- Email: [email protected]
Pitfalls in the preoperative evaluation of subarachnoid hemorrhage without digital subtraction angiography: report on 2 cases
Surgical Neurology 68 (2007) 344 – 348 www.surgicalneurology-online.com
Imaging
Pitfalls in the preoperative evaluation of subarachnoid hemorrhage without digital subtraction angiography: report on 2 cases Yuji Hashimoto, MD4, Sonen Kin, MD, Koichi Haraguchi, MD, Jun Niwa, MD Department of Neurosurgery, Hakodate Municipal Hospital, Hokkaido 041-8680, Japan Received 9 February 2006; accepted 6 October 2006
Abstract
Background: Digital subtraction angiography has been used in the diagnosis of aneurysmal SAH and as a preoperative imaging method. However, new methods such as MRA and CTA are now deemed by many institutions to provide sufficient information to allow surgery to go ahead without a preliminary DSA scan. We report on 2 cases of SAH in which there were additional lesions that were difficult to evaluate because of the lack of DSA information. Case Descriptions: The fist patient demonstrated SAH with IVH. Computed tomographic angiography revealed an ACoA aneurysm with a bleb. We first thought that the SAH and IVH were both caused by a ruptured ACoA aneurysm but noted that hemorrhage pattern was inconsistent with the location and orientation of the aneurysm. A DSA scan revealed a dural arteriovenous fistula in the region of the craniocervical junction, supplied by the right occipital artery. We surmised that the SAH and IVH were caused by a large varix of DAVF and that the ACoA aneurysm would be unruptured. The second patient presented with a 1-week history of headaches and nausea and was diagnosed to have an SAH caused by a ruptured MCA aneurysm. We suspected vasospasm in the second portion of the MCA on CTA, but could not precisely evaluate the affected lesions. A diffusion-weighted MRI scan 4 days after surgery revealed a high-intensity area in the region of the right MCA. The MCA had already seemed to be affected at admission because vasospasm rarely develops within 4 days of the onset of SAH. Conclusions: As long as the CTA scan is of adequate quality and shows the aneurysm clearly, we consider that an additional DSA provides little useful information for surgery. However, in such cases, the information from a DSA scan is needed for the evaluation of secondary factors that are not directly associated with the aneurysm. D 2007 Elsevier Inc. All rights reserved.
Keywords:
Subarachnoid hemorrhage; Computed tomographic angiography; Digital subtraction angiography
1. Introduction Digital subtraction angiography has been used for the diagnosis of aneurysmal SAH and to decide how it should Abbreviations: ACA, anterior cerebral artery; ACoA, anterior communicating artery; CT, computed tomography; CTA, computed tomographic angiography; DAVF, dural arteriovenous fistula; DSA, digital subtraction angiography; DWI, diffusion weighted imaging; IVH, intraventricular hemorrhage; MCA, middle cerebral artery; MD, multidetector-row; MRA, magnetic resonance angiography; MRI, magnetic resonance imaging; SAH, subarachnoid hemorrhage. 4 Corresponding author. Tel.: +81 0138 45 9239; fax: +81 0138 43 4426. E-mail address: [email protected] (Y. Hashimoto). 0090-3019/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2006.10.057
be treated. However, with recent developments in diagnostic equipment, some facilities are now allowing surgical treatment to be performed on the basis of MRA and CTA scans without recourse to DSA. These techniques are good at extracting the morphology of aneurysms and their relationship with surrounding arteries, so the addition of DSA yields little extra information. Hard-to-recognize medical conditions that are not directly related to aneurysms can present with similar imaging findings. Having studied 2 cases where adequate preoperative evaluation could not be achieved with CTA alone, we report in this article on the clinical course and image findings in these cases.
Y. Hashimoto et al. / Surgical Neurology 68 (2007) 344 – 348
We used a single-detector helical scanner (SOMATOM Plus 4; Siemens, Erlangen, Germany) to perform CTA with the following parameters: a 512 512 matrix, 140 kV, 200 mA, 1-mm section collimation, 1-mm pitch, 0.5-mm reconstruction pitch, and 3608 reconstruction interposition. A total of 100 mL of Iopamidol (Iopamiron 300 mg/mL) was injected intravenously via a 20-gauge needle positioned in the medial cubital vein at a rate of 3 mL/s and with an 18-second delay before scanning. A volume-rendering method was used for the extraction of cerebral arteries. Digital subtraction angiography was performed transbrachially with a 4F catheter, using 4 to 7 mL of ioxaglic acid (Hexabrix 320 mg/mL) per injection. Selective 3- or 4 -vessel DSA scans with anteroposterior, lateral, and oblique projections were obtained.
2. Case reports 2.1. Case 1 A 68-year-old woman developed the sudden onset of consciousness disturbance and was rushed to our hospital. She had no particular history of such events. On arrival at hospital, the patient was deeply comatose and tetraparetic
345
and exhibited a decerebrated posture to pain stimuli. Because her breathing was irregular, she had to be intubated. A CT scan revealed a diffuse SAH predominantly in the posterior fossa and IVH from the 4th ventricle to the lateral ventricle (Fig. 1A). In CTA, we observed an ACoA aneurysm projecting to the right side and with a bleb (Fig. 1B). We first scheduled a clipping operation, considering that the SAH and IVH were caused by the rupture of this ACoA aneurysm. However, the SAH was sparse in the interhemispheric fissure but strong in the posterior fossa, whereas the IVH was denser in the 4th ventricle than in the lateral ventricle, which seemed to be inconsistent with the hematoma distribution of this ACoA aneurysm rupture. Because of the discordant imaging findings, a DSA study was performed. The DSA scan revealed a DAVF at the craniocervical junction, fed by the occipital artery (Fig. 1C). The dilated draining vein formed a large varix and was flowing back into the cortical veins of the posterior fossa. We could not detect the DAVF in the CTA scan, even in a retrospective review. On the day the symptoms occurred, only drainage of both lateral ventricles was performed to treat hydrocephalus because the patient was in a severe neurologic state. On the 11th day, a T2-weighted MRI scan revealed that the upper cervical cord
Fig. 1. A: Computed tomographic scan showing a diffuse SAH predominantly in the posterior fossa and ventricular hemorrhage. B: Computed tomographic angiography scan showing an ACoA aneurysm projecting to the right side. C: Right external carotid angiogram, lateral view, showing a DAVF fed by the right occipital artery with cortical venous reflex and a varix. D: T2-weighted MRI scan 11 days after the hemorrhage showing a high-intensity area in the medulla oblongata and upper cervical cord, caused by compression of the DAVF at the craniocervical junction.
346
Y. Hashimoto et al. / Surgical Neurology 68 (2007) 344 – 348
Fig. 2. A: Admission CT showing SAH located in the right sylvian fissure. B: A CTA scan showing a right MCA aneurysm. Slight narrowing can be seen in the 2nd segment of the right MCA (M2). C: Diffusion-weighted MRI scan 4 days after surgery showing a high intensity area in the region of the MCA. D: A CT scan 26 days after surgery showing small infarctions in the right insular cortex.
and the medulla oblongata had been pushed forward significantly by the AVF and appeared in the scan as a high-intensity region (Fig. 1D). At this point, the source of the hemorrhage was finally judged to be the DAVF. Three weeks later, transarterial embolization was performed on the DAVF. The aneurysm was considered not to have ruptured, and its progress was monitored with MRA. At follow-up in 7 months, the patient had recovered to clear consciousness and could communicate with very little difficulty. However, gait disturbance had persisted because of residual faint right hemiparesis. 2.2. Case 2 A 54-year-old female who had been experiencing headaches and nausea for the previous week was admitted after experiencing a disturbance of consciousness. Apart from a smoking habit, she had no medical history. The patient was found to exhibit somnolence, left-side hemiparesis, and a conjugated deviation toward the right side. A CT scan revealed a SAH localized at the right sylvian fissure (Fig. 2A), and a CTA scan revealed an MCA aneurysm (Fig. 2B). In the second portion of MCA, the right side has the impression of being narrower overall, compared with the left side, although this was not proved conclusively. A clipping operation was performed on the day of admission. After surgery, no new neurologic symptoms appeared, and
on the following day, the hemiplegia also disappeared. During the operation, deposits of hemosiderin were found on the surface of the brain, suggesting that bleeding had occurred before this time. It was thus thought that a mild vasospasm had occurred, so after surgery, we endeavored to prevent the progress of vasospasm by intravenously administering the myosin channel blocker fasudil hydrochloride (90 mg/d), and the volume was expanded with lowmolecular-weight dextran. Although not reflected in the neurologic symptoms, a diffusion-weighted MRI scan conducted on day 4 after surgery revealed an extensive high-intensity area in the right MCA region (Fig. 2C). Thereafter, dopamine was administered for 14 days to bring about induced hypertension and keep the systolic blood pressure at 180 mm Hg or higher. Although a small cerebral infarction occurred, the patient was able to return home without neurologic deficits (Fig. 2D). 3. Discussion The first steps in the treatment of aneurysmal SAH are to identify the location of the aneurysm and prevent rebleeding from occurring. Hitherto, the sources of hemorrhages generally have been searched for on the basis of DSA scans. However, the improved resolution of modern MRA and CTA technology means that in nearly
Y. Hashimoto et al. / Surgical Neurology 68 (2007) 344 – 348
all cases, it is possible to perform surgery that do not involve DSA [8,13,15]. Compared with surgical observations or the results of pre- and postoperative DSA scans, the aneurysm diagnosis performance of CTA has a sensitivity of 95% to 100% for ruptured aneurysms or 90% to 96% for amalgamated unruptured aneurysms, and reportedly, it is equal or superior to that of DSA [3,4,13,14]. Furthermore, it can produce clear visualizations of the morphology of aneurysms and their anatomical relationship to marginal vascular structures of importance for therapeutic decisions and is particularly useful for the anterior communicating artery aneurysm where the vascular system is complex [4]. Computed tomographic angiography involves none of the catheter complications that may occur with DSA and can be performed immediately after diagnosis by the CT. That there are fewer effects on circulatory dynamics because of movement and the time of exploration is shorter than DSA results in a reduced danger of rebleeding [4,15]. Situations in which it is preferable to use DSA include cases where it is necessary to confirm the degree of development of a Labbe´ vein or the circulatory flow direction of a vein [8,9], cases where a combination of bypass surgery is being considered for large aneurysms [3,8] or aneurysms are located close to the skull [3,4,8], cases where aneurysms are to be dissected [8], and cases where endovascular treatment is selected [4]. Except for these considerations, if these aneurysms can be imaged with suitable image quality by CTA, we believe that the addition of DSA provides no further information needed for surgery [8,13,15]. On the other hand, it has been pointed out that with CTA, it is possible to overlook amalgamated lesions because the imaging conditions are aimed at aneurysms and that the technique has poor clarity for stenotic lesions of the intracranial arteries [3,8]. However, these lesions are not directly associated with aneurysms and have no effect on how aneurysms are treated during surgery. This results in a reduction in the importance of preoperative recognition, which can ultimately become a pitfall in the treatment of aneurysms by CTA alone. At our institution, we use DSA to search for the sources of hemorrhages as a rule, but surgery is performed based on CTA scans alone in cases such as in elderly patients where catheter procedures are made more difficult because of arteriosclerosis, in neurologically poorgrade patients whose examination must be completed in a short time, or when there is a localized hematoma and the position of the aneurysm can be predicted from a CT scan. In this policy, to ascertain the disease state correctly, we have experienced 2 cases of SAH where it was necessary to perform a DSA scan in addition to a CTA scan. In one case, a DAVF was not visualized by CTA, but there were doubts about the consistency of the aneurysm and hematoma distribution, and the DAVF was detected for the first time by performing a supplementary DSA scan. In cases where multiple hemorrhagic conditions are present,
347
such as multiple aneurysms, it is almost possible to determine which lesion is responsible for the hemorrhaging, based on information such as the positions and orientations of the aneurysms, the direction in which the bleeding is spreading, the location of any hemorrhagic lesions, and the ease with which hemorrhaging occurs. Moreover, the cause of SAH is attributable to saccular aneurysms in 85% of cases, nonaneurysmal perimesencephalic hemorrhage in 10%, and a variety of rare conditions such as arterial dissection, cerebral arteriovenous malformation, and DAVF in 5% [16]. It therefore seems that this case was very unusual. At our institution, CTA imaging is obtained parallel to the orbitomeatal line, with scanning from the posterior inferior cerebellar arteries up to the peripheral part (A2-A3 segment) of the ACA. Almost all aneurysms are included in this range, but in the future, it will be necessary to include at least the lower end of the craniocervical junction into the imaging range. It has been pointed out that it is worth remembering that CTA scans only evaluate a limited imaging region, such as the common locations for the occurrence of aneurysms [6,7]. Vasospasms after SAH commonly occur between days 7 and 10 and seldom occur before day 4 [5]. In the other of the 2 cases, a DWI scan revealed a broad high-intensity area in the right MCA region, and it is thought that vasospasm had already occurred when the patient was admitted. A recent study demonstrated that CTA scans were a reliable means of detecting the severity of intracranial stenosis and occlusion [2]. Computed tomographic angiography scans have been shown to be 98% sensitive for stenosis (N 30% reduction) and 100% sensitive for occlusion, with DSA as the reference standard. This study was evaluated for intracranial stenotic lesions in patients with atherosclerotic disease. However, known and suspected vasospasm was excluded. Computed tomographic angiography scans are also useful for evaluating vasospasms [10,12] and produce results consistent with those of DSA scans in 90% of cases where there is at least 50% constriction and no spasm. However, this decreases to about 60% when the constriction is less than 50%, and with mild spasm, it is likely to produce an inadequate assessment [1,12]. In addition, a close correspondence with DSA is achieved in the basilar artery, the internal carotid artery, or the first segments of the ACA and MCA (A1, M1), but the correspondence is lower in peripheral parts such as A2 and M2 [1]. Recently, MD CTA performed with an MD CT scanner can be revealed more efficiently in the diagnosis of intracranial vasospasm. Yoon et al [17] demonstrated that the agreements between MD CTA and DSA scans in the overall, proximal, and distal segments of the cerebral arteries were 95.2%, 96.0%, and 94.1%, respectively. In addition, they achieved 98.0% accuracy in the evaluation of severe vasospasms (at least 51% reduction). In another study, it was reported that the overall agreement between
348
Y. Hashimoto et al. / Surgical Neurology 68 (2007) 344 – 348
MD CTA and DSA scans was 91.6%, indicating that MD CTA can detect angiographic vasospasm after SAH with accuracy equal to that of DSA [11]. These studies were based on a comparison of corresponding arterial segments in preoperative and postoperative studies where vasospasms were not diagnosed in the first MD CTA. In the CTA performed in this second case, we gained the impression that the M2 on the right was diffuse narrower than the one on the left. However, the difference was very slight and could not be judged to have any connection with the vasospasm. During surgery, we observed locations with deposits of hemosiderin, and once again, we recognized the possibility of vasospasm being exhibited. It is unclear whether performing a DSA scan would have led to a diagnosis; however, except for cases where the image clearly shows rebleeding or ischemic symptoms, it is impossible to diagnose mild vasospasm in the initial CTA with relatively little contrast. References [1] Anderson GB, Asforth R, Steinke DE, Findlay JM. CT angiography for the detection of cerebral vasospasm in patients with acute subarachnoid hemorrhage. AJNR Am J Neuroradiol 2000;21:1011 - 5. [2] Bash S, Villablanca JP, Jahan R, Duckwiler G, Tillis M, Kidwell C, Saver J, Sayre J. Intracranial vascular stenosis and occlusive disease: evaluation with CT angiography, MR angiography, and digital subtraction angiography. AJNR Am J Neuroradiol 2005;26:1012 - 21. [3] Dehdashti AR, Rufenacht DA, Delavelle J, Reverdin A, DE Tribolet N. Therapeutic decision and management of aneurysmal subarachnoid hemorrhage based on computed tomographic angiography. Br J Neurosurg 2003;17:46 - 53. [4] Gonza´lez-Darder JM, Pesudo-Martı´nez JV, Feliu-Tatay JV. Microsurgical management of cerebral aneurysms based in CT angiography with three-dimensional reconstruction (3D-CTA) and without preoperative cerebral angiography. Acta Neurochir (Wien) 2001;143:673 - 9. [5] Kassel NF, Sasaki T, Colohan AR, Nasar G. Cerebral vasospasm following aneurysmal subarachnoid hemorrhage. Stroke 1985;16: 562 - 72.
[6] Kim MS, Han DH, Han MH, Oh CW. Posterior fossa hemorrhage caused by dural arteriovenous fistula: case reports. Surg Neurol 2005; 59:512 - 7. [7] Kinouchi H, Mizoi K, Takahashi A, Nagamine Y, Koshu K, Yoshimoto T. Dural arteriovenous shunt at the craniocervical junction. J Neurosurg 1998;89:755 - 61. [8] Matsumoto M, Sato M, Nakano M, Endo Y, Watanabe Y, Sasaki T, Suzuki K, Kodama N. Three-dimensional computerized tomography angiography-guided surgery of acutely ruptured cerebral aneurysms. J Neurosurg 2001;94:718 - 27. [9] Miyamoto S, Yamada K, Kikuta K, Kataoka D, Satow T, Hashimoto N. Strategy for the proper and safe treatment of cerebral aneurysm. Jpn J Neurosurg (Tokyo) 2003;12:412 - 8 [Jpn, with Eng. abstract]. [10] Ochi RP, Vieco PT, Gross CE. CT angiography of cerebral vasospasm with conventional angiography comparison. AJNR Am J Neuroradiol 1997;18:265 - 9. [11] Otawara Y, Ogasawara K, Ogawa A, Sasaki M, Takahashi K. Evaluation of vasospasm after subarachnoid hemorrhage by use of multislice computed tomography angiography. Neurosurgery 2002;51:939 - 43. [12] Takagi R, Hayashi H, Kobayashi H, Kumazaki T, Isayama K, Ikeda Y, Teramoto A. Three-dimensional CT angiography of intracranial vasospasm following subarachnoid haemorrhage. Neuroradiology 1998;40:631 - 5. [13] Tanabe S, Ohtaki M, Uede T, Hashi K, Suzuki S, Takahashi H. Diagnosis of ruptured and unruptured cerebral aneurysms with threedimensional CT angiography. No Shinkei Geka 1995;23:787 - 95 [Jpn, with Eng abstract]. [14] Velthuis BK, Rinkel GJE, Romos LMP, Witkamp TD, van der Sprenkel JWB, Vandertop WP, van Leeuwen MS. Subarachnoid hemorrhage: aneurysm detection and preoperative evaluation with CT angiography. Radiology 1998;208:423 - 30. [15] Velthuis BK, van Leeuwen MS, Witkamp TD, Romos LMP, van der Sprenkel JWB, Vandertop WP, Rinkel GJE. Computerized tomography angiography in patients with subarachnoid hemorrhage: from aneurysm detection to treatment without conventional angiography. J Neurosurg 1999;91:761 - 7. [16] von Gijn J, Rinkel GJE. Subarachnoid haemorrhage: diagnosis, causes and management. Brain 2001;124:249 - 78. [17] Yoon DY, Choi CS, Kim KH, Cho BM. Multidetector-row CT angiography of cerebral vasospasm after aneurysmal subarachnoid hemorrhage: comparison of volume-rendered images and digital subtraction angiography. AJNR Am J Neuroradiol 2006;27:370 - 7.
Imaging
Pitfalls in the preoperative evaluation of subarachnoid hemorrhage without digital subtraction angiography: report on 2 cases Yuji Hashimoto, MD4, Sonen Kin, MD, Koichi Haraguchi, MD, Jun Niwa, MD Department of Neurosurgery, Hakodate Municipal Hospital, Hokkaido 041-8680, Japan Received 9 February 2006; accepted 6 October 2006
Abstract
Background: Digital subtraction angiography has been used in the diagnosis of aneurysmal SAH and as a preoperative imaging method. However, new methods such as MRA and CTA are now deemed by many institutions to provide sufficient information to allow surgery to go ahead without a preliminary DSA scan. We report on 2 cases of SAH in which there were additional lesions that were difficult to evaluate because of the lack of DSA information. Case Descriptions: The fist patient demonstrated SAH with IVH. Computed tomographic angiography revealed an ACoA aneurysm with a bleb. We first thought that the SAH and IVH were both caused by a ruptured ACoA aneurysm but noted that hemorrhage pattern was inconsistent with the location and orientation of the aneurysm. A DSA scan revealed a dural arteriovenous fistula in the region of the craniocervical junction, supplied by the right occipital artery. We surmised that the SAH and IVH were caused by a large varix of DAVF and that the ACoA aneurysm would be unruptured. The second patient presented with a 1-week history of headaches and nausea and was diagnosed to have an SAH caused by a ruptured MCA aneurysm. We suspected vasospasm in the second portion of the MCA on CTA, but could not precisely evaluate the affected lesions. A diffusion-weighted MRI scan 4 days after surgery revealed a high-intensity area in the region of the right MCA. The MCA had already seemed to be affected at admission because vasospasm rarely develops within 4 days of the onset of SAH. Conclusions: As long as the CTA scan is of adequate quality and shows the aneurysm clearly, we consider that an additional DSA provides little useful information for surgery. However, in such cases, the information from a DSA scan is needed for the evaluation of secondary factors that are not directly associated with the aneurysm. D 2007 Elsevier Inc. All rights reserved.
Keywords:
Subarachnoid hemorrhage; Computed tomographic angiography; Digital subtraction angiography
1. Introduction Digital subtraction angiography has been used for the diagnosis of aneurysmal SAH and to decide how it should Abbreviations: ACA, anterior cerebral artery; ACoA, anterior communicating artery; CT, computed tomography; CTA, computed tomographic angiography; DAVF, dural arteriovenous fistula; DSA, digital subtraction angiography; DWI, diffusion weighted imaging; IVH, intraventricular hemorrhage; MCA, middle cerebral artery; MD, multidetector-row; MRA, magnetic resonance angiography; MRI, magnetic resonance imaging; SAH, subarachnoid hemorrhage. 4 Corresponding author. Tel.: +81 0138 45 9239; fax: +81 0138 43 4426. E-mail address: [email protected] (Y. Hashimoto). 0090-3019/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2006.10.057
be treated. However, with recent developments in diagnostic equipment, some facilities are now allowing surgical treatment to be performed on the basis of MRA and CTA scans without recourse to DSA. These techniques are good at extracting the morphology of aneurysms and their relationship with surrounding arteries, so the addition of DSA yields little extra information. Hard-to-recognize medical conditions that are not directly related to aneurysms can present with similar imaging findings. Having studied 2 cases where adequate preoperative evaluation could not be achieved with CTA alone, we report in this article on the clinical course and image findings in these cases.
Y. Hashimoto et al. / Surgical Neurology 68 (2007) 344 – 348
We used a single-detector helical scanner (SOMATOM Plus 4; Siemens, Erlangen, Germany) to perform CTA with the following parameters: a 512 512 matrix, 140 kV, 200 mA, 1-mm section collimation, 1-mm pitch, 0.5-mm reconstruction pitch, and 3608 reconstruction interposition. A total of 100 mL of Iopamidol (Iopamiron 300 mg/mL) was injected intravenously via a 20-gauge needle positioned in the medial cubital vein at a rate of 3 mL/s and with an 18-second delay before scanning. A volume-rendering method was used for the extraction of cerebral arteries. Digital subtraction angiography was performed transbrachially with a 4F catheter, using 4 to 7 mL of ioxaglic acid (Hexabrix 320 mg/mL) per injection. Selective 3- or 4 -vessel DSA scans with anteroposterior, lateral, and oblique projections were obtained.
2. Case reports 2.1. Case 1 A 68-year-old woman developed the sudden onset of consciousness disturbance and was rushed to our hospital. She had no particular history of such events. On arrival at hospital, the patient was deeply comatose and tetraparetic
345
and exhibited a decerebrated posture to pain stimuli. Because her breathing was irregular, she had to be intubated. A CT scan revealed a diffuse SAH predominantly in the posterior fossa and IVH from the 4th ventricle to the lateral ventricle (Fig. 1A). In CTA, we observed an ACoA aneurysm projecting to the right side and with a bleb (Fig. 1B). We first scheduled a clipping operation, considering that the SAH and IVH were caused by the rupture of this ACoA aneurysm. However, the SAH was sparse in the interhemispheric fissure but strong in the posterior fossa, whereas the IVH was denser in the 4th ventricle than in the lateral ventricle, which seemed to be inconsistent with the hematoma distribution of this ACoA aneurysm rupture. Because of the discordant imaging findings, a DSA study was performed. The DSA scan revealed a DAVF at the craniocervical junction, fed by the occipital artery (Fig. 1C). The dilated draining vein formed a large varix and was flowing back into the cortical veins of the posterior fossa. We could not detect the DAVF in the CTA scan, even in a retrospective review. On the day the symptoms occurred, only drainage of both lateral ventricles was performed to treat hydrocephalus because the patient was in a severe neurologic state. On the 11th day, a T2-weighted MRI scan revealed that the upper cervical cord
Fig. 1. A: Computed tomographic scan showing a diffuse SAH predominantly in the posterior fossa and ventricular hemorrhage. B: Computed tomographic angiography scan showing an ACoA aneurysm projecting to the right side. C: Right external carotid angiogram, lateral view, showing a DAVF fed by the right occipital artery with cortical venous reflex and a varix. D: T2-weighted MRI scan 11 days after the hemorrhage showing a high-intensity area in the medulla oblongata and upper cervical cord, caused by compression of the DAVF at the craniocervical junction.
346
Y. Hashimoto et al. / Surgical Neurology 68 (2007) 344 – 348
Fig. 2. A: Admission CT showing SAH located in the right sylvian fissure. B: A CTA scan showing a right MCA aneurysm. Slight narrowing can be seen in the 2nd segment of the right MCA (M2). C: Diffusion-weighted MRI scan 4 days after surgery showing a high intensity area in the region of the MCA. D: A CT scan 26 days after surgery showing small infarctions in the right insular cortex.
and the medulla oblongata had been pushed forward significantly by the AVF and appeared in the scan as a high-intensity region (Fig. 1D). At this point, the source of the hemorrhage was finally judged to be the DAVF. Three weeks later, transarterial embolization was performed on the DAVF. The aneurysm was considered not to have ruptured, and its progress was monitored with MRA. At follow-up in 7 months, the patient had recovered to clear consciousness and could communicate with very little difficulty. However, gait disturbance had persisted because of residual faint right hemiparesis. 2.2. Case 2 A 54-year-old female who had been experiencing headaches and nausea for the previous week was admitted after experiencing a disturbance of consciousness. Apart from a smoking habit, she had no medical history. The patient was found to exhibit somnolence, left-side hemiparesis, and a conjugated deviation toward the right side. A CT scan revealed a SAH localized at the right sylvian fissure (Fig. 2A), and a CTA scan revealed an MCA aneurysm (Fig. 2B). In the second portion of MCA, the right side has the impression of being narrower overall, compared with the left side, although this was not proved conclusively. A clipping operation was performed on the day of admission. After surgery, no new neurologic symptoms appeared, and
on the following day, the hemiplegia also disappeared. During the operation, deposits of hemosiderin were found on the surface of the brain, suggesting that bleeding had occurred before this time. It was thus thought that a mild vasospasm had occurred, so after surgery, we endeavored to prevent the progress of vasospasm by intravenously administering the myosin channel blocker fasudil hydrochloride (90 mg/d), and the volume was expanded with lowmolecular-weight dextran. Although not reflected in the neurologic symptoms, a diffusion-weighted MRI scan conducted on day 4 after surgery revealed an extensive high-intensity area in the right MCA region (Fig. 2C). Thereafter, dopamine was administered for 14 days to bring about induced hypertension and keep the systolic blood pressure at 180 mm Hg or higher. Although a small cerebral infarction occurred, the patient was able to return home without neurologic deficits (Fig. 2D). 3. Discussion The first steps in the treatment of aneurysmal SAH are to identify the location of the aneurysm and prevent rebleeding from occurring. Hitherto, the sources of hemorrhages generally have been searched for on the basis of DSA scans. However, the improved resolution of modern MRA and CTA technology means that in nearly
Y. Hashimoto et al. / Surgical Neurology 68 (2007) 344 – 348
all cases, it is possible to perform surgery that do not involve DSA [8,13,15]. Compared with surgical observations or the results of pre- and postoperative DSA scans, the aneurysm diagnosis performance of CTA has a sensitivity of 95% to 100% for ruptured aneurysms or 90% to 96% for amalgamated unruptured aneurysms, and reportedly, it is equal or superior to that of DSA [3,4,13,14]. Furthermore, it can produce clear visualizations of the morphology of aneurysms and their anatomical relationship to marginal vascular structures of importance for therapeutic decisions and is particularly useful for the anterior communicating artery aneurysm where the vascular system is complex [4]. Computed tomographic angiography involves none of the catheter complications that may occur with DSA and can be performed immediately after diagnosis by the CT. That there are fewer effects on circulatory dynamics because of movement and the time of exploration is shorter than DSA results in a reduced danger of rebleeding [4,15]. Situations in which it is preferable to use DSA include cases where it is necessary to confirm the degree of development of a Labbe´ vein or the circulatory flow direction of a vein [8,9], cases where a combination of bypass surgery is being considered for large aneurysms [3,8] or aneurysms are located close to the skull [3,4,8], cases where aneurysms are to be dissected [8], and cases where endovascular treatment is selected [4]. Except for these considerations, if these aneurysms can be imaged with suitable image quality by CTA, we believe that the addition of DSA provides no further information needed for surgery [8,13,15]. On the other hand, it has been pointed out that with CTA, it is possible to overlook amalgamated lesions because the imaging conditions are aimed at aneurysms and that the technique has poor clarity for stenotic lesions of the intracranial arteries [3,8]. However, these lesions are not directly associated with aneurysms and have no effect on how aneurysms are treated during surgery. This results in a reduction in the importance of preoperative recognition, which can ultimately become a pitfall in the treatment of aneurysms by CTA alone. At our institution, we use DSA to search for the sources of hemorrhages as a rule, but surgery is performed based on CTA scans alone in cases such as in elderly patients where catheter procedures are made more difficult because of arteriosclerosis, in neurologically poorgrade patients whose examination must be completed in a short time, or when there is a localized hematoma and the position of the aneurysm can be predicted from a CT scan. In this policy, to ascertain the disease state correctly, we have experienced 2 cases of SAH where it was necessary to perform a DSA scan in addition to a CTA scan. In one case, a DAVF was not visualized by CTA, but there were doubts about the consistency of the aneurysm and hematoma distribution, and the DAVF was detected for the first time by performing a supplementary DSA scan. In cases where multiple hemorrhagic conditions are present,
347
such as multiple aneurysms, it is almost possible to determine which lesion is responsible for the hemorrhaging, based on information such as the positions and orientations of the aneurysms, the direction in which the bleeding is spreading, the location of any hemorrhagic lesions, and the ease with which hemorrhaging occurs. Moreover, the cause of SAH is attributable to saccular aneurysms in 85% of cases, nonaneurysmal perimesencephalic hemorrhage in 10%, and a variety of rare conditions such as arterial dissection, cerebral arteriovenous malformation, and DAVF in 5% [16]. It therefore seems that this case was very unusual. At our institution, CTA imaging is obtained parallel to the orbitomeatal line, with scanning from the posterior inferior cerebellar arteries up to the peripheral part (A2-A3 segment) of the ACA. Almost all aneurysms are included in this range, but in the future, it will be necessary to include at least the lower end of the craniocervical junction into the imaging range. It has been pointed out that it is worth remembering that CTA scans only evaluate a limited imaging region, such as the common locations for the occurrence of aneurysms [6,7]. Vasospasms after SAH commonly occur between days 7 and 10 and seldom occur before day 4 [5]. In the other of the 2 cases, a DWI scan revealed a broad high-intensity area in the right MCA region, and it is thought that vasospasm had already occurred when the patient was admitted. A recent study demonstrated that CTA scans were a reliable means of detecting the severity of intracranial stenosis and occlusion [2]. Computed tomographic angiography scans have been shown to be 98% sensitive for stenosis (N 30% reduction) and 100% sensitive for occlusion, with DSA as the reference standard. This study was evaluated for intracranial stenotic lesions in patients with atherosclerotic disease. However, known and suspected vasospasm was excluded. Computed tomographic angiography scans are also useful for evaluating vasospasms [10,12] and produce results consistent with those of DSA scans in 90% of cases where there is at least 50% constriction and no spasm. However, this decreases to about 60% when the constriction is less than 50%, and with mild spasm, it is likely to produce an inadequate assessment [1,12]. In addition, a close correspondence with DSA is achieved in the basilar artery, the internal carotid artery, or the first segments of the ACA and MCA (A1, M1), but the correspondence is lower in peripheral parts such as A2 and M2 [1]. Recently, MD CTA performed with an MD CT scanner can be revealed more efficiently in the diagnosis of intracranial vasospasm. Yoon et al [17] demonstrated that the agreements between MD CTA and DSA scans in the overall, proximal, and distal segments of the cerebral arteries were 95.2%, 96.0%, and 94.1%, respectively. In addition, they achieved 98.0% accuracy in the evaluation of severe vasospasms (at least 51% reduction). In another study, it was reported that the overall agreement between
348
Y. Hashimoto et al. / Surgical Neurology 68 (2007) 344 – 348
MD CTA and DSA scans was 91.6%, indicating that MD CTA can detect angiographic vasospasm after SAH with accuracy equal to that of DSA [11]. These studies were based on a comparison of corresponding arterial segments in preoperative and postoperative studies where vasospasms were not diagnosed in the first MD CTA. In the CTA performed in this second case, we gained the impression that the M2 on the right was diffuse narrower than the one on the left. However, the difference was very slight and could not be judged to have any connection with the vasospasm. During surgery, we observed locations with deposits of hemosiderin, and once again, we recognized the possibility of vasospasm being exhibited. It is unclear whether performing a DSA scan would have led to a diagnosis; however, except for cases where the image clearly shows rebleeding or ischemic symptoms, it is impossible to diagnose mild vasospasm in the initial CTA with relatively little contrast. References [1] Anderson GB, Asforth R, Steinke DE, Findlay JM. CT angiography for the detection of cerebral vasospasm in patients with acute subarachnoid hemorrhage. AJNR Am J Neuroradiol 2000;21:1011 - 5. [2] Bash S, Villablanca JP, Jahan R, Duckwiler G, Tillis M, Kidwell C, Saver J, Sayre J. Intracranial vascular stenosis and occlusive disease: evaluation with CT angiography, MR angiography, and digital subtraction angiography. AJNR Am J Neuroradiol 2005;26:1012 - 21. [3] Dehdashti AR, Rufenacht DA, Delavelle J, Reverdin A, DE Tribolet N. Therapeutic decision and management of aneurysmal subarachnoid hemorrhage based on computed tomographic angiography. Br J Neurosurg 2003;17:46 - 53. [4] Gonza´lez-Darder JM, Pesudo-Martı´nez JV, Feliu-Tatay JV. Microsurgical management of cerebral aneurysms based in CT angiography with three-dimensional reconstruction (3D-CTA) and without preoperative cerebral angiography. Acta Neurochir (Wien) 2001;143:673 - 9. [5] Kassel NF, Sasaki T, Colohan AR, Nasar G. Cerebral vasospasm following aneurysmal subarachnoid hemorrhage. Stroke 1985;16: 562 - 72.
[6] Kim MS, Han DH, Han MH, Oh CW. Posterior fossa hemorrhage caused by dural arteriovenous fistula: case reports. Surg Neurol 2005; 59:512 - 7. [7] Kinouchi H, Mizoi K, Takahashi A, Nagamine Y, Koshu K, Yoshimoto T. Dural arteriovenous shunt at the craniocervical junction. J Neurosurg 1998;89:755 - 61. [8] Matsumoto M, Sato M, Nakano M, Endo Y, Watanabe Y, Sasaki T, Suzuki K, Kodama N. Three-dimensional computerized tomography angiography-guided surgery of acutely ruptured cerebral aneurysms. J Neurosurg 2001;94:718 - 27. [9] Miyamoto S, Yamada K, Kikuta K, Kataoka D, Satow T, Hashimoto N. Strategy for the proper and safe treatment of cerebral aneurysm. Jpn J Neurosurg (Tokyo) 2003;12:412 - 8 [Jpn, with Eng. abstract]. [10] Ochi RP, Vieco PT, Gross CE. CT angiography of cerebral vasospasm with conventional angiography comparison. AJNR Am J Neuroradiol 1997;18:265 - 9. [11] Otawara Y, Ogasawara K, Ogawa A, Sasaki M, Takahashi K. Evaluation of vasospasm after subarachnoid hemorrhage by use of multislice computed tomography angiography. Neurosurgery 2002;51:939 - 43. [12] Takagi R, Hayashi H, Kobayashi H, Kumazaki T, Isayama K, Ikeda Y, Teramoto A. Three-dimensional CT angiography of intracranial vasospasm following subarachnoid haemorrhage. Neuroradiology 1998;40:631 - 5. [13] Tanabe S, Ohtaki M, Uede T, Hashi K, Suzuki S, Takahashi H. Diagnosis of ruptured and unruptured cerebral aneurysms with threedimensional CT angiography. No Shinkei Geka 1995;23:787 - 95 [Jpn, with Eng abstract]. [14] Velthuis BK, Rinkel GJE, Romos LMP, Witkamp TD, van der Sprenkel JWB, Vandertop WP, van Leeuwen MS. Subarachnoid hemorrhage: aneurysm detection and preoperative evaluation with CT angiography. Radiology 1998;208:423 - 30. [15] Velthuis BK, van Leeuwen MS, Witkamp TD, Romos LMP, van der Sprenkel JWB, Vandertop WP, Rinkel GJE. Computerized tomography angiography in patients with subarachnoid hemorrhage: from aneurysm detection to treatment without conventional angiography. J Neurosurg 1999;91:761 - 7. [16] von Gijn J, Rinkel GJE. Subarachnoid haemorrhage: diagnosis, causes and management. Brain 2001;124:249 - 78. [17] Yoon DY, Choi CS, Kim KH, Cho BM. Multidetector-row CT angiography of cerebral vasospasm after aneurysmal subarachnoid hemorrhage: comparison of volume-rendered images and digital subtraction angiography. AJNR Am J Neuroradiol 2006;27:370 - 7.