Transverse-sigmoid sinus dural arteriovenous fistulae. World Neurosurg 2010;74:297–305. [22] Daniels DJ, Vellimana AK, Zipfel GJ, Lanzino G. Intracranial hemorrhage from dural arteriovenous fistulas: clinical features and outcome. Neurosurg Focus 2013;34:E15. [23] Choi JH, Mast H, Sciacca RR, Hartmann A, Khaw AV, Mohr JP, et al. Clinical outcome after first and recurrent hemorrhage in patients with untreated brain arteriovenous malformation. Stroke 2006;37:1243–7. [24] King WA, Martin NA. Intracerebral hemorrhage due to dural arteriovenous malformations and fistulae. Neurosurg Clin N Am 1992;3:577–90. [25] Brasse G, Roth HJ, Fritzsch D, Scheid R. Now you see it, now you don’t – the hidden life of cerebral dural arteriovenous fistulae. Neuroradiology 2009;51:131–3. [26] Sarma D, ter Brugge K. Management of intracranial dural arteriovenous shunts in adults. Eur J Radiol 2003;46:206–20. [27] Urtasun F, Biondi A, Casaco A, Houdart E, Caputo N, Aymard A, et al. Cerebral dural arteriovenous fistulas: percutaneous transvenous embolization. Radiology 1996;199:209–17. [28] Pandey P, Steinberg GK, Westbroek EM, Dodd R, Do HM, Marks MP. Intraoperative angiography for cranial dural arteriovenous fistula. Am J Neuroradiol 2011;32:1091–5. [29] Suh DC, Lee JH, Kim SJ, Chung SJ, Choi CG, Kim HJ, et al. New concept in cavernous sinus dural arteriovenous fistula: correlation with presenting symptom and venous drainage patterns. Stroke 2005;36:1134–9. [30] Halbach VV, Higashida RT, Hieshima GB, Hardin CW, Pribram H. Transvenous embolization of dural fistulas involving the cavernous sinus. AJNR 1989;10:377–83. [31] Piippo A, Laakso A, Seppa K, Rinne J, Jaaskelainen JE, Hernesniemi J, et al. Early and long-term excess mortality in 227 patients with intracranial dural arteriovenous fistulas. J Neurosurg 2013;119:164–71. [32] Klisch J, Huppertz HJ, Spetzger U, Hetzel A, Seeger W, Schumacher M. Transvenous treatment of carotid cavernous and dural arteriovenous fistulae: results for 31 patients and review of the literature. Neurosurgery 2003;53:836–56, discussion 856–837. [33] 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;47:56–66, discussion 66–57. [34] Lucas CP, Zabramski JM, Spetzler RF, Jacobowitz R. Treatment for intracranial dural arteriovenous malformations: a meta-analysis from the English language literature. Neurosurgery 1997;40:1119–30, discussion 1130–1112. [35] Javadpour M, Jain H, Wallace MC, Willinsky RA, ter Brugge KG, Tymianski M. Analysis of cost related to clinical and angiographic outcomes of aneurysm patients enrolled in the international subarachnoid aneurysm trial in a North American setting. Neurosurgery 2005;56:886–94, discussion 886–894.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Diagnosis and management of dural arteriovenous fistulas: A 10 years single-center experience F. Signorelli, G.M. Della Pepa ∗ , G. Sabatino, E. Marchese, G. Maira, A. Puca, A. Albanese Institute of Neurosurgery, Catholic University of Rome, Rome, Italy
a r t i c l e
i n f o
Article history: Received 29 September 2014 Received in revised form 1 November 2014 Accepted 16 November 2014 Available online 24 November 2014 Keywords: Dural arteriovenous fistulas DAVF Intracranial dural arteriovenous fistula Hemorrhage Cortical venous drainage Surgery Embolization
a b s t r a c t Objectives: Dural arteriovenous fistulas (DAVFs) are a challenging condition in vascular neurosurgery. Disease natural history and its management is still debated. In the present paper we report our center series on DAVFs over a period of 10 years. Our data were compared with relevant literature. Patient and methods: Our series includes 45 cases: 14 cavernous sinus, 11 transverse-sigmoid, 8 patients tentorial, 6 anterior cranial fossa, 5 patients spinal, 1 patient foramen magnum. Results and conclusions: DVAFs distribution, clinical presentation and hemorrhagic risk are discussed. Cavernous sinus DAVFs are the most common site in our series. Other locations in order of frequency are transverse-sigmoid sinus, tentorial, anterior cranial fossa, spinal and foramen magnum. The majority of patients presented with non-aggressive symptoms. 18% presented with intracranial hemorrhage: all the hemorrhages occurred in high-grade DAVFs. For most patients, endovascular treatment, transarterial or transvenous, was the first option. Surgery was performed for the anterior cranial fossa DAVFs and other complex lesions draining mostly transverse-sigmoid sinus and tentorium. In 7% of cases a combination of endovascular + surgical treatment was used. Our series has been carefully analyzed in comparison ‘side by side’ with most relevant literature on DVAFs, focusing particularly on management strategies, therapeutic options and risks related to treatment. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Dural arteriovenous fistulas (DAVFs) are pathologic shunts between dural arteries and dural venous sinuses, meningeal veins or cortical veins [1,2]. They account for approximately 10–15% of intracranial vascular malformations and are more frequent among middle-aged and older patients, though children can be affected [1,2]. Venous hypertension, neoangiogenesis and risk factors for venous thrombosis predisposes to the development of DAVFs in adulthood. The more recent and most commonly used classification scheme are those developed by Borden et al. and Cognard et al. The Borden classification system has three subtypes and stratifies DAVFs on the basis of the site of venous drainage, the presence of cortical venous reflux (CVR) and number of fistula (single-hole or multiple-hole fistulas) [3]. The advantage of the Borden classification lies in its simplicity without loss of predictability [4]. Cognard classification is based on the direction of dural sinus drainage,
∗ Corresponding author at: Institute of Neurosurgery, Catholic University of Rome, Largo A. Gemelli 8, 00168 Rome, Italy. Tel.: +39 0630154120; fax: +39 063051343. E-mail address: [email protected] (G.M. Della Pepa). http://dx.doi.org/10.1016/j.clineuro.2014.11.011 0303-8467/© 2014 Elsevier B.V. All rights reserved.
the presence or absence of CVR and the venous outflow architecture (nonectatic cortical veins, ectatic cortical veins, or spinal perimedullary veins) [5]. Cognard type I lesions drain into the dural sinus with an anterograde flow direction and lack CVR. Type IIa lesions drain retrogradely into a dural sinus without CVR; type IIb lesions drain antegradely into a dural sinus with CVR; type IIa + b lesions drain retrogradely into a dural sinus with CVR. Type III drain directly in cortical veins without dural venous drainage. Types IV and V also lack dural venous drainage, have CVR and varying cortical venous outflow architecture (venous ectasias and spinal perimedullary drainage respectively in types IV and V). This classification system permits an accurate comparison of clinical and radiological parameters. Indeed lack of CVR (Cognard types I and IIa) is associated with a benign natural history with an extremely low risk of intracranial hemorrhage. Conversely the presence of CVR (Cognard type IIb-V) is an aggressive feature and is associated with a high risk of hemorrhage. An annual mortality rate of 10.4%, an annual risk of intracranial hemorrhage of 8.1% and annual risk of non-hemorrhagic neurologic deficit (seizures, Parkinsonism, cerebellar symptoms, failure to thrive and cranial nerve deficits) of 6.9% have been reported [5,6]. These lesions could have a dynamic course, so the risk stratification is not rigorous. The persistence of arterialized blood flow in
124
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
the involved dural sinus can lead over time to mechanical obstruction of the sinus and result in retrograde drainage of blood away from the sinus and into the cortical vein, predisposing to intracranial hemorrhage. Thus type I lesions can develop CVR with time. The risk of conversion is low, having only been reported in 2% of low-grade lesions [7]. Conversely, cases of spontaneous thrombosis/resolution of DAVFs have also been reported [8,9]. Any change in patient’s symptoms can reflect a variation of the venous drainage pattern and requires further diagnostic examinations. Initial radiologic investigation includes CT and MR imaging. Noncontrast CT is limited to rule out intracranial hemorrhage and edema due to venous congestion. MR imaging can demonstrate dilated vessels, venous pouches, vascular enhancement and sign of venous hypertension. Conventional angiography remains the gold standard for detection and classification of DAVFs. In addition to diagnosis, it permits an anatomical and hemodynamic assessment of the fistula which has a crucial relevance for planning therapy. Therapeutic armamentarium for management of DAVFs includes conservative treatment, endovascular procedures, surgical treatment and radiation therapy. In the present paper we retrospectively analyze our center series, over 10 years, and discuss the pertinent literature.
2. Materials and methods Between January 2003 and December 2013, 45 patients with intracranial DAVFs were enrolled in this study. These patients included 28 males and 17 females, with a mean age of 58.6 years, ranging from 31 to 82 years. We reviewed hospital records and imaging studies during a follow-up period of 48 months, ranging from 4 to 96 months. Conventional angiography was performed in all patients with intracranial DAVFs. Patients were stratified into seven groups according with the Cognard classification and were categorize into six groups based on anatomic location: transverse-sigmoid sinus, cavernous sinus, tentorial, anterior cranial fossa, foramen magnum and spinal. Fistulas were localized to the tentorium if the draining veins emerged from the superior or inferior tentorial surface, the area of the tentorial incisura, including the galenic system, and the tentorial attachment. Both angiographic and clinical assessment of treatment outcomes was performed. Statistical analyses were performed using Fisher exact test. Statistical significance was set at a probably value (p value) less than 0.05. Endovascular embolization and surgical interruption, alone or in combination, were performed according to clinical setting, lesion location and venous drainage pattern.
Conservative treatment was indicated in patients harboring a low-grade fistula (Cognard types I and IIa) or in some higher-grade cases located in cavernous sinus. 3. Results According to the location of DAVFs, cavernous sinus DAVFs were the most common. 14 patients (31%) had cavernous sinus DAVFs; 11 patients (24%) had transverse-sigmoid sinus DAVFs; 8 patients (18%) had tentorial DAVFs; 6 patients (13%) had anterior cranial fossa DAVFs; 5 patients (11%) had spinal DAVFs and 1 patient (2%) had foramen magnum DAVF. A majority of patients presented with non-aggressive symptoms. 8 patients (18%) had intracranial hemorrhage: 3 patients with tentorial DAVF, 3 with anterior cranial fossa DAVF, 1 with transverse-sigmoid sinus DAVF and one with spinal DAVF (Table 1) All the hemorrhages occurred in high-grade DAVFs (Table 1) but the hemorrhage with respect to the lesion location was not significant for the transverse-sigmoid sinus (p = 0.65), for the tentorium (p = 0.14), anterior cranial fossa (p = 0.063) and spine (p = 0.56). A significative absence of hemorrhage was observed in cavernous sinus DAVFs (p = 0.04). For most patients, endovascular treatment, transarterial or transvenous, was the first option. Surgery was performed for the anterior cranial fossa DAVFs and other complex lesions draining mostly transverse-sigmoid sinus and tentorium. 39 patients (87%) underwent surgical treatment, endovascular embolization or a combination treatment. In 7 cases (15%) only surgical treatment was performed. 29 patients (64%) underwent only endovascular embolization, via transarterious or transvenous route, or both. 3 patients (7%) who had shown incomplete obliteration with a single treatment modality, were treated with combined therapy. Conservative treatment was performed in 5 patients (11%). 2 patients (4%) refused endovascular treatment, 1 with cavernous sinus DAVF (type IIa) and 1 with transverse-sigmoid sinus DAVF (type I). A summary of the anatomic locations and treatment strategies of the DAVFs is shown in Table 2. Complete occlusion demonstrated at follow-up by angiography was achieved in 19 patients (65%) treated with embolization alone, in 2 patients (67%) treated with embolization and surgery, and in 6 patients (87%) treated with surgery alone. In one case of tentorial DAVF, endovascular procedure was complicated by an intracranial hemorrhage. Then surgical intervention warranted hemorrhage removal and fistula exclusion. Another patient with type IV anterior cranial fossa DAVF and hemorrhage developed transient right hemiparesis 4 days after surgical intervention due to vasospasm. She gradually recovered and in fifteenth day after surgery was discharged with no neurological deficit.
Table 1 Symptoms of DAVFs. Symptom
Cavernous sinus
Sigmoid transverse
Tentorial
Anterior cranial fossa
Headache Ocular Cranial nerves Incidental Bruises/tinnitus Aphasia Seizure Dysarhtria Anosmia Syncope Motor deficits Sensitive deficits Sphinteric deficits Hemorrhage
2 (I, III) 13 (IIa, IIb, III) 7 (IIa, IIb, IIa + b)
5 (IIa + b, IV)
4 (III, IV)
3 (IV)
1 (I) 6 (I, IIa, III) 1 (IIa + b)
1 (IV)
1 (IV)
Spinal
Total
1 (V) 2 (V) 1 (V) 1 (V)
14 (I, IIa + b, III, IV) 13 (IIa, IIb, III) 7 (IIa, IIb, IIa + b) 3 (I, IV) 6 (I, IIa, III) 1 (IIa + b) 2 (III) 2 (III) 1 (III) 1 (IV) 1 (V) 2 (V) 1 (V) 8 (IIa + b, III, IV, V)
2 (III) 2 (III) 1 (III) 1 (IV)
1 (IIa + b)
3 (III, IV)
3 (III, IV)
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129 Table 2 Treatment strategies according with anatomic location of DAVFs. Cognard
Conservative
Endovascular
I IIa IIb IIa + b III
3 (2ST, 1FM) 2 (1ST, 1CS)
3 (2ST, 1 CS) 4 (4CS) 6 (5CS, 1T) 1 (1ST) 5 (1ST, 1CS, 2T, 1 ACF) 6 (2ST, 3T, 1ACF) 1 26
IV V Total
2 (2CS)
2 9
Surgery
1 (1ST) 1 (1ACF) 3 (3ACF) 2 7
Endo + surgery
1 (1T) 2 (1ST, 1T) 3
Total 6 6 6 2 9 11 5 45
ST: sigmoid-transverse; FM: foramen magnum; CS: cavernous sinus; T: tentorial; ACF: anterior cranial fossa.
Overall, complication rate of DAVFs was 4%. Patients who had incomplete obliteration underwent only clinical and angiographic follow-up. No hemorrhages were recorded in the group of patients who had incomplete obliteration or those who did not undergo treatment. 4. Discussion Since an acquired etiology for DAVFs was first proposed by Castaigne and Djindjian in the late 1970s [10], several etiologic hypotheses have been formulated over years. Some authors sustained that physiologic arteriovenous shunts between meningeal arteries and dural venous sinuses enlarge as a response to local venous hypertension, resulting in a pathologic shunt [11,12]. Others underlined the role of neoangiogenesis as a response to the decreased cerebral perfusion secondary to venous hypertension [11,13]. Condition associated with DAVFs includes also stenosis or occlusion of the draining dural sinuses, sinus thrombosis, prothrombin gene mutation, resistance to activated protein (APCR), mutation in factor V gene, presence of antiphospholipid antibodies [1,10,14,15]. Probably a predisposing hypercoagulability status leads to venous sinus thrombosis. The subsequent venous hypertension opens up the microvascular shunts within the dura recruiting pial supply from parenchimal vessels [16]. The involved dural sinus receives arterialized blood flow that can lead to mechanical obstruction of the sinus and result in retrograde drainage of blood into the cortical veins. In addition, as a response, dilatation of the cortical veins may occur, predisposing to intracranial hemorrhage [10]. This pathogenetic mechanism explains the above mentioned dynamic features of DAVFs. Patients with DAVFs typically present between the fifth and the sixth decade of life, with symptoms depending on the characteristics of the venous outflow and the anatomic location. Similarly to other series, in our study patients with DAVFs presented at a mean age of 58.6 years, and the majority of lesions were located in the cavernous and the transverse-sigmoid sinuses (Table 3). Several studies showed no gender preponderance [11], some other demonstrated a female preponderance. A female-to-male ratio of 2:1 is reported for cavernous and transverse-sigmoid sinuses [17]. Kirsch reported on a series of 141 patents with cavernous sinus DAVFs, 112 of whom (79%) were women [18]. In our series, conversely, there was a male preponderance (62%). 4.1. Transverse-sigmoid sinus DAVFs Pulsatile tinnitus was the most common symptom of transversesigmoid sinus DAVFs. It is an experiencing auditory sensation in the absence of external stimuli synchronous with the patient’s heartbeat [19] which results from increased blood flow through the dural venous sinuses close to the middle ear [2,5]. Transverse-sigmoid
125
sinus DAVFs are the most common cause of objective pulsatile tinnitus in patients with normal otoscopic exam [19]. In Kirsch series, this symptom was present in 121 of 150 patients harboring transverse-sigmoid sinus (81%), and in 63% as the only symptom [18]. In our series, pulsatile tinnitus was complained by 54% of patients, followed by headache (45%). Other authors reported symptoms are bruit, insomnia, visual decline, seizure and dementia [20,21]. Intracranial hemorrhage is the presenting symptom in 10% of cases [19,20], mainly occurring in those lesions with a retrograde leptomeningeal venous drainage. In our series intracranial hemorrhage was recorded in one case (9%) of high grade DAVF (IIa + b Cognard); the patient underwent endovascular treatment, had no neurological deficits and was discharged in good conditions. An example is shown in Fig. 1. Despite intracranial hemorrhage represents one of the most serious sequelae of intracranial DAVFs, the prognosis of these patients is significantly better than that of other vascular lesions such as ruptured aneurysms [22,23]. This is probably related to the bleeding site being venous (although arterialized) rather than from a direct arterial source [22,24]. Nonetheless there is evidence that rebleeding may occur early after presentation, so treatment should be carried out soon after the diagnosis. Spontaneous regression of transverse-sigmoid sinus DAVFs is rare and usually occurs following a hemorrhage [21]. Brasse et al. documented the repeat spontaneous closure and reopening of a DAVF [25]. We based our therapeutic strategy on the pattern of venous drainage, since it determines the natural history of the DAVF. All cases of high grade transverse-sigmoid sinus DAVFs underwent treatment and three cases of low grade lesions had a conservative management. Low grade DAVFs, which drain directly into a dural venous sinus with normal anterograde flow, underwent treatment when the symptoms, especially tinnitus and bruits, were not well tolerated by the patient. In his series of 95 patients, Oh found a correlation between the transverse sigmoid sinus location and intracranial hemorrhage, besides the venous drainage pattern. Therefore, due to the low rate of spontaneous regression and the relatively high grade of aggressive symptoms, he suggests that all transverse-sigmoid sinus DAVFs require treatment [2]. Nonetheless the treatment-associated risk has to be weighed against the risk of the disease itself [26], which is very low in lowgrade DAVFs. Lv suggests transarterial embolization alone in patients with a low grade DAVF without risk of intracranial hemorrhage because it carries a low complication rate [21]. This was the predominant endovascular procedure in the beginning of the 1990s, when the experience with intracranial coil was limited [18]. Kirsch, in his series of 150 transverse-sigmoid DAVFs managed endovascularly, treated 33 patients with transarterial embolization alone, the majority of whom harboring a Cognard I DAVF and complaining non-aggressive neurological deficits [18]. Transarterial embolization of transverse-sigmoid sinus alone is also effective in palliating disabling symptoms even without completely curing the DAVF and is also indicated in the case of incomplete interruption of the arteriovenous shunt after transvenous coil occlusion or as adjunctive therapy for surgical treatment or radiosurgery in complex DAVFs that are not accessible by percutaneous transvenous catheterization [17]. Frequently, DAVFs involve a multitude of feeders, which often arise as multiple, small tributaries from major cerebral arteries that are not amenable to transarterial embolization alone. DAVFs that are partially treated with transarterial embolization alone may later recur and result in hemorrhage [17]. Moreover transarterial embolization with Onyx carries the risk of migration of the glue from the sinus to the arterialized draining
126
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
Table 3 Literature review. Author
Patients
Sex
Emorrhage
Cavernous Sinus
Transverse sigmoid
ACF
Oh et al. Kim et al. Davies et al. Piippo et al. (ref) Kirsch et al.
95 53 98 237 291
44 M 51 F 20 M 33 F 53 M 45 F 90 M 127 F 100 M 191 F
15 (16%) 15 (28%) 16 (16%) 33 (15%) 0 (CS) 15 (TS 10%) Nr Nr Nr 8 (18%)
15 (16%) 34 (64%) 28 (29%) 23 (10%) 141 (48%)
32 (34%) 11 (22%) 43 (44%) 153 (67%) 150 (52%)
2 (2%)
3 (3%)
4 (2%)
14 (14%) 6 (3%)
Chung et al. Urtasun et al. Klisch et al. Present series
60 24 31 45
Nr Nr Nr 28 M 17 F
34 (57%) 17 (55%) 14 (31%)
Nr 20 (83%) 14 (45%) 11 (24%)
Tentorial
Foramen magnum
SSS 14 (13.5%) 8 (14%)
Other 24 (25%) 13 (13%)
4 (2%)
6 (3%)
4 (17%) 6 (13%)
8 (18%)
1 (2%)
5 (11%)
ACF: anterior cranial fossa; SSS: superior sagittal sinus.
veins, distal venous occlusion and consequent venous infarction or hemorrhage. Growing experience with tranvenous procedures made it the primary therapy and transarterial embolization as an additional option for residual arterial feeders only [18]. The management of high grade DAVFs, where there is reflux through the sinus or reflux straight into a cortical vein, is more complex. Some authors consider arterial embolization followed by transvenous disconnection of the involved dural sinus the most efficient strategy for high grade DAVFs [21].
In his series of 24 patients who underwent percutaneous transvenous embolization, 20 of whom harboring transverse sigmoid DAVF, Urtasun obtained a complete occlusion in 17, while 20 patients were clinically cured [27]. The goals of surgical treatment of DAVFs include ligation of the point of fistula [17]. Because partial excision/obliteration of DAVFs does not offer protection against future hemorrhage risk, but rather increases it, a complete resection is of paramount importance. Recently, intraoperative angiography has been advocated as a useful tool for assessing the adequacy of surgical treatment
Fig. 1. A complex DVAF type 2a + b 8 (according to Cognard classification) of the left transverse-sigmoid sinus. The feeding arteries originate from the following branches: multiple petrosal branches of left middle meningeal artery, dural branches of the occipital artery, the hypertrophic ascending pharyngeal artery, the posterior meningeal artery (from the origin to the left vertebral artery), some pious-dural branches of the left posterior cerebral artery, dural branches from right occipital artery. The early venous drainage in the dural sinuses is associated with retrograde reflux into the superior sagittal sinus and in the deep veins, though the rectus sinus, the ampulla of Galen and basal veins.
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
not only of cerebral aneurysms and AVMs, but also DAVFs. In his series of 29 patients who underwent surgery for intracranial davfs, Pandey found a higher incidence of residual fistula sin patients with cavernous, tenurial, and SSS fistulas and a lower incidence for transverse-sigmoid DAVFs [28]. We performed surgical treatment in two cases of high grade DAVF when transarterial and transvenous embolization failed or were hazardous. 4.2. Cavernous sinus DAVFs In our study, the most common symptoms of cavernous sinus DAVFs were ocular symptoms, exophtalmos and headache. Reversed blood flow in the superior ophthalmic vein results in various ophthalmic symptoms such as proptosis, chemosis, arterialized conjunctival veins, ocular paresis, retroorbital pain, elevated intraocular pressure and diminished visual acuity, which can at least temporarily aggravate by partial or complete spontaneous thrombosis of draining veins. On the other hand, cavernous sinus DAVFs may cause serious morbidity including blindness, seizures, stroke and intracranial hemorrhage, when associated with CVR [18]. An example is shown in Fig. 2. In our series, aggressive neurologic symptoms were rare and there was no occurrence of intracranial hemorrhage. Interestingly, we observed no hemorrhage in this population of patients despite
127
57% of them (8 of 14) harbored high-grade DAVFs (Cognard types IIb–III). Conversely, 26% of all patients harboring a high-grade DAVF had intracranial hemorrhage. Suh et al. already reported that, although a cavernous sinus DAVF has a high-grade angiographic finding, it can have a benign course because it has sufficient venous drainage routes [29]. The only 2 patients with high-grade DAVF who underwent conservative management harbored a cavernous sinus lesion, and both cases displayed no clinical neither angiographic evolutions at 56 months follow-up. In a series of 141 patients harboring cavernous sinus DAVFs [18], of which 34% with CVR, no patient presented with intracranial hemorrhage. Interestingly, a single endovascular procedure was performed in 87% of patients and a complete occlusion of the fistula was achieved in 81% of cases. Current treatment is primarily endovascular and the transvenous route is preferred to the transarterial route, due to higher clinical and anatomical cure rates and a lower incidence of complications. The transvenous approach may be technically challenging but offers a high success rate with a relatively low likelihood of serious complications compared with transarterial embolization [18,30,31]. When petrosal sinus retrograde catheterization proves impossible, the cavernous sinus may still be approached through the superior opthalmic vein, the superior orbital fissure, the pterygoid
Fig. 2. A cavernous sinus fistula fed by multiple thin dural branches of the ascending pharyngeal artery and thin dural meningo-pituitary branches of right carotid artery. Left inferior petrosal sinus is occluded, so the fistula from the cavernous sinus drains exclusively in the omolateral superior ophthalmic vein, which appears ectasic, with reverse flow and drainage into angular and facial veins. No intracranial pial reflux is evident.
128
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
plexus, the facial vein or the superficial temporal vein [19]. As a last resort, a surgical transphenoidal approach may allow access to the cavernous sinus, or similarly a subtemporal approach may allow exposure to the superficial middle cerebral vein [19]. In his series of 17 patients harboring cavernous sinus DAVF, Klisch performed different transvenous approaches obtaining the best results via the superior opthalmic vein approach which was associated with a rate of complete fistula occlusion of 80% [32]. Endovascular treatment seems to be suitable also for highgrade/symptomatic cavernous sinus DAVF even if a complete embolization is not achievable or too risky, with angiographic follow-up of the residual fistula. Considering the extremely low risk of hemorrhage, the treatment purpose of cavernous sinus DAVFs should be considered the clinical improvement rather than the angiographic cure. Nonetheless, for the same reasons, a conservative management should be reasonable not only in asymptomatic cavernous sinus DAVFs but also in low grade mildly symptomatic DAVFs with symptoms not greatly impacting the quality of life [19] and some high grade lesions when the endovascular procedure is too hazardous. 4.3. Tentorial and anterior cranial fossa DAVFs In our series tentorial and anterior cranial fossa were the most common location for DAVFs presenting with hemorrhage (37.5% and 37.5%).
The venous drainage od anterior cranial fossa DVAFs is always via cortical venous drainage through the frontal and olfactory veins. This constant presence of CVR explains the high incidence of aggressive clinical behavior for lesions in this location and aggressive treatment is thus warranted. Anterior cranial fossa DAVFs are more amenable to surgery because are supplied by the ophthalmic arteries, in which catheterization is difficult and dangerous [33,34]. Transarterial embolization is usually limited to ECA branches. Embolization of ophthalmic artery branches is usually avoided as it carries the risk of central retinal artery occlusion. Transvenous embolization has been described but is not possible for most lesions due to the lack of venous access. Surgery is associated with high cure rates and low complication rates and is the treatment of choice for most anterior fossa DAVFs. This is performed through a low unilateral frontal or bifrontal craniotomy in the supine position with minimal retraction of the frontal lobes followed by disconnection of the arterialized leptomeningeal veins which should be coagulated and divided. In this study four patients with an anterior cranial fossa DAVFs underwent successful surgery and achieved complete exclusion. An example is shown in Fig. 3. A patient with type IV anterior cranial fossa DAVF and hemorrhage developed transient right hemiparesis 4 days after surgical intervention due to vasospasm. Probably the vasospasm was a delayed response to the pre-existent hemorrhage, rather than
Fig. 3. Anterior cranial fossa dural fistula type 4 (according to Cognard classification) with “foot” of the fistula localized medially on the cribriform plate. The fistula is fed by multiple ethmoidal branches originating from the distal portion of both the internal maxillary artery and to a lesser extent by the distal branches of both ophthalmic arteries. The fronto-basal pial venous drainage, dilated and tortuous, soon bifurcates into a frontal branch that flows into the superior sagittal sinus and in a left one which has an aneurysmal dilatation. This venous drainage flows into the left cavernous sinus with secondary reflux into the contralateral cavernous sinus. In the right lower box the CT scan at admission shows a deep left frontal hemorrhage.
F. Signorelli et al. / Clinical Neurology and Neurosurgery 128 (2015) 123–129
a true procedure complication. She gradually recovered and in 15th day after surgery was discharged with no neurological deficit. Tentorial DAVFs display very often a leptomeningeal venous drainage involving the basal vein of Rosenthal, the lateral mesencephalic veins and the vein of Galen. The presence of CVR explains the high incidence of intracranial hemorrhage and neurologic deficit with these lesions. Transarterial embolization of the arterial feeders is often the preferred strategy to palliatively treat tentorial DAVFs without CVR [35]. Surgical exposure of tentorial DAVFs may be performed using a posterior parasagittal craniotomy and interhemispheric approach followed by CVR disconnection. In one case of tentorial DAVF, endovascular procedure was complicated by an intracranial hemorrhage. Then surgical intervention warranted hemorrhage removal and fistula exclusion. There is evidence that rebleeding may occur early after presentation, so it is reasonable to recommend treatment soon after diagnosis [6]. Moreover, tentorial and anterior cranial fossa DAVFs are among the most frequently presenting with hemorrhage, in our as in other series. Therefore, due to the difficult endovascular access route, surgical treatment should be recommended for recently diagnosed tentorial DAVFs and for residual ones after incomplete endovascular embolization. 5. Conclusions DAVFs encompass a broad spectrum of symptoms and clinical manifestations depending on their location and hemodynamic features. Thus strategies for treatment should be decided according to the characteristics of the DAVFs, subjective symptoms of the patients and after careful analysis of risk rate of each procedure. Disclosure statement The authors declare no interest to disclose, they have any personal or institutional financial interest in drugs, materials, or devices described in this submission and that they did not receive any specific funding. The paper and any of its contents have been presented previously. References [1] Gandhi D, Chen J, Pearl M, Huang J, Gemmete JJ, Kathuria S. Intracranial dural arteriovenous fistulas: classification, imaging findings, and treatment. Am J Neuroradiol 2012;33:1007–13. [2] Oh JT, Chung SY, Lanzino G, Park KS, Kim SM, Park MS, et al. Intracranial dural arteriovenous fistulas: clinical characteristics and management based on location and hemodynamics. J Cerebrovasc Endovasc Neurosurg 2012;14: 192–202. [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;82:166–79. [4] 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;85:830–7. [5] Cognard C, Gobin YP, Pierot L, Bailly AL, Houdart E, Casasco A, et al. Cerebral dural arteriovenous fistulas: clinical and angiographic correlation with a revised classification of venous drainage. Radiology 1995;194:671–80. [6] van Dijk JM, terBrugge KG, Willinsky RA, Wallace MC. Clinical course of cranial dural arteriovenous fistulas with long-term persistent cortical venous reflux. Stroke 2002;33:1233–6. [7] Satomi J, van Dijk JM, terbrugge KG, Willinsky RA, Wallace MC. Benign cranial dural arteriovenous fistulas: outcome of conservative management based on the natural history of the lesion. J Neurosurg 2002;97:767–70.
129
[8] Saito A, Furuno Y, Nishimura S, Kamiyama H, Nishijima M. Spontaneous closure of transverse sinus dural arteriovenous fistula: case report. Neurol Med Chir (Tokyo) 2008;48:564–8. [9] Olutola PS, Eliam M, Molot M, Talalla A. Spontaneous regression of a dural arteriovenous malformation. Neurosurgery 1983;12:687–90. [10] Gupta A, Periakaruppan A. Intracranial dural arteriovenous fistulas: a review. Indian J Radiol Imaging 2009;19:43–8. [11] Chung SJ, Kim JS, Kim JC, Lee SK, Kwon SU, Lee MC, et al. Intracranial dural arteriovenous fistulas: analysis of 60 patients. Cerebrovasc Dis 2002;13:79–88. [12] Kerber CW, Newton TH. The macro and microvasculature of the dura mater. Neuroradiology 1973;6:175–9. [13] Kojima T, Miyachi S, Sahara Y, Nakai K, Okamoto T, Hattori K, et al. The relationship between venous hypertension and expression of vascular endothelial growth factor: hemodynamic and immunohistochemical examinations in a rat venous hypertension model. Surg Neurol 2007;68:277–84, discussion 284. [14] Fujita A, Kuwamura K, Saitoh M, Sakagami Y, Takaishi Y, Suzuki S, et al. Cerebral sinus thrombosis in a patient with protein S deficiency: a case report. No Shinkei Geka 1997;25:467–72. [15] Gerlach R, Yahya H, Rohde S, Bohm M, Berkefeld J, Scharrer I, et al. Increased incidence of thrombophilic abnormalities in patients with cranial dural arteriovenous fistulae. Neurol Res 2003;25:745–8. [16] Houser OW, Campbell JK, Campbell RJ, Sundt Jr TM. Arteriovenous malformation affecting the transverse dural venous sinus–an acquired lesion. Mayo Clin Proc 1979;54:651–61. [17] Ghobrial GM, Marchan E, Nair AK, Dumont AS, Tjoumakaris SI, Gonzalez LF, et al. Dural arteriovenous fistulas: a review of the literature and a presentation of a single institution’s experience. World Neurosurg 2013;80:94–102. [18] Kirsch M, Liebig T, Kuhne D, Henkes H. Endovascular management of dural arteriovenous fistulas of the transverse and sigmoid sinus in 150 patients. Neuroradiology 2009;51:477–83. [19] Hacein-Bey L, Konstas AA, Pile-Spellman J. Natural history, current concepts, classification, factors impacting endovascular therapy, and pathophysiology of cerebral and spinal dural arteriovenous fistulas. Clin Neurol Neurosurg 2014;121:64–75. [20] Bernstein R, Dowd CF, Gress DR. Rapidly reversible dementia. Lancet 2003;361:392. [21] Lv X, Jiang C, Li Y, Liu L, Liu J, Wu Z. Transverse-sigmoid sinus dural arteriovenous fistulae. World Neurosurg 2010;74:297–305. [22] Daniels DJ, Vellimana AK, Zipfel GJ, Lanzino G. Intracranial hemorrhage from dural arteriovenous fistulas: clinical features and outcome. Neurosurg Focus 2013;34:E15. [23] Choi JH, Mast H, Sciacca RR, Hartmann A, Khaw AV, Mohr JP, et al. Clinical outcome after first and recurrent hemorrhage in patients with untreated brain arteriovenous malformation. Stroke 2006;37:1243–7. [24] King WA, Martin NA. Intracerebral hemorrhage due to dural arteriovenous malformations and fistulae. Neurosurg Clin N Am 1992;3:577–90. [25] Brasse G, Roth HJ, Fritzsch D, Scheid R. Now you see it, now you don’t – the hidden life of cerebral dural arteriovenous fistulae. Neuroradiology 2009;51:131–3. [26] Sarma D, ter Brugge K. Management of intracranial dural arteriovenous shunts in adults. Eur J Radiol 2003;46:206–20. [27] Urtasun F, Biondi A, Casaco A, Houdart E, Caputo N, Aymard A, et al. Cerebral dural arteriovenous fistulas: percutaneous transvenous embolization. Radiology 1996;199:209–17. [28] Pandey P, Steinberg GK, Westbroek EM, Dodd R, Do HM, Marks MP. Intraoperative angiography for cranial dural arteriovenous fistula. Am J Neuroradiol 2011;32:1091–5. [29] Suh DC, Lee JH, Kim SJ, Chung SJ, Choi CG, Kim HJ, et al. New concept in cavernous sinus dural arteriovenous fistula: correlation with presenting symptom and venous drainage patterns. Stroke 2005;36:1134–9. [30] Halbach VV, Higashida RT, Hieshima GB, Hardin CW, Pribram H. Transvenous embolization of dural fistulas involving the cavernous sinus. AJNR 1989;10:377–83. [31] Piippo A, Laakso A, Seppa K, Rinne J, Jaaskelainen JE, Hernesniemi J, et al. Early and long-term excess mortality in 227 patients with intracranial dural arteriovenous fistulas. J Neurosurg 2013;119:164–71. [32] Klisch J, Huppertz HJ, Spetzger U, Hetzel A, Seeger W, Schumacher M. Transvenous treatment of carotid cavernous and dural arteriovenous fistulae: results for 31 patients and review of the literature. Neurosurgery 2003;53:836–56, discussion 856–837. [33] 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;47:56–66, discussion 66–57. [34] Lucas CP, Zabramski JM, Spetzler RF, Jacobowitz R. Treatment for intracranial dural arteriovenous malformations: a meta-analysis from the English language literature. Neurosurgery 1997;40:1119–30, discussion 1130–1112. [35] Javadpour M, Jain H, Wallace MC, Willinsky RA, ter Brugge KG, Tymianski M. Analysis of cost related to clinical and angiographic outcomes of aneurysm patients enrolled in the international subarachnoid aneurysm trial in a North American setting. Neurosurgery 2005;56:886–94, discussion 886–894.
