Acta Neurol Belg DOI 10.1007/s13760-014-0387-7
ORIGINAL ARTICLE
Four-dimensional computed tomography angiography is valuable in intracranial dural arteriovenous fistula diagnosis and fistula evaluation Xianwang Ye • Haifeng Wang • Qiuli Huang Maoqing Jiang • Xiang Gao • Jie Zhang • Shengjun Zhou • Zhiqing Lin
•
Received: 8 June 2014 / Accepted: 23 October 2014 Ó Belgian Neurological Society 2014
Abstract This study was to evaluate the value of fourdimensional computed tomography angiography (4DCTA) in the diagnosis of intracranial dural arteriovenous fistula (DAVF). This study included 16 patients who were diagnosed to have intracranial DAVF by digital subtraction angiography (DSA). The 4D-CTA was performed by Aquilion ONE multi-detector CT scanner (Toshiba Medical Systems, Japan) equipped with 320 9 0.5 mm detector rows. Standard biplane fluoroscopy equipments (Infinix, Toshiba Medical Systems, Japan and ADVANTX LC/LP, GE Medical Systems, Milwaukee, WI, USA) were applied in the diagnosis of intra-arterial DSA. Examinations were performed to evaluate the findings of DSA and 4D-CTA in each patient. The examination results were read by two independent readers in a blind manner. All results were documented on standardized scoring sheets. In all 16 cases, the same diagnosis results of intracranial DAVF were obtained from DSA and 4D-CTA. The results of subtype (Borden and Cognard classification), venous reflux and fistula sites were also accurately exhibited in 4D-CTA. In addition, there was a little discrepancy in identifying
X. Ye and H. Wang contributed equally to this work. X. Ye Q. Huang (&) M. Jiang J. Zhang Department of Radiology, No. 1 Hospital of Ningbo, No. 59 Liuting Road, Haishu District, Ningbo 315010, Zhejiang, People’s Republic of China e-mail: [email protected]; [email protected] H. Wang Department of Medical College, Ningbo University, Ningbo, Zhejiang, People’s Republic of China X. Gao S. Zhou Z. Lin Department of Neurosurgery, No. 1 Hospital of Ningbo, Ningbo, Zhejiang, People’s Republic of China
smaller and specific arterial branches and in distinguishing fistula type (focal or diffuse) using 4D-CTA. Good-toexcellent agreements were made between 4D-CTA and DSA. Therefore, 4D-CTA could be a feasible tool for the characterization of intracranial DAVF, with respect to determining fistula site and venous drainage. Keywords Four-dimensional computed tomography angiography Digital subtraction angiography Intracranial dural arteriovenous fistula Fistula site
Introduction Intracranial dural arteriovenous fistula (intracranial DAVF) is an abnormal arteriovenous anastomosis between dural arteries and venous sinuses or cortical veins, which is generally responsible for approximately 10–15 % [12] of intracranial vascular malformations. DAVFs with retrograde cortical venous drainage are severe lesions, generally accompanied by intracranial hemorrhage, seizures, progressive neurologic deficits, intracranial hypertension or dementia. Current treatments for intracranial DAVF include endovascular embolization and surgical resection. The aim of these treatments is to obliterate arteriovenous shunt that is responsible for hemorrhage. Therefore, it is important to determine the lesion, and to identify the specific vascular angioarchitecture, especially the fistula site, in the diagnosis of DAVF. Digital subtraction angiography (DSA) has been recognized as a unique super-selective technique with superior resolution. DSA has always been the ‘‘gold standard’’ in determining cerebral vascular malformations, including intracranial DAVF [4, 5]. However, as an invasive inspection, DSA may cause unavoidable complications such as embolism and vascular injury
123
Acta Neurol Belg
[2]. Time-resolved magnetic resonance angiography (MRA) seems to be available in detecting DAVF [7, 11, 13, 15]. However, MRA is also limited by low resolution, restricted field of view (FOV) and saturation artifacts. Time-resolved MRA is less sensitive for slow-flow shunts and incapable to evaluate hemodynamics. The newly emerged dynamic computed tomography angiography (CTA) exhibited its potential value in the accurate diagnosis of brain arteriovenous malformation (bAVM) [6, 19, 21] and DAVF [1, 3, 8, 17, 20]. In this study, the materials of four-dimensional CTA (4D-CTA) are reviewed in 16 patients with intracranial DAVF, aiming to show the angioarchitecture of DAVF and the value of 4D-CTA.
rotation. Imaging was carried out according to the following procedures: first, 60 ml non-ionic contrast medium and 20 ml saline were intravenously administrated, followed by a dynamic acquisition sequence (22 dynamic volumes each protocol, 80 kVp, 120 mAs). The gantry rotation speed was 1 rotation per second (one mask volume, 80 kVp, 300 mAs). The mask volume was subtracted from the dynamic volumes. A total of 7,040 (22 dynamic volume 9 320 slices) images (temporal resolution was 3 frames per second and spatial resolution was 1024 9 1024) were stored as 22 DICOM files. Finally, with these results, time-resolved (arterial to venous) maximum intensity projection at different viewing angles could be generated with standard software in the scanner. These general procedures and parameters were as previously described [20].
Materials and methods DSA Patients From September 2010 to August 2012, 16 patients were enrolled in this study (Table 1). Inclusion criteria: previously untreated DAVF diagnosed by DSA. Exclusion criteria: ages less than 18 years; treatment for diabetes mellitus (cardiovascular risk may be increased); large intracranial hematoma pending emergent surgical treatment thus without DSA inspection results; allergy to iodinated contrast agents; impaired renal function (as indicated by baseline serum creatinine greater than 133 lmol/L); and inability to provide proper informed consent. The presented cases constituted a consecutive series, rather than hand-picked one. All results of 4D-CTA and DSA were blind-reviewed by two independent readers (one radiologist who did not manipulate the procedure of inspection and one neurosurgeon who was blind to the materials of these patients) in anonymous manner. The results were documented on the same scoring sheet. Several items were included in the scoring sheet, including fistula site, arterial supply, and venous drainage inspected by both 4D-CTA and DSA, as well as Borden classification and final treatment. All patients were treated by endovascular embolization or surgical resection. All procedures were approved by the Ethics Committee of the No. 1 Hospital of Ningbo. Informed consents were obtained from all patients or their families.
4D-CTA The 4D-CTA was performed by Aquilion ONE multidetector CT scanner (Toshiba Medical Systems, Japan) equipped with 320 9 0.5 mm detector rows. The whole brain could be covered for the volume of 16 cm per
123
Standard biplane fluoroscopy equipments (Infinix, Toshiba Medical Systems, Japan and ADVANTX LC/LP, GE Medical Systems, Milwaukee, WI, USA) were applied in the diagnosis of intra-arterial DSA. Seldinger technique was also involved. The 5-F catheter sheath and a 5-F duct were inserted with a single bend. With fluoroscopy and insoluble contrast medium (Ethiodol 350 mgl/ml, Savage Laboratories, Melville, NY, USA), angiography was performed for the aortic arch (20 ml/s, 25 ml), anteroposterior and bilateral internal carotid artery (ICA), external carotid artery (ECA; cervical segment 4 ml/s, 6 ml; intracranial 6 ml/s, 9 ml) and at least one vertebral artery (VA; cervical segment 4 ml/s, 6 ml; intracranial 6 ml/s, 9 ml) in turn. For each inspection, anteroposterior and lateral projections were obtained at 3 frames per second (matrix1024 9 1024, FOV of 2.5 Lp/cm).
Results Consistent borden classifications are achieved from DSA and 4D-CTA To determine Borden classification of DAVFs, all patients with DAVFs were examined with DSA and 4D-CTA in a blind manner. According to Borden classification for DAVFs, 4 cases were classified into Borden type II and the rest were classified into Borden type III. Full agreement could be achieved between the results of DSA and 4DCTA. Regarding feeding arteries, the number of major contributing arteries identified with 4D-CTA was C1 in all cases. However, for DSA, additional arterial feeders could be determined from internal maxillary artery (IMA) in 2 frontal cranial cases (patients 2 and 9). Extra feeding supply was confirmed from middle meningeal artery
55/M
11
Borden
(R) occipital dural
(L) parietal dural
(R) TS
(L) TS
(R) TS
(R) parietal dural
(R) frontal cranial base
(R) parietal dural (L) frontal cranial base
(L) middle cranial base
(L) TS (R) frontal cranial base
(R) frontal cranial base
(R) tentorial dural
(L) frontal cranial base
(R) frontal cranial base
(R) OCA
(L) MMA ? (L) OCA
(R) MMA ? (R) OCA
(L) OCA
(R) MMA ? (R) OCA
(L ? R) MMA ? (L ? R) PMA ? (L ? R) OCA
(R) OA
(L) OA ? (L) IMA
(R) MMA
(L) MMA
(L) MMA ? (L) OCA (R) OA
(R) OA
(R) MMA ? (R) OCA
(L) OA ? (L) IMA
(L) OA
SSS ? (L) TS
SSS
(R) SS
(L) SS
SSS ? (R) SS
SSS
SSS
SSS
SSS
(L) OV
(L) TS ? SSS SSS
SSS
SSS
SSS
SSS
III
III
II
II
III
III
II
III
III
II
II III
III
III
III
III
(R) occipital dural
(L) parietal dural
(R) TS
(L) TS
(R) TS
(R) parietal dural
(R) frontal cranial base
(R) parietal dural (L) frontal cranial base
(L) middle cranial base
(L) TS (R) frontal cranial base
(R) frontal cranial base
(R) tentorial dural
(L) frontal cranial base
(R) frontal cranial base
Fistula site
Venous drainage
Fistula site
Arterial supply
4D-CTA
DSA
(R) OCA
(L) OCA
(R) OCA
(L) OCA
(R) ECA
(L ? R) MMA ? (L ? R) PMA ? (L ? R) OCA
(R) OA
(L) OA
(R) MMA
(L) MMA
(L) MMA ? (L) OCA (R) OA
(R) OA
(R) MMA ? (R) OCA
(L) OA
(L) OA
Arterial supply
SSS ? (L) TS
SSS
(R) SS
(L) SS
SSS ? (R) SS
SSS
SSS
SSS
SSS
(L) OV
(L) TS ? SSS SSS
SSS
SSS
SSS
SSS
Venous drainage
III
III
II
II
III
III
III
III
III
II
II III
III
III
III
III
Borden
Resection
Embolization
Embolization
Embolization
embolization
Resection
Resection
Resection
Resection
Resection
Embolization Resection
Resection
Resection
Embolization
Resection
Treatment
a
Case 1;
b
Case 2; c Case 3
DAVF dural arteriovenous fistula, DSA digital subtraction angiography, 4D-CTA four-dimensional computed tomography angiography, CTA computed tomography angiography, IMA internal maxillary artery, ECA external carotid artery, MMA middle meningeal artery, PMA posterior meningeal artery, OA ophthalmic artery, OCA occipital artery, TS transverse sinus, SSS superior sagittal sinus, SS sigmoid sinus, L left, R right
38/M
70/M
10
16
67/F
9c
24/M
67/M
8
15
55/M
7
62/F
66/F 61/M
5b 6
14
61/M
4
47/F
61/M
3a
79/M
64/M
2
12
70/M
1
13
Age/ sex
No.
Table 1 Detailed information of 16 patients diagnosed as intracranial DAVF
Acta Neurol Belg
123
Acta Neurol Belg Fig. 1 A 61-year-old male with severe headache. a Plain head CT revealing a right occipital hemorrhage. b Lateral view of DSA after ECA injection, demonstrating DAVF at the level of right tentorial dural with cortical venous reflux (small white arrow). It is supplied by the occipital artery (large arrow head) and the middle meningeal artery (small arrow head). c 4DCTA identified right tentorial DAVF (Cognard type IV or Borden type III) with single point connection, draining though CVD into superior sagittal sinus. The fistula site and arterial feeders from occipital artery and middle meningeal artery were clear and identical to DSA
(MMA) in 2 cases (patients 14 and 15). MMA and occipital artery (OCA) could be distinguished from ECA in one case (patient 12). In these cases, large supplying arteries of DAVF could be correctly reflected by 4D-CTA. However, 4D-CTA failed to identify latent smaller feeders (Table 1). These data suggested that consistent Borden classifications were achieved from DSA and 4D-CTA. DSA and 4D-CTA can achieve the same location and type of fistula, which are identified by two individual readers To detect the location and type of fistula (focal or diffuse), DSA and 4D-CTA examinations were performed. Consistency has been achieved between the results of DSA and 4D-CTA, although the results were individually identified by each reader. The results of fistula sites were as follows: frontal cranial base (n = 6), middle cranial base (n = 1), occipital dural (n = 1), parietal dural (n = 3), tentorial dural (n = 1), and transverse sinus (n = 4) (Figs. 1, 2 and 3) (Table 1). However, two observers had disagreements in only 2 cases, in which the character of fistula could not be determined as focal or
123
diffuse. These data indicated that DSA and 4D-CTA achieved the same location and type of fistula, which were identified by two individual readers. Good outcome is obtained in all patients without further neurological defects To evaluate treatment outcomes, the results were blind reviewed by two independent readers. There was full agreement between the results from the two readers. Suitable treatment strategies were selected based on the clinical manifestations, current status of patients and subtype of DAVF lesions. In this study, there were 10 patients with fistula sites of frontal or middle cranial base and cerebral convex (Table 1). They were treated with surgical resection. Endovascular embolization was carried out in 6 patients with transverse sinus-based fistula sites. All patients presented totally blocked fistula or resection of deformed vessels 7 days after the surgery in CTA reexamination with a similar procedure to CTA before the surgery. These results demonstrated that good outcome was obtained in all patients without further neurological defects.
Acta Neurol Belg Fig. 2 Imaging of a 66-yearold woman presenting left-sided pulse-synchronous tinnitus. a Lateral DSA projection after left ECA injection, demonstrating DAVF (Borden type II and Cognard type IIa ? b) at the level of the left transverse sinus (large hollow arrow) with cortical venous reflux (small white arrow). The lesion is mainly supplied by branches of the occipital artery (large arrow head) and to a lesser degree by a branch from the middle meningeal artery (small arrow head). b Lateral view of 4D-CTA showing diffuse fistula involving the transverse sinus. The venous drainage was distributed earlier into the left transverse sinus (large hollow arrow) and sagittal sinus. Compared with DSA, 4D-CTA is in agreement in terms of the fistula site and subtype of DAVF, as well as to its dominant feeder from the occipital artery (large arrow head). However, the smaller feeder from the middle meningeal artery failed to be recorded
Discussion There may be complications during the inspection with DSA. However, DSA is still the first choice in the diagnosis of intracranial DAVF. The newly emerged time-resolved techniques, 4D-CTA and 4D-MRA, enabled additional dynamic methods to investigate vascular abnormality. Willems et al. [20] studied and compared the images of DAVF between DSA and 4D-CTA, indicating that full agreement was reached in 10 out of 11 cases in terms of Borden classification. Fujiwara et al. [8] considered 320-detector row CT as a valuable non-invasive tool for the diagnosis and grading of cerebral DAVF, regarding the determination, location, and subtype of fistula. In addition, its results were highly correlated with DSA. Reinacher and co-workers [11] indicated that DAVF could be correctly graded and classified by time-resolved MRA, although some small feeders might be overlooked. Evans et al. [6] proved that dynamic MRA was an alternative method for identifying shunting vascular abnormalities such as bAVM and DAVF. Compared with on-dynamic CT(A) and
MR(A), 4D-CTA can significantly increase diagnosis information. For the 16 cases in the present study, the results of all lesions detected by 4D-CTA were highly consistent with those by DSA (Table 1). Larger arterial feeders and venous outflow were displayed clearly while only a few small arteries failed to be examined (No. 2, 9, 12, and 14). Therefore, it seemed that the diagnostic value of 4D-CTA for intracranial DAVF was very similar to that of DSA. Intracranial DAVF can be treated by endovascular embolization or surgical resection [9, 10, 14, 16]. The treatment of DAVF aims to completely eliminate arteriovenous shunt. On the contrary, incomplete treatment results in the recruitment of collateral vessels and increases persistent risk of hemorrhage [18]. Therefore, pre-operative inspection should be able to clearly identify not only arterial feeders and venous drainage, but also the sites of fistula. For the cases in this study, fistula sites and venous drainages were successfully detected by 4D-CTA. The accuracy of 4D-CTA in reflecting fistula point has been
123
Acta Neurol Belg Fig. 3 Images of a 67-year-old female with one-week history of severe headache. a Axial nonenhanced CT showing an intraparenchymal lesion in the left frontal lobe. b Diagnostic right lateral ICA angiogram reflecting anterior cranial fossa DAVF. This fistula is supplied by left ophthalmic artery via the left ethmoidal artery. Venous reflux goes though frontal cortical vein into superior sagittal sinus. Angiogram reveals another contribution of the left internal maxillary artery (large arrow head). c 4D-CTA detected the major contribution of the left ethmoidal artery from left OA, fistula site, DAVF subtype (Cognard type IV or Borden type III) and drainage pattern correctly but failed to record branches of the left internal maxillary artery as another contribution (coronal projection in C). Despite such disadvantages, it did not affect surgical resection
demonstrated. It is facile for CTA to detect fistula site with a single point connection. For DAVF in cranial base, smaller feeding arteries may be overlooked by 4D-CTA, but anatomic detail of fistula site and venous pattern could be precisely delineated. For DAVF involving venous sinus such as transverse or sigmoid sinus, it is hard to identify all arterial branches due to the interference of complex vascular structures. However, the fistula sites, the larger feeding arteries and retrograde venous flow could be well identified, even for a diffuse fistula. All cases in this study were classified into high-grade DAVF. The diagnostic value of 4D-CTA could not be determined for patients in Cognard I and IIa. Beijer [1] reported that distinct patterns of venous drainage could be differentiated and subtypes of DAVF could be identified by 4D-CTA. In our study, the classification results of 4D-CTA in all cases were the same as those of DSA inspection. It would be conductive to determine treatment strategies. However, there are some limitations in this study. The study was performed in a single center with a small study population. All cases were patients with high-grade DAVF and some DAVFs (Borden type I) were missed in our study group. In addition, DAVF was diagnosed on all patients after examination for cerebral hemorrhage, after which emergency surgery such as interventional surgery or
123
craniotomy was performed immediately. Therefore, disease progression afterward cannot be observed. In conclusion, 4D-CTA and DSA showed good agreement between each other. The findings suggest that 4DCTA could be a complementary tool for characterizing intracranial DAVF on fistula site and venous drainage, while 4D-CTA could be applied in the primary diagnosis of intracranial DAVF. However, further technical advances are required to improve the visualization and imaging of arterial feeders. Acknowledgments This work was supported by Ningbo Social Developmental Research Project (No. 2012C50027). Conflict of interest All authors declare no financial competing interests. All authors declare no non-financial competing interests.
References 1. Beijer TR, van Dijk EJ, de Vries J, Vermeer SE, Prokop M, Meijer FJ (2013) 4D-CT angiography differentiating arteriovenous fistula subtypes. Clin Neurol Neurosurg 115:1313–1316 2. Bendszus M, Koltzenburg M, Burger R, Warmuth-Metz M, Hofmann E, Solymosi L (1999) Silent embolism in diagnostic cerebral angiography and neurointerventional procedures: a prospective study. Lancet 354:1594–1597 3. Brouwer PA, Bosman T, van Walderveen MA, Krings T, Leroux AA, Willems PW (2010) Dynamic 320-section CT angiography
Acta Neurol Belg
4.
5.
6.
7.
8.
9.
10.
11.
12.
in cranial arteriovenous shunting lesions. AJNR Am J Neuroradiol 31:767–770 Brown RD Jr, Flemming KD, Meyer FB, Cloft HJ, Pollock BE, Link ML (2005) Natural history, evaluation, and management of intracranial vascular malformations. Mayo Clin Proc 80:269–281 Endo S, Kuwayama N, Takaku A, Nishijima M (1998) Direct packing of the isolated sinus in patients with dural arteriovenous fistulas of the transverse-sigmoid sinus. J Neurosurg 88:449–456 Evans AL, Coley SC, Wilkinson ID, Griffiths PD (2005) Firstline investigation of acute intracerebral hemorrhage using dynamic magnetic resonance angiography. Acta Radiol 46:625–630 Farb RI, Agid R, Willinsky RA, Johnstone DM, Terbrugge KG (2009) Cranial dural arteriovenous fistula: diagnosis and classification with time-resolved MR angiography at 3T. AJNR Am J Neuroradiol 30:1546–1551 Fujiwara H, Momoshima S, Akiyama T, Kuribayashi S (2013) Whole-brain CT digital subtraction angiography of cerebral dural arteriovenous fistula using 320-detector row CT. Neuroradiology 55:837–843 Hatano T, Bozinov O, Burkhardt JK, Bertalanffy H (2013) Surgical treatment of tentorial dural arteriovenous fistulae located around the tentorial incisura. Neurosurg Rev 36:429–435 Houdart E, Saint-Maurice JP, Chapot R, Ditchfield A, Blanquet A, Lot G, Merland JJ (2002) Transcranial approach for venous embolization of dural arteriovenous fistulas. J Neurosurg 97:280–286 Kwon BJ, Han MH, Kang HS, Chang KH (2005) MR imaging findings of intracranial dural arteriovenous fistulas: relations with venous drainage patterns. AJNR Am J Neuroradiol 26:2500–2507 Lucas CP, Zabramski JM, Spetzler RF, Nishimura S, Jacobowitz R (1997) Treatment for intracranial dural arteriovenous malformations: a meta-analysis from the English language literature. Neurosurgery 40(6):1119–1130
13. Nishimura S, Hirai T, Sasao A, Kitajima M, Morioka M, Kai Y, Omori Y, Okuda T, Murakami R, Fukuoka H, Awai K, Kuratsu JI, Yamashita Y (2010) Evaluation of dural arteriovenous fistulas with 4D contrast-enhanced MR angiography at 3T. AJNR Am J Neuroradiol 31:80–85 14. Pradilla G, Coon AL, Huang J, Tamargo RJ (2012) Surgical treatment of cranial arteriovenous malformations and dural arteriovenous fistulas. Neurosurg Clin N Am 23:105–122 15. Reinacher PC, Stracke P, Reinges MH, Hans FJ, Krings T (2007) Contrast-enhanced time-resolved 3-D MRA: applications in neurosurgery and interventional neuroradiology. Neuroradiology 49(Suppl 1):S3–S13 16. Salomon EJ, Barfett J, Willems PW, Geibprasert S, Bacigaluppi S, Krings T (2009) Dynamic CT angiography and CT perfusion employing a 320-detector row CT: protocol and current clinical applications. Klin Neuroradiol 19:187–196 17. Siebert E, Bohner G, Dewey M, Masuhr F, Hoffmann KT, Mews J, Engelken F, Bauknecht HC, Diekmann S, Klingebiel R (2009) 320-slice CT neuroimaging: initial clinical experience and image quality evaluation. Br J Radiol 82:561–570 18. Wachter D, Hans F, Psychogios MN, Knauth M, Rohde V (2011) Microsurgery can cure most intracranial dural arteriovenous fistulae of the sinus and non-sinus type. Neurosurg Rev 34(3):337–345 19. Wang H, Ye X, Gao X, Zhou S, Lin Z (2014) The diagnosis of arteriovenous malformations by 4D-CTA: a clinical study. J Neuroradiol 41:117–123 20. Willems PW, Brouwer PA, Barfett JJ, terBrugge KG, Krings T (2011) Detection and classification of cranial dural arteriovenous fistulas using 4D-CT angiography: initial experience. AJNR Am J Neuroradiol 32:49–53 21. Willems PW, Taeshineetanakul P, Schenk B, Brouwer PA, Terbrugge KG, Krings T (2012) The use of 4D-CTA in the diagnostic work-up of brain arteriovenous malformations. Neuroradiology 54:123–131
123
ORIGINAL ARTICLE
Four-dimensional computed tomography angiography is valuable in intracranial dural arteriovenous fistula diagnosis and fistula evaluation Xianwang Ye • Haifeng Wang • Qiuli Huang Maoqing Jiang • Xiang Gao • Jie Zhang • Shengjun Zhou • Zhiqing Lin
•
Received: 8 June 2014 / Accepted: 23 October 2014 Ó Belgian Neurological Society 2014
Abstract This study was to evaluate the value of fourdimensional computed tomography angiography (4DCTA) in the diagnosis of intracranial dural arteriovenous fistula (DAVF). This study included 16 patients who were diagnosed to have intracranial DAVF by digital subtraction angiography (DSA). The 4D-CTA was performed by Aquilion ONE multi-detector CT scanner (Toshiba Medical Systems, Japan) equipped with 320 9 0.5 mm detector rows. Standard biplane fluoroscopy equipments (Infinix, Toshiba Medical Systems, Japan and ADVANTX LC/LP, GE Medical Systems, Milwaukee, WI, USA) were applied in the diagnosis of intra-arterial DSA. Examinations were performed to evaluate the findings of DSA and 4D-CTA in each patient. The examination results were read by two independent readers in a blind manner. All results were documented on standardized scoring sheets. In all 16 cases, the same diagnosis results of intracranial DAVF were obtained from DSA and 4D-CTA. The results of subtype (Borden and Cognard classification), venous reflux and fistula sites were also accurately exhibited in 4D-CTA. In addition, there was a little discrepancy in identifying
X. Ye and H. Wang contributed equally to this work. X. Ye Q. Huang (&) M. Jiang J. Zhang Department of Radiology, No. 1 Hospital of Ningbo, No. 59 Liuting Road, Haishu District, Ningbo 315010, Zhejiang, People’s Republic of China e-mail: [email protected]; [email protected] H. Wang Department of Medical College, Ningbo University, Ningbo, Zhejiang, People’s Republic of China X. Gao S. Zhou Z. Lin Department of Neurosurgery, No. 1 Hospital of Ningbo, Ningbo, Zhejiang, People’s Republic of China
smaller and specific arterial branches and in distinguishing fistula type (focal or diffuse) using 4D-CTA. Good-toexcellent agreements were made between 4D-CTA and DSA. Therefore, 4D-CTA could be a feasible tool for the characterization of intracranial DAVF, with respect to determining fistula site and venous drainage. Keywords Four-dimensional computed tomography angiography Digital subtraction angiography Intracranial dural arteriovenous fistula Fistula site
Introduction Intracranial dural arteriovenous fistula (intracranial DAVF) is an abnormal arteriovenous anastomosis between dural arteries and venous sinuses or cortical veins, which is generally responsible for approximately 10–15 % [12] of intracranial vascular malformations. DAVFs with retrograde cortical venous drainage are severe lesions, generally accompanied by intracranial hemorrhage, seizures, progressive neurologic deficits, intracranial hypertension or dementia. Current treatments for intracranial DAVF include endovascular embolization and surgical resection. The aim of these treatments is to obliterate arteriovenous shunt that is responsible for hemorrhage. Therefore, it is important to determine the lesion, and to identify the specific vascular angioarchitecture, especially the fistula site, in the diagnosis of DAVF. Digital subtraction angiography (DSA) has been recognized as a unique super-selective technique with superior resolution. DSA has always been the ‘‘gold standard’’ in determining cerebral vascular malformations, including intracranial DAVF [4, 5]. However, as an invasive inspection, DSA may cause unavoidable complications such as embolism and vascular injury
123
Acta Neurol Belg
[2]. Time-resolved magnetic resonance angiography (MRA) seems to be available in detecting DAVF [7, 11, 13, 15]. However, MRA is also limited by low resolution, restricted field of view (FOV) and saturation artifacts. Time-resolved MRA is less sensitive for slow-flow shunts and incapable to evaluate hemodynamics. The newly emerged dynamic computed tomography angiography (CTA) exhibited its potential value in the accurate diagnosis of brain arteriovenous malformation (bAVM) [6, 19, 21] and DAVF [1, 3, 8, 17, 20]. In this study, the materials of four-dimensional CTA (4D-CTA) are reviewed in 16 patients with intracranial DAVF, aiming to show the angioarchitecture of DAVF and the value of 4D-CTA.
rotation. Imaging was carried out according to the following procedures: first, 60 ml non-ionic contrast medium and 20 ml saline were intravenously administrated, followed by a dynamic acquisition sequence (22 dynamic volumes each protocol, 80 kVp, 120 mAs). The gantry rotation speed was 1 rotation per second (one mask volume, 80 kVp, 300 mAs). The mask volume was subtracted from the dynamic volumes. A total of 7,040 (22 dynamic volume 9 320 slices) images (temporal resolution was 3 frames per second and spatial resolution was 1024 9 1024) were stored as 22 DICOM files. Finally, with these results, time-resolved (arterial to venous) maximum intensity projection at different viewing angles could be generated with standard software in the scanner. These general procedures and parameters were as previously described [20].
Materials and methods DSA Patients From September 2010 to August 2012, 16 patients were enrolled in this study (Table 1). Inclusion criteria: previously untreated DAVF diagnosed by DSA. Exclusion criteria: ages less than 18 years; treatment for diabetes mellitus (cardiovascular risk may be increased); large intracranial hematoma pending emergent surgical treatment thus without DSA inspection results; allergy to iodinated contrast agents; impaired renal function (as indicated by baseline serum creatinine greater than 133 lmol/L); and inability to provide proper informed consent. The presented cases constituted a consecutive series, rather than hand-picked one. All results of 4D-CTA and DSA were blind-reviewed by two independent readers (one radiologist who did not manipulate the procedure of inspection and one neurosurgeon who was blind to the materials of these patients) in anonymous manner. The results were documented on the same scoring sheet. Several items were included in the scoring sheet, including fistula site, arterial supply, and venous drainage inspected by both 4D-CTA and DSA, as well as Borden classification and final treatment. All patients were treated by endovascular embolization or surgical resection. All procedures were approved by the Ethics Committee of the No. 1 Hospital of Ningbo. Informed consents were obtained from all patients or their families.
4D-CTA The 4D-CTA was performed by Aquilion ONE multidetector CT scanner (Toshiba Medical Systems, Japan) equipped with 320 9 0.5 mm detector rows. The whole brain could be covered for the volume of 16 cm per
123
Standard biplane fluoroscopy equipments (Infinix, Toshiba Medical Systems, Japan and ADVANTX LC/LP, GE Medical Systems, Milwaukee, WI, USA) were applied in the diagnosis of intra-arterial DSA. Seldinger technique was also involved. The 5-F catheter sheath and a 5-F duct were inserted with a single bend. With fluoroscopy and insoluble contrast medium (Ethiodol 350 mgl/ml, Savage Laboratories, Melville, NY, USA), angiography was performed for the aortic arch (20 ml/s, 25 ml), anteroposterior and bilateral internal carotid artery (ICA), external carotid artery (ECA; cervical segment 4 ml/s, 6 ml; intracranial 6 ml/s, 9 ml) and at least one vertebral artery (VA; cervical segment 4 ml/s, 6 ml; intracranial 6 ml/s, 9 ml) in turn. For each inspection, anteroposterior and lateral projections were obtained at 3 frames per second (matrix1024 9 1024, FOV of 2.5 Lp/cm).
Results Consistent borden classifications are achieved from DSA and 4D-CTA To determine Borden classification of DAVFs, all patients with DAVFs were examined with DSA and 4D-CTA in a blind manner. According to Borden classification for DAVFs, 4 cases were classified into Borden type II and the rest were classified into Borden type III. Full agreement could be achieved between the results of DSA and 4DCTA. Regarding feeding arteries, the number of major contributing arteries identified with 4D-CTA was C1 in all cases. However, for DSA, additional arterial feeders could be determined from internal maxillary artery (IMA) in 2 frontal cranial cases (patients 2 and 9). Extra feeding supply was confirmed from middle meningeal artery
55/M
11
Borden
(R) occipital dural
(L) parietal dural
(R) TS
(L) TS
(R) TS
(R) parietal dural
(R) frontal cranial base
(R) parietal dural (L) frontal cranial base
(L) middle cranial base
(L) TS (R) frontal cranial base
(R) frontal cranial base
(R) tentorial dural
(L) frontal cranial base
(R) frontal cranial base
(R) OCA
(L) MMA ? (L) OCA
(R) MMA ? (R) OCA
(L) OCA
(R) MMA ? (R) OCA
(L ? R) MMA ? (L ? R) PMA ? (L ? R) OCA
(R) OA
(L) OA ? (L) IMA
(R) MMA
(L) MMA
(L) MMA ? (L) OCA (R) OA
(R) OA
(R) MMA ? (R) OCA
(L) OA ? (L) IMA
(L) OA
SSS ? (L) TS
SSS
(R) SS
(L) SS
SSS ? (R) SS
SSS
SSS
SSS
SSS
(L) OV
(L) TS ? SSS SSS
SSS
SSS
SSS
SSS
III
III
II
II
III
III
II
III
III
II
II III
III
III
III
III
(R) occipital dural
(L) parietal dural
(R) TS
(L) TS
(R) TS
(R) parietal dural
(R) frontal cranial base
(R) parietal dural (L) frontal cranial base
(L) middle cranial base
(L) TS (R) frontal cranial base
(R) frontal cranial base
(R) tentorial dural
(L) frontal cranial base
(R) frontal cranial base
Fistula site
Venous drainage
Fistula site
Arterial supply
4D-CTA
DSA
(R) OCA
(L) OCA
(R) OCA
(L) OCA
(R) ECA
(L ? R) MMA ? (L ? R) PMA ? (L ? R) OCA
(R) OA
(L) OA
(R) MMA
(L) MMA
(L) MMA ? (L) OCA (R) OA
(R) OA
(R) MMA ? (R) OCA
(L) OA
(L) OA
Arterial supply
SSS ? (L) TS
SSS
(R) SS
(L) SS
SSS ? (R) SS
SSS
SSS
SSS
SSS
(L) OV
(L) TS ? SSS SSS
SSS
SSS
SSS
SSS
Venous drainage
III
III
II
II
III
III
III
III
III
II
II III
III
III
III
III
Borden
Resection
Embolization
Embolization
Embolization
embolization
Resection
Resection
Resection
Resection
Resection
Embolization Resection
Resection
Resection
Embolization
Resection
Treatment
a
Case 1;
b
Case 2; c Case 3
DAVF dural arteriovenous fistula, DSA digital subtraction angiography, 4D-CTA four-dimensional computed tomography angiography, CTA computed tomography angiography, IMA internal maxillary artery, ECA external carotid artery, MMA middle meningeal artery, PMA posterior meningeal artery, OA ophthalmic artery, OCA occipital artery, TS transverse sinus, SSS superior sagittal sinus, SS sigmoid sinus, L left, R right
38/M
70/M
10
16
67/F
9c
24/M
67/M
8
15
55/M
7
62/F
66/F 61/M
5b 6
14
61/M
4
47/F
61/M
3a
79/M
64/M
2
12
70/M
1
13
Age/ sex
No.
Table 1 Detailed information of 16 patients diagnosed as intracranial DAVF
Acta Neurol Belg
123
Acta Neurol Belg Fig. 1 A 61-year-old male with severe headache. a Plain head CT revealing a right occipital hemorrhage. b Lateral view of DSA after ECA injection, demonstrating DAVF at the level of right tentorial dural with cortical venous reflux (small white arrow). It is supplied by the occipital artery (large arrow head) and the middle meningeal artery (small arrow head). c 4DCTA identified right tentorial DAVF (Cognard type IV or Borden type III) with single point connection, draining though CVD into superior sagittal sinus. The fistula site and arterial feeders from occipital artery and middle meningeal artery were clear and identical to DSA
(MMA) in 2 cases (patients 14 and 15). MMA and occipital artery (OCA) could be distinguished from ECA in one case (patient 12). In these cases, large supplying arteries of DAVF could be correctly reflected by 4D-CTA. However, 4D-CTA failed to identify latent smaller feeders (Table 1). These data suggested that consistent Borden classifications were achieved from DSA and 4D-CTA. DSA and 4D-CTA can achieve the same location and type of fistula, which are identified by two individual readers To detect the location and type of fistula (focal or diffuse), DSA and 4D-CTA examinations were performed. Consistency has been achieved between the results of DSA and 4D-CTA, although the results were individually identified by each reader. The results of fistula sites were as follows: frontal cranial base (n = 6), middle cranial base (n = 1), occipital dural (n = 1), parietal dural (n = 3), tentorial dural (n = 1), and transverse sinus (n = 4) (Figs. 1, 2 and 3) (Table 1). However, two observers had disagreements in only 2 cases, in which the character of fistula could not be determined as focal or
123
diffuse. These data indicated that DSA and 4D-CTA achieved the same location and type of fistula, which were identified by two individual readers. Good outcome is obtained in all patients without further neurological defects To evaluate treatment outcomes, the results were blind reviewed by two independent readers. There was full agreement between the results from the two readers. Suitable treatment strategies were selected based on the clinical manifestations, current status of patients and subtype of DAVF lesions. In this study, there were 10 patients with fistula sites of frontal or middle cranial base and cerebral convex (Table 1). They were treated with surgical resection. Endovascular embolization was carried out in 6 patients with transverse sinus-based fistula sites. All patients presented totally blocked fistula or resection of deformed vessels 7 days after the surgery in CTA reexamination with a similar procedure to CTA before the surgery. These results demonstrated that good outcome was obtained in all patients without further neurological defects.
Acta Neurol Belg Fig. 2 Imaging of a 66-yearold woman presenting left-sided pulse-synchronous tinnitus. a Lateral DSA projection after left ECA injection, demonstrating DAVF (Borden type II and Cognard type IIa ? b) at the level of the left transverse sinus (large hollow arrow) with cortical venous reflux (small white arrow). The lesion is mainly supplied by branches of the occipital artery (large arrow head) and to a lesser degree by a branch from the middle meningeal artery (small arrow head). b Lateral view of 4D-CTA showing diffuse fistula involving the transverse sinus. The venous drainage was distributed earlier into the left transverse sinus (large hollow arrow) and sagittal sinus. Compared with DSA, 4D-CTA is in agreement in terms of the fistula site and subtype of DAVF, as well as to its dominant feeder from the occipital artery (large arrow head). However, the smaller feeder from the middle meningeal artery failed to be recorded
Discussion There may be complications during the inspection with DSA. However, DSA is still the first choice in the diagnosis of intracranial DAVF. The newly emerged time-resolved techniques, 4D-CTA and 4D-MRA, enabled additional dynamic methods to investigate vascular abnormality. Willems et al. [20] studied and compared the images of DAVF between DSA and 4D-CTA, indicating that full agreement was reached in 10 out of 11 cases in terms of Borden classification. Fujiwara et al. [8] considered 320-detector row CT as a valuable non-invasive tool for the diagnosis and grading of cerebral DAVF, regarding the determination, location, and subtype of fistula. In addition, its results were highly correlated with DSA. Reinacher and co-workers [11] indicated that DAVF could be correctly graded and classified by time-resolved MRA, although some small feeders might be overlooked. Evans et al. [6] proved that dynamic MRA was an alternative method for identifying shunting vascular abnormalities such as bAVM and DAVF. Compared with on-dynamic CT(A) and
MR(A), 4D-CTA can significantly increase diagnosis information. For the 16 cases in the present study, the results of all lesions detected by 4D-CTA were highly consistent with those by DSA (Table 1). Larger arterial feeders and venous outflow were displayed clearly while only a few small arteries failed to be examined (No. 2, 9, 12, and 14). Therefore, it seemed that the diagnostic value of 4D-CTA for intracranial DAVF was very similar to that of DSA. Intracranial DAVF can be treated by endovascular embolization or surgical resection [9, 10, 14, 16]. The treatment of DAVF aims to completely eliminate arteriovenous shunt. On the contrary, incomplete treatment results in the recruitment of collateral vessels and increases persistent risk of hemorrhage [18]. Therefore, pre-operative inspection should be able to clearly identify not only arterial feeders and venous drainage, but also the sites of fistula. For the cases in this study, fistula sites and venous drainages were successfully detected by 4D-CTA. The accuracy of 4D-CTA in reflecting fistula point has been
123
Acta Neurol Belg Fig. 3 Images of a 67-year-old female with one-week history of severe headache. a Axial nonenhanced CT showing an intraparenchymal lesion in the left frontal lobe. b Diagnostic right lateral ICA angiogram reflecting anterior cranial fossa DAVF. This fistula is supplied by left ophthalmic artery via the left ethmoidal artery. Venous reflux goes though frontal cortical vein into superior sagittal sinus. Angiogram reveals another contribution of the left internal maxillary artery (large arrow head). c 4D-CTA detected the major contribution of the left ethmoidal artery from left OA, fistula site, DAVF subtype (Cognard type IV or Borden type III) and drainage pattern correctly but failed to record branches of the left internal maxillary artery as another contribution (coronal projection in C). Despite such disadvantages, it did not affect surgical resection
demonstrated. It is facile for CTA to detect fistula site with a single point connection. For DAVF in cranial base, smaller feeding arteries may be overlooked by 4D-CTA, but anatomic detail of fistula site and venous pattern could be precisely delineated. For DAVF involving venous sinus such as transverse or sigmoid sinus, it is hard to identify all arterial branches due to the interference of complex vascular structures. However, the fistula sites, the larger feeding arteries and retrograde venous flow could be well identified, even for a diffuse fistula. All cases in this study were classified into high-grade DAVF. The diagnostic value of 4D-CTA could not be determined for patients in Cognard I and IIa. Beijer [1] reported that distinct patterns of venous drainage could be differentiated and subtypes of DAVF could be identified by 4D-CTA. In our study, the classification results of 4D-CTA in all cases were the same as those of DSA inspection. It would be conductive to determine treatment strategies. However, there are some limitations in this study. The study was performed in a single center with a small study population. All cases were patients with high-grade DAVF and some DAVFs (Borden type I) were missed in our study group. In addition, DAVF was diagnosed on all patients after examination for cerebral hemorrhage, after which emergency surgery such as interventional surgery or
123
craniotomy was performed immediately. Therefore, disease progression afterward cannot be observed. In conclusion, 4D-CTA and DSA showed good agreement between each other. The findings suggest that 4DCTA could be a complementary tool for characterizing intracranial DAVF on fistula site and venous drainage, while 4D-CTA could be applied in the primary diagnosis of intracranial DAVF. However, further technical advances are required to improve the visualization and imaging of arterial feeders. Acknowledgments This work was supported by Ningbo Social Developmental Research Project (No. 2012C50027). Conflict of interest All authors declare no financial competing interests. All authors declare no non-financial competing interests.
References 1. Beijer TR, van Dijk EJ, de Vries J, Vermeer SE, Prokop M, Meijer FJ (2013) 4D-CT angiography differentiating arteriovenous fistula subtypes. Clin Neurol Neurosurg 115:1313–1316 2. Bendszus M, Koltzenburg M, Burger R, Warmuth-Metz M, Hofmann E, Solymosi L (1999) Silent embolism in diagnostic cerebral angiography and neurointerventional procedures: a prospective study. Lancet 354:1594–1597 3. Brouwer PA, Bosman T, van Walderveen MA, Krings T, Leroux AA, Willems PW (2010) Dynamic 320-section CT angiography
Acta Neurol Belg
4.
5.
6.
7.
8.
9.
10.
11.
12.
in cranial arteriovenous shunting lesions. AJNR Am J Neuroradiol 31:767–770 Brown RD Jr, Flemming KD, Meyer FB, Cloft HJ, Pollock BE, Link ML (2005) Natural history, evaluation, and management of intracranial vascular malformations. Mayo Clin Proc 80:269–281 Endo S, Kuwayama N, Takaku A, Nishijima M (1998) Direct packing of the isolated sinus in patients with dural arteriovenous fistulas of the transverse-sigmoid sinus. J Neurosurg 88:449–456 Evans AL, Coley SC, Wilkinson ID, Griffiths PD (2005) Firstline investigation of acute intracerebral hemorrhage using dynamic magnetic resonance angiography. Acta Radiol 46:625–630 Farb RI, Agid R, Willinsky RA, Johnstone DM, Terbrugge KG (2009) Cranial dural arteriovenous fistula: diagnosis and classification with time-resolved MR angiography at 3T. AJNR Am J Neuroradiol 30:1546–1551 Fujiwara H, Momoshima S, Akiyama T, Kuribayashi S (2013) Whole-brain CT digital subtraction angiography of cerebral dural arteriovenous fistula using 320-detector row CT. Neuroradiology 55:837–843 Hatano T, Bozinov O, Burkhardt JK, Bertalanffy H (2013) Surgical treatment of tentorial dural arteriovenous fistulae located around the tentorial incisura. Neurosurg Rev 36:429–435 Houdart E, Saint-Maurice JP, Chapot R, Ditchfield A, Blanquet A, Lot G, Merland JJ (2002) Transcranial approach for venous embolization of dural arteriovenous fistulas. J Neurosurg 97:280–286 Kwon BJ, Han MH, Kang HS, Chang KH (2005) MR imaging findings of intracranial dural arteriovenous fistulas: relations with venous drainage patterns. AJNR Am J Neuroradiol 26:2500–2507 Lucas CP, Zabramski JM, Spetzler RF, Nishimura S, Jacobowitz R (1997) Treatment for intracranial dural arteriovenous malformations: a meta-analysis from the English language literature. Neurosurgery 40(6):1119–1130
13. Nishimura S, Hirai T, Sasao A, Kitajima M, Morioka M, Kai Y, Omori Y, Okuda T, Murakami R, Fukuoka H, Awai K, Kuratsu JI, Yamashita Y (2010) Evaluation of dural arteriovenous fistulas with 4D contrast-enhanced MR angiography at 3T. AJNR Am J Neuroradiol 31:80–85 14. Pradilla G, Coon AL, Huang J, Tamargo RJ (2012) Surgical treatment of cranial arteriovenous malformations and dural arteriovenous fistulas. Neurosurg Clin N Am 23:105–122 15. Reinacher PC, Stracke P, Reinges MH, Hans FJ, Krings T (2007) Contrast-enhanced time-resolved 3-D MRA: applications in neurosurgery and interventional neuroradiology. Neuroradiology 49(Suppl 1):S3–S13 16. Salomon EJ, Barfett J, Willems PW, Geibprasert S, Bacigaluppi S, Krings T (2009) Dynamic CT angiography and CT perfusion employing a 320-detector row CT: protocol and current clinical applications. Klin Neuroradiol 19:187–196 17. Siebert E, Bohner G, Dewey M, Masuhr F, Hoffmann KT, Mews J, Engelken F, Bauknecht HC, Diekmann S, Klingebiel R (2009) 320-slice CT neuroimaging: initial clinical experience and image quality evaluation. Br J Radiol 82:561–570 18. Wachter D, Hans F, Psychogios MN, Knauth M, Rohde V (2011) Microsurgery can cure most intracranial dural arteriovenous fistulae of the sinus and non-sinus type. Neurosurg Rev 34(3):337–345 19. Wang H, Ye X, Gao X, Zhou S, Lin Z (2014) The diagnosis of arteriovenous malformations by 4D-CTA: a clinical study. J Neuroradiol 41:117–123 20. Willems PW, Brouwer PA, Barfett JJ, terBrugge KG, Krings T (2011) Detection and classification of cranial dural arteriovenous fistulas using 4D-CT angiography: initial experience. AJNR Am J Neuroradiol 32:49–53 21. Willems PW, Taeshineetanakul P, Schenk B, Brouwer PA, Terbrugge KG, Krings T (2012) The use of 4D-CTA in the diagnostic work-up of brain arteriovenous malformations. Neuroradiology 54:123–131
123
