Dural Arteriovenous Fistulas: the value of Carotid Ultrasonography.

Transcription

1 Dural Arteriovenous Fistulas: the value of Carotid Ultrasonography. Poster No.: C-2199 Congress: ECR 2014 Type: Educational Exhibit Authors: J. P. Filipe, T. Parreira, C. Andrade, R. Santos, E. Azevedo; Porto/ PT Keywords: Ultrasound-Spectral Doppler, Ultrasound-Colour Doppler, Vascular, Neuroradiology brain, Head and neck, Catheter arteriography, Diagnostic procedure, Arteriovenous malformations, Fistula, Hemodynamics / Flow dynamics DOI: /ecr2014/C-2199 Any information contained in this pdf file is automatically generated from digital material submitted to EPOS by third parties in the form of scientific presentations. References to any names, marks, products, or services of third parties or hypertext links to thirdparty sites or information are provided solely as a convenience to you and do not in any way constitute or imply ECR's endorsement, sponsorship or recommendation of the third party, information, product or service. ECR is not responsible for the content of these pages and does not make any representations regarding the content or accuracy of material in this file. As per copyright regulations, any unauthorised use of the material or parts thereof as well as commercial reproduction or multiple distribution by any traditional or electronically based reproduction/publication method ist strictly prohibited. You agree to defend, indemnify, and hold ECR harmless from and against any and all claims, damages, costs, and expenses, including attorneys' fees, arising from or related to your use of these pages. Please note: Links to movies, ppt slideshows and any other multimedia files are not available in the pdf version of presentations. Page 1 of 39

2 Learning objectives (a) Review the clinical presentation, classification and imaging findings of dural arteriovenous fistulas (DAVFs); (b) Summarize and illustrate ultrasonographic suggestive features of DAVFs; (c) Highlight the current progress regarding carotid ultrasonography in the evaluation of DAVFs. Please note that DAVFs in the cavernous sinus (Carotid-cavernous fistulas Barrow type B-D) are out of the scope of the present work, due to the extremely limited value of carotid ultrasonography in those lesions. Background 1) Introduction Intracranial dural arteriovenous fistulas (DAVFs) are pathologic shunts between dural arteries and dural venous sinuses, meningeal veins or cortical veins. Their true incidence is unknown, but they are estimated to account for 10%-15% of all intracranial arteriovenous malformations (AVMs), comprising arround 6% of supratentorial and 35% of infratentorial AVMs. DAVFs tend to present later in life than AVMs and are distinguished from parenchymal or pial AVMs mostly by the presence of a dural arterial supply and the absence of a parenchymal nidus. Althouth clinical presentation is diverse, patients may develop intracranial haemorrhage (overall risk ~ 1.5% per year), sinus thrombosis or venous infarct as their first presentation (fig. 4, fig. 5). As so, early diagnosis and management of DAVFs may prevent serious complications, as the occurrence of stroke. Page 2 of 39

3 While invasiveness of conventional cerebral catheter angiography dificults regular investigation of patients with suspected DAVF, more studies have attempted to use ultrasonography in order to detect early hemodynamic changes sugestive of DAVF. 2) Etiology and Patophysiology Most the lesions are considered idiopathic but some associated conditions have been reported: dural sinus thrombosis; venous hypertension; previous craniotomy; head trauma; ear infection; secondary to aneurysmal rupture; heritable risk factors for venous thrombosis (such as antithrombin, protein C and protein S deficiencies, factor V Leiden); The pathophysiology of DAVFs is believed to derive from venous thrombosis of a dural sinus. The two most accepted hypotheses are: Venous thrombosis promotes venous hypertension, which may induce the opening of physiologic arteriovenous connections (microscopic dural sinus's vasa vasorum or the dural venous plexus embryonic remnants). Maturation of these channels, secondary to venous stenosis or occlusion, results in the development of direct pathologic shunts. Venous hypertension, due to outflow obstruction, causes decreased cerebral perfusion with the subsequent release of angiogenic factors (vascular endothelial growth factor, basic fibroblast growth factor, etc.) promoting neoangiogenesis and the development of a DAVF. Generally, the arterial supply to DAVFs is by meningeal arteries and the fistulous connection is contained in the leaflets of the dura mater. Occasionally, as the fistula grows or becomes larger, pial recruitment from parenchymal vessels and dilatation of cortical veins can occur. 3) Classification The two commonly used classification systems are the more user-friendly Borden classification (fig. 1-3) and the more detailed Cognard system. Page 3 of 39

4 While the Borden classification system stratifies lesions based on the local and direction of venous drainage, and the presence or absence of cortical venous drainage (CVD), the Cognard classification adds the venous outflow architecture to the equation. The DAVF venous drainage pattern determines the severity of symptoms and provides the foundation for both classification schemes and both of them associate CVD with increased risk of intracranial hemorrhage and nonhemorrhagic neurologic deficits. 3.1) Borden classification system Description CVD? / High risk lesion? Borden 1 Drainage of meningeal arteries directly into dural sinus or meningeal vein, with anterograde flow. Borden 2 Anterograde drainage YES into dural venous sinus + retrograde flow into cortical subarachnoid veins (with CVR). Borden 3 Direct retrograde flow from meningeal arteries into cortical veins or an "isolated" sinus segment (CVR only). YES Page 4 of 39

5 Fig. 1: A 58 year-old woman with a Borden 1 DAVF. Left ECA angiogram, lateral projection showing a osteodural AVF fistulous area on the left sigmoid sinus (red circle). We can see enlarged occipital artery with multiple transosseous branches feeding the davf (blue arrow), feeding branches from the superficial temporal and middle meningeal arteries (white arrows) and feeding branches from the posterior auricular artery (green arrow). The venous drainage is anterograde throw the jugular vein (grey arrow). [Patient A in Fig. 1] References: Neuroradiology, Hospital de São João - Porto/PT Page 5 of 39

6 Fig. 2: A 44 year-old woman with Borden 2 DAVF. Left ECA angiogram, lateral projection (A) showing arterial feeding mainly from branches of the middle meningeal artery (white arrow) and transosseous branches of the occipital artery (green arrow). The fistulous area is despicable in a compartment of incarcerated left transverse sinus (red circle) with cortical venous reflux and venous ectasias (blue) present. [Patient B in Fig. 2, Fig. 4 and Fig. 6] References: Neuroradiology, Hospital de São João - Porto/PT Page 6 of 39

7 Fig. 3: A 76 year-old man with a Borden 3 DAVF. Right ECA angiogram, lateral projection (A) and right ICA angiogram, lateral projection (B) showing arterial supply to the fistula via the middle meningeal and accessory meningeal arteries (white arrows) and also from the tentorium cerebelli dural branches arising from the meningohypophyseal trunk of the ICA (blue arrow). The fistulous point seems to be located over the anterior and medial portion of tentorium cerebelli (red circle) and the venous drainage is made directly to an enlarged cortical vein (green arrows) that ultimately drains to the straight sinus (grey arrow). [Patient C in Fig. 3, Fig. 7, Fig. 8 and Fig. 9] References: Neuroradiology, Hospital de São João - Porto/PT 3.2) Cognard classification system Description CVD? / High risk lesion? Type I Direct anterograde flow into a dural venous sinus Type II a Drainage into a sinus with retrograde flow within the sinus Type II b Drainage into a sinus with YES retrograde flow into cortical veins Type II a+b Drainage into a sinus with retrograde flow within the sinus and cortical veins YES Page 7 of 39

8 Type III Direct drainage into a cortical vein without venous ectasia YES Type IV Direct drainage into a cortical vein with ectasia >5mm and 3x larger than the diameter of the draining vein YES Type V Direct drainage into spinal perimedullary veins YES 4) Natural History Even nowadays, data on DAVFs natural history is limited. Lack of CVD (Borden type I, Cognard types I and IIa) is a favorable feature and is associated with a benign natural history and the presence of CVD (Borden type 2 and 3, Cognard types IIb to V) is an aggressive feature that places DAVFs in a high-risk category. Regarding high-risk lesions, some values have been estimated: annual mortality rate of 10.4%; annual risk of intracranial hemorrhage of 8.1%; annual risk of Nonhemorrhagic neurologic deficits of 6.9%; annual hemorrhage risk is diferente between symptomatic (aprox 7.5%) and asymptomatic patients (aprox 1.5%). risk of rebleeding within 2 weeks after inicial hemorrhage of 35% Although classifying DAVFs is helpful for risk stratification, one should be aware that these lesions have a dynamic nature: the risk of conversion from a benign to an aggressive DAVF is small (~ 2%) but sufficient to warrant repeat investigation if the clinical picture changes. 5) Clinical Presentation The majority of patients with DAVFs present in the fifth and sixth decades of life with symptoms related to lesion location and pattern of venous drainage. The clinical picture may include: Page 8 of 39

9 - Pulsatile tinnitus (sometimes associated with an audible bruit). A common symptom that results from high flow in AV shunts, particularly those drainaging to the transverse or sigmoid sinus. - Ocular/visual symptoms (such as ophthalmoplegia, proptosis, chemosis, retro-orbital pain or decreased visual acuity), more commonly found in cavernous sinus DAVFs. - Intracranial hemorrhage (including subarachnoid, subdural or intraparenchymal), probably caused by rupture of draining veins, as the source of hemorrhage is usually not the nidus but rather the distended leptomeningeal venous varices (fig. 6). - Nonhemorrhagic neurologic deficits (seizures, parkinsonism, cerebellar symptoms, apathy, failure to thrive and cranial nerve abnormalities, including rare cases of trigeminal neuralgia), probably due to arterial steal. - Dementia-like syndrome with cognitive decline (that may improve after treatment). Common clinical Main presentation supply Transversepulsatile tinnitus, sigmoid junction headaches. DAVFs arterial Venous drainage Transmastoid branches of the occipital art, branches of the middle meningeal art, meningeal Particular attention should be made to the direction of flow in the vein of Labbé branches of the ascending pharyngeal artery, branches of the vertebral artery Carotid-cavernous chemosis, DAVFs exophthalmos, third and/or sixth cranial nerve palsies, decreased visual acuity, headache, Middle or accessory meningeal artery, artery of the foramen rotundum, ascending pharyngeal artery Superior ophtalmic vein or inferior ophthalmic vein into facial venous system and external jugular vein Page 9 of 39

10 tinnitus and/or bruit. Superior sagittal intracranial sinus DAVFs hemorrhage, progressive neurological deficits. Bilateral supply from branches of the middle meningeal art, anterior falcine artery from the ophtalmic art Tentorial DAVFs non-specific symptoms headache, ataxia, trigeminal neuralgia, etc.) Anterior cranial headaches, fossa (ethmoidal) seizures, DAVFs ocular symptoms (such as exophthalmos). atypical symptoms (such as proptosis, chemosis, elevated IOP, and loss of vision Tentorial artery Superior petrosal sinus or pontine, (Bernasconi and perimesencephalic Cassinari), petrosal and basal veins into and the petrosquamosal branches of the galenic system middle meningeal artery Posterior anterior and Main drainage into the frontal cortical veins and ethmoidal branches secondarily into of Oph art. thesuperior sagital Secondary supply sinus or the anterior from the ethmoidal branches of the pontomesencephalic internal and middle vein meningeal arts or superior temporal artery high incidence of ICH Superior petrosal headaches, sinus DAVFs pulsatile tinnitus, Tentorial artery Superior petrosal sinus, pontine and (Bernasconi and perimesencephalic Cassinari), petrosal trigeminal neuralgia, and veins. It also drains petrosquamosal retrograde to the ocular symptoms, branches of the vein of Galen and middle meningeal antegrade intracranial artery hemorrhage down the internal jugular vein Page 10 of 39

11 Common clinical presentations by anatomical localization Indicators for early treatment include: intracranial hemorrhage; increased intracranial pressure; increased intraocular pressure causing visual loss; neurovascular compression producing neurological deficits; presence of leptomeningeal drainage / CVD. 6) Diagnosis / Imaging Because clinical and imaging features can be non-specific, the diagnosis of a DAVF is often delayed or missed. Occasionally plain films can demonstrate grooving within the skull vault due to chronic compression from enlarged middle meningeal vessels. Cross-sectional imaging techniques (CT and MR imaging) aid in the diagnosis and mostly constitute the initial radiological evaluation. Head computed tomography and magnetic resonance imaging studies have a reported sensitivity for detection of DAVFs of 70% and 81%, respectively. Conventional angiography remains the most accurate method for complete characterization and classification of DAVFs. Doppler sonography represents a minimally invasive test that can be applied as a diagnostic screening tool, and treatment monitoring in patients with DAVF, as we will address in a separated topic under "Findings and procedure details". 6.1) CT Non-contrast CT is limited to the identification of complications such as intracranial hemorrhage, venous sinus thrombosis or edema due to venous congestion. Linear bony defects formed by enlarged emissary veins, similar to the grooving abnormality seen on plain film, can also be found. Page 11 of 39

12 Fig. 4: CT scan of a 44 year-old woman with a Borden 2 DAVF, later confirmed by digital subtraction angiography, who presented with focal neurological signs secondary to the brain hemorrhage seen in the left parietal lobe (white arrow). [Patient B in Fig. 2, Fig. 4 and Fig. 6] References: Neuroradiology, Hospital de São João - Porto/PT Page 12 of 39

13 Fig. 5: Initial CT scan of a 53 year-old man revealing a hyperdense region in the right sigmoid and lateral sinus. Broader study with gadolinium injection revealed a large thrombus extending from the transverse, lateral and sigmoid sinus to the cervical portion of the internal jugular vein. [Patient D in Fig. 5]] References: Neuroradiology, Hospital de São João - Porto/PT 6.2) MR imaging Page 13 of 39

14 An increasing number of DAVFs are being diagnosed incidentally due to the more frequente use of high-resolution MRI and the wider availability of the technique. T2-weighted MRI is more sensitive to the white matter changes of venous congestion or infarction when compared to CT. Findings vary with the type of DAVF and include: dilated cortical vessels, appearing as flow-voids located within the cortical sulci without a true parenchymal nidus (originally referred to as a "pseudophlebitic" pattern); venous pouches; prominent occipital arteries; vascular enhancement; osseous abnormalities due to ecstatic or hypertrophied vessels; sinus stenosis or thrombosis; signs of venous hypertension (as white matter hyperintensity, intracranial hemorrhage or venous infarction). Fig. 6: MRI of a 44 year-old woman with Borden 2 DAVF, later confirmed by digital subtraction angiography (C). Axial DP (A) and Sagital contrast-enhanced T1 (B) showing parenchymal (blue arrow) and subdural hematoma (white arrow). We can clearly depict the flow void pouch on axial DP (A), enhancing after contrast (B), probably corresponding to the rupture point of the venous pouch of this cortical vein. Angiographic study (C) confirms davf with cortical venous reflux and venous ectasia (red arrow) as the likely point of rupture. [Patient B in Fig. 2, Fig. 4 and Fig. 6] References: Neuroradiology, Hospital de São João - Porto/PT Page 14 of 39

15 Fig. 7: A 76 year-old man with a Borden 3 DAVF, later confirmed by digital subtraction angiography. Axial FLAIR showing mesencephalic edema (red arrow) and flow voids corresponding to a venous ectasia. [Patient C in Fig. 3, Fig. 7, Fig. 8 and Fig. 9] References: Neuroradiology, Hospital de São João - Porto/PT Lesions Borden type 1 and 2 are more likely to show flow-void clustering, engorged ophthalmic veins or proptosis, while aggressive type 2 or 3 lesions may reveal dilated vessels, prominente vascular enhancement and hemorrhage. 6.3) CT angiography Page 15 of 39

16 Any suspicious flow void cluster around the dural venous sinus should prompt additional evaluation with dynamic CTA, MRA, or DSA. CTA is particularly useful in treatment planning by helping define the arteriovenous shunt relative to surrounding brain and skull. Fig. 8: A 76 year-old man with a Borden 3 DAVF. Initial CT scan (A) revealing subarachnoid hemorrhage in perimesencephalic cistern bilaterally (white arrow). CT angiography (B) showing the presence of venous ectasia with wall irregularities, located around the perimesencephalic cistern, more pronounced at the right (blue arrow), apparently converging to an aneurysmatic dilatation (green arrow). We cannot depict the arteriovenous nidus in this study. [Patient C in Fig. 3, Fig. 7, Fig. 8 and Fig. 9] References: Neuroradiology, Hospital de São João - Porto/PT Multi-detector CTA can provide high resolution for vascular anatomy and, in the investigation of tinnitus, has the additional advantage of evaluating inner and middle ear abnormalities. Arterialised venous blood within the veins draining a DAVF has increased density when compared to non-arterialised blood and that can be assessed by carefull evaluation using narrow window settings. Recent publications on 4D CTA by using 320-section multidetector row CT angiography have highlighted its potential to diagnosis, classification and treatment planning. More reported experience is needed, though. 6.4) MR angiography Page 16 of 39

17 Studies have reported that MRA has more sensitivity than CTA for the detection of DAVFs (50% vs 15.4%). Due to current limitations of low resolution, restricted FOV, and saturation artifacts, the negative predictive value of MRA is still inadequate to exclude DAVFs. Page 17 of 39

18 Page 18 of 39

19 Fig. 9: A 76 year-old man with a Borden 3 DAVF. TOF-3D MPR axial MIP 4mm (A) TOF-3D MPR oblique MIP 7mm (B) and corresponding right ICA angiogram, lateral projection (C) showing a DAVF located over the anterior and medial portion of tentorium cerebelli (red circle). We can see and compare the arterial supply from the meningohypophyseal trunk of the ICA, via the tentorium cerebelli dural branches (blue arrows), and the venous drainage directly to an enlarged cortical vein (green arrows). [Patient C in Fig. 3, Fig. 7, Fig. 8 and Fig. 9] References: Neuroradiology, Hospital de São João - Porto/PT 3D TOF MRA is sensitive enough to detect flow signal in arterialized veins, especially with reversal of flow direction, but this technique currently has a high false-positive rate of 14%. Time-resolved MRA techniques are promising and may be reliable for DAVF screening and surveillance in the future. 6.5) Digital subtraction angiography Benign disease, without cortical venous reflux can be missed using both CT and MRI. Therefore, conventional angiography remains the most accurate method for detection and classification of DAVFs (fig. 1-3, fig. 11). However, their invasiveness and high cost make it less attractive as a screening tool. The pretreatment angiographic evaluation must include: Bilateral ICA and ECAs, and it can also include the vertebral arteries when evaluating DAVFs in the posterior fossa or tentorium. Subsequently, superselective angiography is required to access the arterial supply (including anastomoses with arterial distributions to the brain, orbit, or cranial nerves), and the venous anatomy (with respect to the pattern of venous drainage and the adequacy of normal venous flow from the brain). Arterial and venous anatomy can affect the selection of the ideal treatment modality and it's extremely important in order to avoid treatment complications. Angiographic features associated with aggressive behaviour: - leptomeningeal retrograde; - venous drainage, variceal or aneurysmal venous structures; - galenic venous drainage. Page 19 of 39

20 Images for this section: Fig. 1: A 58 year-old woman with a Borden 1 DAVF. Left ECA angiogram, lateral projection showing a osteodural AVF fistulous area on the left sigmoid sinus (red circle). We can see enlarged occipital artery with multiple transosseous branches feeding the davf (blue arrow), feeding branches from the superficial temporal and middle meningeal arteries (white arrows) and feeding branches from the posterior auricular artery (green arrow). The venous drainage is anterograde throw the jugular vein (grey arrow). [Patient A in Fig. 1] Page 20 of 39

21 Fig. 2: A 44 year-old woman with Borden 2 DAVF. Left ECA angiogram, lateral projection (A) showing arterial feeding mainly from branches of the middle meningeal artery (white arrow) and transosseous branches of the occipital artery (green arrow). The fistulous area is despicable in a compartment of incarcerated left transverse sinus (red circle) with cortical venous reflux and venous ectasias (blue) present. [Patient B in Fig. 2, Fig. 4 and Fig. 6] Page 21 of 39

22 Fig. 3: A 76 year-old man with a Borden 3 DAVF. Right ECA angiogram, lateral projection (A) and right ICA angiogram, lateral projection (B) showing arterial supply to the fistula via the middle meningeal and accessory meningeal arteries (white arrows) and also from the tentorium cerebelli dural branches arising from the meningohypophyseal trunk of the ICA (blue arrow). The fistulous point seems to be located over the anterior and medial portion of tentorium cerebelli (red circle) and the venous drainage is made directly to an enlarged cortical vein (green arrows) that ultimately drains to the straight sinus (grey arrow). [Patient C in Fig. 3, Fig. 7, Fig. 8 and Fig. 9] Page 22 of 39

23 Fig. 4: CT scan of a 44 year-old woman with a Borden 2 DAVF, later confirmed by digital subtraction angiography, who presented with focal neurological signs secondary to the brain hemorrhage seen in the left parietal lobe (white arrow). [Patient B in Fig. 2, Fig. 4 and Fig. 6] Page 23 of 39

24 Fig. 5: Initial CT scan of a 53 year-old man revealing a hyperdense region in the right sigmoid and lateral sinus. Broader study with gadolinium injection revealed a large thrombus extending from the transverse, lateral and sigmoid sinus to the cervical portion of the internal jugular vein. [Patient D in Fig. 5]] Page 24 of 39

25 Fig. 6: MRI of a 44 year-old woman with Borden 2 DAVF, later confirmed by digital subtraction angiography (C). Axial DP (A) and Sagital contrast-enhanced T1 (B) showing parenchymal (blue arrow) and subdural hematoma (white arrow). We can clearly depict the flow void pouch on axial DP (A), enhancing after contrast (B), probably corresponding to the rupture point of the venous pouch of this cortical vein. Angiographic study (C) confirms davf with cortical venous reflux and venous ectasia (red arrow) as the likely point of rupture. [Patient B in Fig. 2, Fig. 4 and Fig. 6] Page 25 of 39

26 Fig. 7: A 76 year-old man with a Borden 3 DAVF, later confirmed by digital subtraction angiography. Axial FLAIR showing mesencephalic edema (red arrow) and flow voids corresponding to a venous ectasia. [Patient C in Fig. 3, Fig. 7, Fig. 8 and Fig. 9] Page 26 of 39

27 Fig. 8: A 76 year-old man with a Borden 3 DAVF. Initial CT scan (A) revealing subarachnoid hemorrhage in perimesencephalic cistern bilaterally (white arrow). CT angiography (B) showing the presence of venous ectasia with wall irregularities, located around the perimesencephalic cistern, more pronounced at the right (blue arrow), apparently converging to an aneurysmatic dilatation (green arrow). We cannot depict the arteriovenous nidus in this study. [Patient C in Fig. 3, Fig. 7, Fig. 8 and Fig. 9] Page 27 of 39

28 Page 28 of 39

29 Fig. 9: A 76 year-old man with a Borden 3 DAVF. TOF-3D MPR axial MIP 4mm (A) TOF-3D MPR oblique MIP 7mm (B) and corresponding right ICA angiogram, lateral projection (C) showing a DAVF located over the anterior and medial portion of tentorium cerebelli (red circle). We can see and compare the arterial supply from the meningohypophyseal trunk of the ICA, via the tentorium cerebelli dural branches (blue arrows), and the venous drainage directly to an enlarged cortical vein (green arrows). [Patient C in Fig. 3, Fig. 7, Fig. 8 and Fig. 9] Page 29 of 39

30 Findings and procedure details Carotid doppler sonography (CDS) represents a minimally invasive screening test that yield hemodynamic information very suggestive of DAVF, such as low resistance and high velocity flow in external carotid arteries (ECA). These hemodynamic changes can be useful for follow-up and evaluation of treatment effectiveness. On the other hand, the most important limitation is that it only shows the hemodynamic changes in the feeding arteries without showing the actual size and patterns of the venous drainage. As CDS has been widely used to evaluate the severity of atherosclerosis in carotid arteries, simultaneous analysis of the ECA Resistance index (RI) during routine CDS may detect occult DAVF in some patients. During follow-up after DAVF diagnosis or treatment, conventional angiography should be performed to revaluate the status of the lesion, in all patients presenting deterioration of CDS results (e.g., lowering of the ECA RI). Some authors advocate that follow-up using CDS should be carried out at least once per year or any time the clinical picture of the patient changes. 6.1) Rationale of using ultrasonography in DAVF diagnosis Diagnosis of DAVF by ultrasonography depends on hemodynamic changes in the feeding arteries and drainage veins as well as shortened cerebral circulation time ) Arterial findings As most DAVFs are feed from branches of the ECA, pathological hemodynamic changes can be accessed in cervical arteries and translate into: Increased flow volume; Increased Peak-systolic velocity (PSV); Increased End-diastolic velocity (EDV); Reduction in flow resistance (due to direct shunting in the absence of capillary beds)*. (*Resistance index (RI) = (PSV# EDV)/PSV) Page 30 of 39

31 6.1.2) Venous findings In normal conditions, the venous system displays low flow velocity (without pulsatile waveforms, as seen in arteries). High flow through the AV shunts may also induce hemodynamic changes in the cerebral sinuses and drainage veins: Arterialized pulsatile waveforms; Increased flow velocity in cervical veins; Venous dilatations; Reversed blood flow. 6.2) CDS in follow-up after treatment Some studies showed that CDS could be used to evaluate effectiveness of DAVFs treatment. In order to do so, CDS should be performed before and immediately after endovascular therapy and the baseline hemodynamic status should be registered, so changes during follow-up can be noticed. It was found a good correlation between serial ECA RI changes and the evolution of clinical symptoms related to dural AVFs. In patients with total resolution of the DAVF, the ECA RI increases after embolization, while the same is not true in lesions that were only partially embolized. 6.3) Ultrasound equipment Looking at literature, the most reported systems have been the HP 4500 (HewlettPackard Co., Palo Alto, CA), the VST Master (Diasonics, Tirat Carmel, Israel) and the SSD-3000 (Aloka Co., Ltd., Tokyo, Japan), but virtually any ultrasound system with carotid doppler capability can be used. Regarding to transducers, those reported usually range from 3 to 11 MHz real time Bmode-transducers associated with 3 to 6 MHz pulsed-doppler-transducers. 6.4) CDS parameters for DAVFs Establishing the optimal cut-off values for the screening, diagnosis and follow-up of DAVFs is the critical point, as those values are slightly different across studies. The table below summarizes some significant data: Page 31 of 39

32 No. of DAFVs studied Proposed parameters Conclusions Tsai et al. (2004) 35 RI of ECA < 0.72 (right) or < 0.71 (left) RI of ECA is the best parameter for predicting DAVFs (74% sensitivity and 89% specificity). Tsai et al. (2005) 29 Change in RI of RI of ECA is correlated with ECA # 0.1 the effectiveness of treatment and (correlates with clinical evolution of DAVF clinical status DAVFs. at follow-up) Arning et al. (2005) 17 Low RI of branches Tsai et al. (2009) 63 RI of ECA < 0,7 ECA Studying ECA branches is more diagnostic (positive predictive value (versus CCA and 100%). ICA) and ICA-to-ECA of RI > 0,9 Combined CDS parameters can be used for screening of DAVF (51% ratio sensitivity and 99% specificity). (validated in subjects) Yeh et al. (2010) 67 RI of ECA < 0,72 Tee et al. (2013) 24 RI of Occipital artery <0.76 RI and EDV of ECA have and high diagnostic sensitivity for EDV of ECA > 21 DAVFs in cm/s patients with pulsatile tinnitus (sensitivity 92% to 95%). RI of Occipital artery can be used to screen for DAVFs having Page 32 of 39

33 the occipital artery as one of feeding arteries (about half of DAVFs) (96% sensitivity and 97% specificity). Although the parameter RI of ECA seems the most sensitive for predicting DAVF, the combination of ECA RI and ICA-to-ECA ratio of RI have highest specificity (99%) and positive predictive value (97%). Therefore, if the diagnosis of DAVF is suspected in patients with equivocal symptoms (such as tinnitus), that set of parameters can be used to enhance our confidence to arrange more invasive and expensive studies. But, since the diagnostic sensitivity is not high (51%), that set of parameters is only recomended as a screening tool and NOT to confirm a diagnosis of DAVF. Page 33 of 39

34 Fig. 10: 42-year-old woman with left pulsatile tinnitus and bilateral occipital bruit. Carotid duplex sonography showed low resistance index (RI = 0.48) and high end diastolic velocity (EDV = 128 cm/s) in the left ECA (B) with an ICA to ECA RI ratio of 1.3 (C). Analyzes of the left Internal Jugular Vein (D) revealed a pseudoarterialized flow pattern and an increased flow velocity. Similar findings were recorded in the right side. [Patient E in Fig. 10 and Fig. 11] References: Neuroradiology, Hospital de São João - Porto/PT Fig. 11: 42-year-old woman with left pulsatile tinnitus and bilateral occipital bruit (seen in fig. 10). Conventional angiography showed a high flow dural arteriovenous fistula, Borden 2, with feeders from the left occipital artery, posterior branches of the middle meningeal artery, neuromeningeal division of the ascending pharyngeal artery (A, left ECA, left lateral view), as well as participation from the right occipital artery (B, right Page 34 of 39

35 ECA and ICA, right lateral view), and from dural branches of the right vertebral artery (C, right vertebral artery, right lateral view). Venous drainage showing early venous filling of the internal jugular vein and retrograde flow throw cortical veins (D, left ECA, anteroposterior view). References: Neuroradiology, Hospital de São João - Porto/PT To summarize the best CDS signs of DAVF: (a) ECA RI < 0.7 and ICA-to-ECA ratio of RI > 0.9 and no other CDS abnormality (cervicofacial vasogenic pathology, cervical tumors feed by ECA, carotid/vertebral artery stenosis) (b) ECA end dyastolic velocity > 21 cm/s To summarize CDS warnings during follow-up: (a) CDS should be performed once per year or whenever the patient clinical picture changes (b) a change in ECA RI # 0.1 over time should prompt additional investigation When, based on CDS findings, a DAVF is highly suspected or the status of a previous known DAVF changes, cerebral angiography should be performed. Images for this section: Page 35 of 39

36 Fig. 10: 42-year-old woman with left pulsatile tinnitus and bilateral occipital bruit. Carotid duplex sonography showed low resistance index (RI = 0.48) and high end diastolic velocity (EDV = 128 cm/s) in the left ECA (B) with an ICA to ECA RI ratio of 1.3 (C). Analyzes of the left Internal Jugular Vein (D) revealed a pseudoarterialized flow pattern and an increased flow velocity. Similar findings were recorded in the right side. [Patient E in Fig. 10 and Fig. 11] Page 36 of 39

37 Fig. 11: 42-year-old woman with left pulsatile tinnitus and bilateral occipital bruit (seen in fig. 10). Conventional angiography showed a high flow dural arteriovenous fistula, Borden 2, with feeders from the left occipital artery, posterior branches of the middle meningeal artery, neuromeningeal division of the ascending pharyngeal artery (A, left ECA, left lateral view), as well as participation from the right occipital artery (B, right ECA and ICA, right lateral view), and from dural branches of the right vertebral artery (C, right vertebral artery, right lateral view). Venous drainage showing early venous filling of the internal jugular vein and retrograde flow throw cortical veins (D, left ECA, anteroposterior view). Page 37 of 39

38 Conclusion DAVFs natural history is complex and they may present in a variety of ways. CTA and MRA can provide useful information for diagnosis, classification and treatment planning, but digital subtraction angiography remains the gold-standard study. Early screening of DAVF may prevent aggressive manifestations, especially in mild symptomatic patients. CDS can play a role as an initial screening and follow-up tool, considering the convenience, noninvasiveness, low cost and reproducibility of its results. Personal information 1 1 João Pedro Filipe, Tiago Parreira, Carlos Andrade 2,3 4 2,3,4, Rosa Santos, Elsa Azevedo 1 Department of Neuroradiology, Hospital de São João, Porto, Portugal; 2 Department of Neurology, Hospital de São João, Porto, Portugal; 3 Faculty of Medicine, Porto University, Porto, Portugal; 4 Neurosonology Unit, Department of Neurology, Hospital de São João, Porto, Portugal; References 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: 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: Page 38 of 39

39 Van Dijk JMC, terbrugge KG, Willinsky RA, et al. Clinical course of cranial dural arteriovenous fistulas with long-term persistent cortical venous reflux. Stroke J Cereb Circ. 2002;33(5): Tsai LK, Jeng JS, Wang HJ, et al. Diagnosis of intracranial dural arteriovenous fistulas by carotid duplex sonography. J Ultrasound Med 2004;23(6): Arning C, Grzyska U, Lachenmayer L. Duplex ultrasound of external carotid artery branches for the detection of dural arteriovenous fistulae. Rofo 2005; 177: Tsai L-K, Liu H-M, Lu C-J, et al. Carotid duplex sonography in the follow-up of intracranial dural arteriovenous fistulae. AJNR Am J Neuroradiol. 2005;26(3): Söderman M, Pavic L, Edner G, et al. Natural history of dural arteriovenous shunts. Stroke J Cereb Circ. 2008;39(6): Tsai LK, Jeng JS, Yip PK. Ultrasonography in intracranial dural arteriovenous fistula. J Med Ultrasound 2008;16: Tsai L-K, Yeh S-J, Chen Y-C, et al. Screen for intracranial dural arteriovenous fistulae with carotid duplex sonography. J Neurol Neurosurg Psychiatry. 2009;80(11): Yeh S-J, Tsai L-K, Jeng J-S. Clinical and carotid ultrasonographic features of intracranial dural arteriovenous fistulas in patients with and without Pulsatile Tinnitus. J Neuroimaging Off J Am Soc Neuroimaging. 2010;20(4): Gandhi D, Chen J, Pearl M, et al. Intracranial dural arteriovenous fistulas: classification, imaging findings, and treatment. AJNR Am J Neuroradiol. 2012;33(6): Santillan A, Nanaszko M, Burkhardt J-K, et al. Endovascular management of intracranial dural arteriovenous fistulas: a review. Clin Neurol Neurosurg. 2013;115(3): Page 39 of 39