- Email: [email protected]
“Stretched loop sign” of the vertebral artery: a predictor of vertebrobasilar insufficiency in atlantoaxial dislocation
Surgical Neurology 66 (2006) 298 – 304 www.surgicalneurology-online.com
Imaging
bStretched loop signQ of the vertebral artery: a predictor of vertebrobasilar insufficiency in atlantoaxial dislocation Vijay Sawlani, MDa, Sanjay Behari, MCh, DNBb,4, Pravin Salunke, MSb, Vijendra K. Jain, MChb, Rajendra V. Phadke, MDc a Department of Neuroradiology, Morriston Hospital, Swansea, SA6 6NL, UK Departments of bNeurosurgery and cNeuroradiology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow 226014, India Received 20 August 2005; accepted 2 February 2006
Abstract
Background: Vertebrobasilar territory infarction is one of the rarer presentations of CVJ anomalies. A new radiologic sign due to stretching of the short third segment of VA detected on MRA/DSA may identify patients of AAD at risk of developing VBI. Methods: Seven patients who presented with VBI were found to have a coexisting mobile (n = 6) or fixed (n = 1) AAD. None of these patients had the presence of any of the known risk factors for cerebrovascular disease. On identification of VBI on CT/MRI, DSA (n = 7) and MRA (n = 1) were performed to assess bilateral vertebral arteries. The course of normal VA was also studied in 5 control patients without AAD or VBI. Results: Digital subtraction angiography/MRA showed obstruction of VA at the C1 through C2 level on one side in each of these cases. The third segment of the contralateral VA showed a shortened and straighter loop termed as the stretched loop sign of the VA. On DSA, the latter manifested as (a) opening of the distal loop of the VA as it emerges from the foramen transversarium of the atlas and traverses on the dorsum of the posterior arch of atlas (n = 3), (b) shortened and stretched VA that runs laterally and posteriorly forming the proximal loop after emerging from the foramen transversarium of the axis (n = 2), or (c) both (n = 2). All patients presented with the clinical manifestations of VBI. Only 2 of these had preexisting myelopathy and long tract signs conventionally attributable to AAD. Conclusion: Vertebrobasilar territory infarction in AAD may occur because of the obstruction of the third segment of VA. A shorter and straighter loop of the third segment of VA coexisting with an abnormal translational mobility between the atlas and the axis may be the etiopathogenetic factor. D 2006 Elsevier Inc. All rights reserved.
Keywords:
Vertebral artery; Ischemia; Atlantoaxial dislocation; Cervical spine
1. Introduction Vertebral artery is particularly vulnerable to obstruction in patients with AAD in whom an abnormal translational Abbreviations: AAD, atlantoaxial dislocation; BA, basilar artery; CT, computed tomography; CVJ, craniovertebral junction; DSA, digital subtraction angiography; MRA, magnetic resonance angiography; MRI, magnetic resonance imaging; VA, vertebral artery; VBI, vertebrobasilar territory infarction. 4 Corresponding author. Fax: +91 522 2668078, 2668017. E-mail addresses: [email protected], [email protected] (S. Behari). 0090-3019/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2006.02.032
movement exists between the atlas and axis [1,4,6,33,34]. The clinical manifestations in AAD are mainly due to compression of the thecal sac and spinal cord, and rarely due to VBI. In the conventional evaluation of the CVJ anomalies, however, the bony anomalies are investigated thoroughly, whereas VA often remains unevaluated because DSA or MRA does not form a part of the conventional radiologic investigative protocol. In this study, the radiologic features of 7 patients with VBI due to AAD have been presented. Their DSA revealed occlusion of VA on one side and an abnormally short and straight loop of the third segment of VA at C1 through C2 level on the other side. The
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
radiologic findings showing the presence of a shorter and straighter loop of the third segment of VA and an abnormal translational mobility between the atlas and axis help in identifying the subset of patients prone to developing VBI. 2. Material and methods Over a 12-year period between 1990 and 2002, in 7 patients with VBI, AAD was implicated as an etiologic factor for the underlying VA occlusion. The patients underwent dynamic (in flexion and extension) trans-table lateral radiographs (n = 7) and intrathecal contrast computerized tomography (n = 2) of the CVJ to assess for reducibility at the C1-2 joints and the associated bony anomalies. An MRI of the cervical spine and brain was also performed to assess for the associated soft-tissue abnormalities, the extent of VBI, and the degree of cervicomedullary compression. On identification of VBI on CT/MRI, DSA (n = 7) and MRA (n = 1) were also performed to study the course of both vertebral arteries. The tests for cerebrovascular risk factors and cardiovascular diseases and the serological tests for vasculitis were negative in these patients. Six patients were managed by stabilizing their neck movements using a hard cervical collar until they underwent posterior occipito-C2 or C1 through C2 stabilization using Jain’s or modified Brooks’ technique, respectively [2,5,13]. In patients with mobile AAD, reducibility
299
was determined by neck flexion and extension only and not cervical traction. The patient with fixed AAD underwent Crutchfield’s cervical traction for a period of 2 weeks. When the AAD was found irreducible even on traction, the patient underwent transoral decompression of odontoid before his posterior stabilization [2,5,13,20,23] under the same anesthesia. In 1 patient, surgery was deferred in view of his poor neurologic status (case 3). The course of normal VA was also studied in 5 control patients without AAD or VBI in whom a cranial DSA was being performed for a supratentorial pathology. 3. Results These patients (male/female ratio, 5:2; age range, 20-62 years; mean age, 33 years) presented with vertigo (n = 6), ataxia (n = 5), spastic quadriparesis (n = 4), dysarthria (n = 4), nystagmus and diplopia (n = 4), and transient loss of consciousness (n = 2). A short neck was seen in 3 patients and a low hairline in 1 patient. Only 2 of these had preexisting myelopathy and long tract signs before the development of VBI. The AAD was reducible in 6 patients and fixed in 1 patient. Occipitalization of atlas was present in 6 of the cases and C2 through C3 fusion in 3 of the cases. The third segment of the left VA was occluded in 4 and the right VA in 3 patients. In all these patients, the third segment of the contralateral VA showed a short and stretched loop at
Fig. 1. Case 1: Axial CT scan A: shows bilateral occipital lobe and B: cerebellar infarction. Left vertebral angiogram lateral view C: shows VA occlusion at the C1 through C2 level (arrow). Right vertebral angiogram anteroposterior D: view shows opening of the distal loop of the third segment (arrow) of VA as it emerges from the foramen transversarium of atlas and traverses on the dorsum of C1 posterior arch.
300
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
the C1 through C2 level. On DSA, the latter was evident by (a) opening of the distal loop of the VA as it emerges from the foramen transversarium of the atlas and traverses on the dorsum of the posterior arch of atlas (n = 3), (b) shortened and stretched VA that runs laterally and posteriorly forming the proximal loop after emerging from the foramen transversarium of axis (n = 2), or (c) both (n = 2). In all the 5 control patients, the third segment of bilateral vertebral arteries showed long and redundant distal and proximal loops. There was no evidence of either opening of the distal loop or shortening and stretching of the proximal loop of the third segment of VA. 3.1. Case 1: KD A 62-year-old woman presented with right-sided weakness and incoordination after an episode of sudden vertigo with a momentary lapse of consciousness 2 weeks before admission. She also had transient dysarthria that improved in a week’s time. There was no previous history of transient ischemic attacks, migraine, diabetes, hypertension, or cardiac disease. The neurologic examination revealed gaze-evoked nystagmus and right-sided ataxic hemiparesis. The CT revealed bilateral occipital lobe and cerebellar infarcts (Fig. 1A and B). The lateral radiographs of the cervical spine showed occipitalization of the atlas with a 10-mm AAD on flexion. The DSA showed a left VA occlusion at the C1 through C2 level (Fig. 1C). The right vertebral angiogram anteroposterior view (Fig. 1D) showed straightening and stretching of the third segment of VA at
the C1 through C2 level manifested as the opening of the distal loop of VA as it emerged from the foramen transversarium of atlas and traversed on the dorsum of the posterior arch of atlas. There have been no further attacks during a follow-up of 2 years after C1 through C2 stabilization. 3.2. Case 2: PBS A 35-year old man developed a sudden episode of headache, vomiting, gait ataxia, and nasal intonation after a jerk to the neck 6 months ago. For the past 1 month before admission, he developed progressive spastic quadriparesis. On examination, he had IX and X cranial nerve paresis with cerebellar signs and spastic quadriparesis. His dynamic trans-table radiograph of the CVJ revealed a mobile AAD of 7 mm with an assimilated posterior arch of atlas. The CT scan of the head revealed infarcts in the occipital and cerebellar regions (Fig. 2A and B). The dynamic MRI of the CVJ in flexed position revealed AAD causing compression of cervicomedullary junction and, in extension, showed reduction of AAD (Fig. 2C and D). The DSA showed occlusion of right VA at the level of axis (Fig. 2E and F). The third segment of left VA was short and straight. This was evident both by the opening of the distal loop of VA as it emerged from the foramen transversarium of atlas and traversed on the dorsum of posterior arch of atlas and by a shortened and stretched portion of VA that ran laterally and posteriorly forming the proximal loop after emerging from the foramen transversarium of axis
Fig. 2. Case 2: Axial CT scan A: shows occipital lobe and B: cerebellar infarction. Dynamic T2-weighted sagittal image in flexion C: shows AAD causing obliteration of subarachnoid space and compression of cervicomedullary junction, and in extension D: shows reduction of AAD. Right vertebral angiogram unsubtracted lateral E: and subtracted anteroposterior F: views show occlusion of VA at the C1 through C2 level (arrow). Left vertebral angiogram, unsubtracted lateral G: and subtracted anteroposterior H: views show straightening and shortening of the third segment of VA manifested both by the opening of the distal loop of VA as it emerges from the C1 foramen transversarium and traverses on the dorsum of C1 posterior arch (straight arrow), and by a shortened and stretched portion of VA that runs laterally and posteriorly forming the proximal loop after emerging from the C2 foramen transversarium (curved arrow).
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
(Fig. 2G and H). He underwent an occipitocervical fusion using sublaminar wires and the placement of strut and onlay bone grafts [2,13]. The patient was kept on a cervical collar for 3 months. There was a significant improvement in spasticity, and his power normalized. However, a mild gait ataxia persisted. 3.3. Case 3: RB This 35-year-old man had sudden onset of severe headache and recurrent vomiting associated with vertigo, scanning speech, and left-sided appendicular ataxia 1 month before admission. After admission, while he was being investigated, he developed a sudden loss of consciousness with absent doll’s eye movements and a decerebrate posturing. His MRI showed multiple brainstem and cerebellar infarcts (Fig. 3A). His dynamic (in flexion and extension) lateral cervical radiographs and the intrathecal dynamic contrast CT of the CVJ revealed a mobile AAD with an occipitalized posterior arch of atlas and C2 through C3 fusion (Fig. 3B and C). There was significant thecal compression in flexion. The left vertebral angiogram showed occlusion of VA at the level of C1 through C2 (Fig. 3D). The right vertebral DSA showed a shortened and stretched portion of VA that ran laterally and posteriorly forming the proximal loop after emerging from the foramen
301
transversarium of axis. There was complete filling of right VA until the origin of the BA with reflux into the left VA until its horizontal segment on the posterior arch of atlas (Fig. 3E). There was an abrupt cutoff of the dye column at the lower one-third of the BA and no forward filling of the BA. The right internal carotid artery injection showed filling of distal BA and the posterior cerebral arteries through the posterior communicating arteries. However, the proximal and mid-segmental basilar trunk showed no filling (Fig. 3F). The angiographic impression was that of VA thrombosis with migration of clot to the basilar trunk. In view of a bad prognosis related to his poor neurologic condition and the presence of established multiple brainstem and cerebellar infarcts, surgery for an occipitocervical posterior stabilization was refused by the patient’s relatives.
4. Discussion 4.1. Anatomical considerations Vertebral artery has traditionally been described in 4 segments. The first part constitutes the segment from its origin to entry into the foramen transversarium of the C6 vertebra; the second part courses vertically through the foramina transversarium of C6 to C2 vertebrae; the third
Fig. 3. Case 3: T2-weighted axial magnetic resonance image A: shows cerebellar infarction. Intrathecal contrast CT scan in flexion B: and extension C: shows mobile AAD with occipitalized posterior arch of atlas and C2 through C3 fusion. There is significant thecal compression in flexed position of the neck. Unsubtracted left vertebral angiogram D: shows occlusion of VA at the level of C1 through C2 (arrow). Subtracted right vertebral angiogram E: shows a shortened and stretched portion of VA that runs laterally and posteriorly forming the proximal loop after emerging from the foramen transversarium of the axis (arrow). There is a complete filling of the right VA until the origin of the BA with reflux into the left VA until its horizontal segment on the posterior arch of atlas, and an abrupt cutoff of the dye column at the lower one-third of BA (arrow). The right internal carotid artery injection F: shows reflux filling of distal BA and posterior cerebral arteries through the posterior communicating arteries. The proximal and mid-segmental basilar trunk shows no filling (arrow).
302
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
part extends from the foramen transversarium of C2 vertebra to its entry into the subarachnoid space; and, the fourth part courses within the subarachnoid space until it joins the opposite VA to form BA at the lower border of pons [7,12,31]. The third segment has a very tortuous course with 2 loops (Fig. 4A-C). After exiting from C2 transverse foramen, this segment of VA runs laterally and posteriorly over a short distance, forming the proximal loop, and then cranially to enter the transverse foramen of atlas. It then runs an oblique course on the dorsum of the posterior arch of atlas forming the distal loop before penetrating the posterior atlantoaxial membrane and the dura to enter the subarachnoid space [12,31]. At the C1 through C2 level, VA is particularly prone to mechanical compression during head and neck rotation. When the head is rotated, the atlantoaxial joint on the side to which the head is turned is fixed. On the opposite side, the C1 arch moves forward on the axis. Thus, the segment of VA between C1 and C2 gets stretched and may become compromised [1,6,18,33,34,39]. Under normal circumstances, the contralateral VA gets compressed at 308 rotation of the neck; with rotation beyond 458, the ipsilateral VA also begins to get compressed [27,38]. The long and redundant
course of the third segment of VA is a physiologic adaptation that is essential to accommodate for the stretching of VA on turning the neck to the opposite side. The laxity present in this segment of VA compensates for the excessive movements of VA between the fixed points provided by the C1 and C2 foramen transversarium as well as the fibrous dural entry of the VA. 4.2. Review of literature Craniovertebral junction anomalies have often been associated with anomalies of VA. In isolated basilar impression, the terminal part of VA, including the first part of BA, may be displaced dorsally [16]; a high incidence of supernumerary coiling of VA has also been reported [24]. In cases with assimilation of the posterior arch of atlas, VA may enter the dura more anteriorly and occasionally through a separate bony canal [25]. Permanent changes in the caliber of the vertebrobasilar vasculature have been noted, leading to obstruction or dissection. The changes range from complex interruption of vascular filling of BA to significant reduction in the lumen of VA with compensatory hypertrophic changes in the other VA [9]. Klausberger and Samec [17] found 30% or greater reduction in the diameter of VA in patients with an occipitalized atlas. Bernini et al [3] found
Fig. 4. Unsubtracted normal anteroposterior A: and lateral B: angiogram showing the third segment of VA with its tortuous course and its proximal (curved arrow) and distal (straight arrow) loops. Schematic diagrams showing C: normal C1 through C2 orientation and the third segment of VA with its normal proximal (curved arrow) and distal (straight arrow) loops, and D: showing AAD with opening of the distal loop of VA as it emerges from the C1 foramen transversarium and traverses on the dorsum of C1 posterior arch (straight arrow) and by a shortened and stretched portion of VA forming the proximal loop after emerging from the C2 foramen transversarium (curved arrow).
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
VA to be anomalous in 24% of their patients with CVJ anomalies. These VA anomalies included narrowing and anomalous ending in posterior inferior cerebellar artery or the presence of a filiform artery. Vertebral artery abnormalities in the form of dolichoectasia, focal narrowing, elongation, arterial looping, or abnormal entry into an occipital artery in cases of CVJ anomalies associated with posterior circulation strokes have been noted. It is postulated that these changes made VA more susceptible to low-grade trauma and slowing of circulation [24]. Thus, CVJ anomalies may often be associated with dissection or occlusion of VA [4,8,10,14,22,29,32]. The vulnerability of VA to occlude at the C1 through C2 level has also been noted [1,15,39]. Menezes et al [20], in their study on CVJ anomalies, performed angiogram in 9 of their symptomatic patients and demonstrated complete block in 2 patients in the third segment. Vertebral artery obstruction at the C1 through C2 level resulting in VBI has also been studied by Sarathchandra et al [24]. Rotational C1 through C2 VA occlusion causing VBI (known as the bBow Hunter’sQ stroke) is a well-described entity [19,21,26, 28,30,36]. 4.3. Vertebral artery obstruction associated with AAD in the present study In our patients with AAD, VA obstruction was also seen in this third segment. The patient’s contralateral VA showed a short and stretched third segment (called the stretched loop sign) of VA at the C1 through C2 level (Fig. 4D). This was evident on angiograms by (a) opening of the distal loop of the VA as it emerges from the foramen transversarium of the atlas and traverses on the dorsum of the posterior arch of atlas, (b) shortened and stretched VA that runs laterally and posteriorly, forming the proximal loop after emerging from the foramen transversarium of the axis, or (c) both. Because the course of bilateral vertebral arteries is generally similar, perhaps this stretching predisposed to thrombosis and occlusion of VA on one side. The occipitoatlantoaxial complex forms the most mobile segment of the spine [38]. When AAD is present, the atlas slips forward, relative to the axis. Thus, an increase in the distance and the obliquity of the respective foramen transversarium of C1 and C2 also occurs. Because the artery is firmly anchored in the transverse foramina of the axis and atlas, an atlas that intermittently slides over the axis brings about stretch and trauma to the small segment of the artery that traverses between these 2 fixed points. A high incidence of occipitalized atlas and C2 through C3 fusion was also noted in our series. In the latter patients, the portions of VA between C2 and C3 vertebral bodies and that beyond the foramen transversarium of atlas are enclosed in separate bony tunnels, leaving a completely unprotected segment between C1 and C2 where all movements of the upper cervical spine are taking place. This probably exaggerates the trauma occurring at this part of the VA.
303
Repeated motion, kinking, and stretching of an already shortened VA may cause intimal tears that serve as a nidus for platelet aggregation and thrombus formation [24]. The thrombosis may extend along distal branches but may often not be detected because an adequate circle of Willis may prevent the development of VBI, and milder symptoms due to posterior circulation infarction are often not investigated [11,35,37]. Multiple, bilateral, posterior circulation infarcts occur mainly because of embolization from the thrombosed artery. In 1 of our patients, the thromboembolic episode also caused a simultaneous proximal basilar block (case 3). In the present study, occipito-C1 through C2 posterior fusion restricted the normal C1 through C2 rotation and the exaggerated translational movement (because of AAD), thus preventing stretching and further injury to the anomalous VA, but at the cost of significant restriction of neck movements [2,5,13,20,23]. 5. Conclusions Vertebrobasilar territory infarction may be the sole manifestation of AAD, as seen in the present series. Atlantoaxial dislocation should be looked for whenever the cause of posterior circulation stroke remains elusive. In this clinical situation, a coexisting AAD may often be missed unless dynamic lateral radiographs of the CVJ are undertaken. The complete diagnostic protocol for evaluation of AAD should include a DSA or MRA. The presence of a bstraight loop signQ of VA may be a valuable pointer toward an impending VBI. C1 through C2 stabilization in these patients may prevent further thromboembolic episodes. References [1] Barton JW, Margolis MT. Rotational obstruction of the vertebral artery at the atlanto-axial joint. Neuroradiology 1975;9:117 - 20. [2] Behari S, Bhargava V, Nayak S, Kirankumar MV, Banerji D, Chhabra DK, Jain VK. Congenital reducible atlantoaxial dislocation: classification and surgical considerations. Acta Neurochir (Wien) 2002;144:1165 - 77. [3] Bernini FP, Elefante R, Smatino F, Tedeschi G. Angiographic study on the vertebral artery in cases of deformities in craniocervical joints. AJR Am J Roentgenol 1969;107:526 - 9. [4] Bhatnagar M, Sponseller PD, Carol C, Tolo VT. Pediatric atlanto axial instability presenting as cerebellar and cerebral infarct. J Pediatr Orthop 1991;11:103 - 7. [5] Brooks AL, Jenkins EB. Atlantoaxial arthrodesis by the wedge compression method. J Bone Joint Surg (Am) 1978;60:279 - 84. [6] Brown BSJ, Tatlow WF. Radiographic studies of the vertebral arteries in cadavers: effect of position and traction on the head. Radiology 1963;81:80 - 8. [7] Cacciola F, Phalke U, Goel A. Vertebral artery in relationship to C1C2 vertebrae: an anatomical study. Neurol India 2004;52:178 - 84. [8] Chakera TMH, Anderson JEM, Edis RH. Atlanto axial dislocations and vertebral artery aneurysm. Br J Radiol 1982;55:863 - 4. [9] Djinjian R, Pansini A. L’arteriographie vertebrale dans les cervicarthrosis et les malformations de la charneiere cervico-occipitale. Rev Neurol 1962;106:691 - 8. [10] Dumas JL, Salama J, Dreyfus P, Thoreux P, Goldlust D, Chevrel JP. Magnetic resonance angiographic analysis of atlanto axial rotation:
304
[11]
[12]
[13]
[14] [15]
[16] [17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304 anatomic bases of compression of the vertebral arteries. Surg Radiol Anat 1996;18:303 - 13. Faris AA, Poser CM, Wilmore DW, Agnew CH. Radiologic visualization of the neck vessels in healthy men. Neurology 1963;13:386 - 96. Hodak JA, Mamourian A, Dean BL. Radiologic evaluation of the craniovertebral junction. In: Dickman CA, Spetzler RF, Sonntag VKH, editors. Surgery of the Craniovertebral Junction. New York7 Thieme; 1998. p. 81 - 102. Jain VK, Behari S. Posterior occipitoaxial fusion for atlantoaxial dislocation associated with occipitalized atlas. In: Wilkins R, Rengachary S, editors. Neurosurgical Operative Atlas, vol. 7. Illinois7 American Association of Neurological Surgeons; 1997. p. 249 - 56. Janeway R, Toole JF, Lainbach LB, Miller HS. Vertebral artery obstruction with basilar impression. Arch Neurol 1966;15:211 - 6. Jones MW, Kaufman JC. Vertebrobasilar artery insufficiency in rheumatoid atlanto axial subluxation. J Neurol Neurosurg Psychiatry 1976;39:122 - 8. Klaus E. Vertebral artery findings in basilar impressions. Radiologe 1969;9:293 - 9. Klausberger E, Samec P. Foramen retroarticulare atlantis and the vertebral angiogram. MMW Munch Med Wochenschr 1975;117: 483 - 6. Kuether TA, Nesbit GM, Clark GM, Barnwell SL. Rotational vertebral artery occlusion: a mechanism of vertebrobasilar insufficiency. Neurosurgery 1997;41:427 - 33. Matsuyama T, Morimoto T, Sakaki T. Comparison of C1-2 posterior fusion and decompression of the vertebral artery in the treatment of bow hunter’s stroke. J Neurosurg 1997;86:619 - 23. Menezes AH, VanGilder JC, Graf CJ, McDonnell DE. Craniocervical anomalies. A comprehensive surgical approach. J Neurosurg 1980;53:444 - 55. Pang D, Li V. Atlantoaxial rotatory fixation: Part 1—Biomechanics of normal rotation at the atlantoaxial joint in children. Neurosurgery 2004;55:614 - 26. Puca A, Scogna A, Rollo M. Craniovertebral junction malformation and rotational occlusion of the vertebral artery. Br J Neurosurg 2000;14:361 - 4. Ransford AO, Crockard HA, Pozo JL, Thomas NP, Nelson IW. Craniocervical instability treated by contoured loop fixation. J Bone Joint Surg (Br) 1986;68:173 - 7. Sarathchandra P, Ramdas GV, Gaikwad SB, Kale SS. Craniovertebral junction anomalies with posterior circulation strokes. Proceedings of the Annual Conference of the Neurological Society of India, Mumbai, 2001. p. 23 - 32 [December 15-18]. Schmidt H, Sartor K, Heckl RW. Bone malformation of the craniocervical junction. In: Vinken PJ, Bruyn GW, editors. Handbook of Clinical Neurology. Amsterdam7 North-Holland; 1978. p. 1 - 84. Seki T, Hida K, Akino M, Iwasaki Y. Anterior decompression of the atlantoaxial vertebral artery to treat bow hunter’s stroke: technical case report. Neurosurgery 2001;49:1474 - 6. Selecki BR. The effects of rotation of the atlas on the axis. Med J Aust 1969;1:1012 - 5.
[28] Shimazu T, Waga S, Kojima T, Niwa S. Decompression of the vertebral artery for bow hunter’s stroke: case report. J Neurosurg 1988;69:127 - 31. [29] Singer WD, Haler JS, Wolpart SM. Occlusive vertebro-basilar disease associated with cervical spine anomaly. Am J Dis Child 1975; 129:492 - 5. [30] Sorenson BF. Bow hunter’s stroke: case report. Neurosurgery 1978;2:259 - 61. [31] Taveras JM, Wood EH. Vertebral angiography. In: Taveras JM, Wood EH, editors. 2nd ed. Diagnostic Neuroradiology, vol. 2. Baltimore7 Williams and Wilkins; 1977. p. 778 - 80. [32] Tetsuya T, Satoru H, Haruhiko H, Koichiro H, Shigiyatsu E, Tadashi S. Os odontoideum with vertebral artery occlusion. Spine 1994; 19:460 - 2. [33] Toole JF, Tucker SH. Influence of head position upon cerebral circulation. Arch Neurol 1960;2:616 - 23. [34] Traflet RF, Babaria AR, Bell RD, Sollenberger SE, Doan HT, Gonzalez CF, Mishkin MM. Vertebral artery dissection after rapid head turning. AJNR Am J Neuroradiol 1989;10:650 - 1. [35] Troost BT. Dizziness and vertigo in vertebrobasilar disease. Part II: Central causes and vertebrobasilar disease. Stroke 1980;11:413 - 5. [36] Vilela MD, Goodkin R, Lundin DA, Newell DW. Rotational vertebrobasilar ischaemia: hemodynamic assessment and surgical treatment. Neurosurgery 2005;56:36 - 45. [37] Vinchon M, Assaker R, Leclerc X, Lejeune JP. Vertebrobasilar insufficiency resulting from traumatic atlantoaxial instability: case report. Neurosurgery 1955;36:839 - 41. [38] White III AA, Panjabi MM. The clinical biomechanics of the occipitoatlantoaxial complex. Orthop Clin North Am 1978;9:867 - 78. [39] Yang PJ, Latack JT, Gabrielson TO. Rotational vertebral artery occlusion at C1-C2. AJNR Am J Neuroradiol 1985;6:96 - 100.
Commentary The authors present 7 individuals who have radiographic evidence of vertebrobasilar abnormality as a result of atlantoaxial craniovertebral abnormality. The routine use of CT angiography and MRA in the evaluation of craniovertebral abnormalities is pertinent, especially with the findings of the authors. The senior author has significant experience in craniovertebral abnormalities, and the findings here reflect their experience. Arnold H. Menezes, MD Department of Neurosurgery University of Iowa Hospital Iowa City, IA 52242, USA
Imaging
bStretched loop signQ of the vertebral artery: a predictor of vertebrobasilar insufficiency in atlantoaxial dislocation Vijay Sawlani, MDa, Sanjay Behari, MCh, DNBb,4, Pravin Salunke, MSb, Vijendra K. Jain, MChb, Rajendra V. Phadke, MDc a Department of Neuroradiology, Morriston Hospital, Swansea, SA6 6NL, UK Departments of bNeurosurgery and cNeuroradiology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow 226014, India Received 20 August 2005; accepted 2 February 2006
Abstract
Background: Vertebrobasilar territory infarction is one of the rarer presentations of CVJ anomalies. A new radiologic sign due to stretching of the short third segment of VA detected on MRA/DSA may identify patients of AAD at risk of developing VBI. Methods: Seven patients who presented with VBI were found to have a coexisting mobile (n = 6) or fixed (n = 1) AAD. None of these patients had the presence of any of the known risk factors for cerebrovascular disease. On identification of VBI on CT/MRI, DSA (n = 7) and MRA (n = 1) were performed to assess bilateral vertebral arteries. The course of normal VA was also studied in 5 control patients without AAD or VBI. Results: Digital subtraction angiography/MRA showed obstruction of VA at the C1 through C2 level on one side in each of these cases. The third segment of the contralateral VA showed a shortened and straighter loop termed as the stretched loop sign of the VA. On DSA, the latter manifested as (a) opening of the distal loop of the VA as it emerges from the foramen transversarium of the atlas and traverses on the dorsum of the posterior arch of atlas (n = 3), (b) shortened and stretched VA that runs laterally and posteriorly forming the proximal loop after emerging from the foramen transversarium of the axis (n = 2), or (c) both (n = 2). All patients presented with the clinical manifestations of VBI. Only 2 of these had preexisting myelopathy and long tract signs conventionally attributable to AAD. Conclusion: Vertebrobasilar territory infarction in AAD may occur because of the obstruction of the third segment of VA. A shorter and straighter loop of the third segment of VA coexisting with an abnormal translational mobility between the atlas and the axis may be the etiopathogenetic factor. D 2006 Elsevier Inc. All rights reserved.
Keywords:
Vertebral artery; Ischemia; Atlantoaxial dislocation; Cervical spine
1. Introduction Vertebral artery is particularly vulnerable to obstruction in patients with AAD in whom an abnormal translational Abbreviations: AAD, atlantoaxial dislocation; BA, basilar artery; CT, computed tomography; CVJ, craniovertebral junction; DSA, digital subtraction angiography; MRA, magnetic resonance angiography; MRI, magnetic resonance imaging; VA, vertebral artery; VBI, vertebrobasilar territory infarction. 4 Corresponding author. Fax: +91 522 2668078, 2668017. E-mail addresses: [email protected], [email protected] (S. Behari). 0090-3019/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2006.02.032
movement exists between the atlas and axis [1,4,6,33,34]. The clinical manifestations in AAD are mainly due to compression of the thecal sac and spinal cord, and rarely due to VBI. In the conventional evaluation of the CVJ anomalies, however, the bony anomalies are investigated thoroughly, whereas VA often remains unevaluated because DSA or MRA does not form a part of the conventional radiologic investigative protocol. In this study, the radiologic features of 7 patients with VBI due to AAD have been presented. Their DSA revealed occlusion of VA on one side and an abnormally short and straight loop of the third segment of VA at C1 through C2 level on the other side. The
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
radiologic findings showing the presence of a shorter and straighter loop of the third segment of VA and an abnormal translational mobility between the atlas and axis help in identifying the subset of patients prone to developing VBI. 2. Material and methods Over a 12-year period between 1990 and 2002, in 7 patients with VBI, AAD was implicated as an etiologic factor for the underlying VA occlusion. The patients underwent dynamic (in flexion and extension) trans-table lateral radiographs (n = 7) and intrathecal contrast computerized tomography (n = 2) of the CVJ to assess for reducibility at the C1-2 joints and the associated bony anomalies. An MRI of the cervical spine and brain was also performed to assess for the associated soft-tissue abnormalities, the extent of VBI, and the degree of cervicomedullary compression. On identification of VBI on CT/MRI, DSA (n = 7) and MRA (n = 1) were also performed to study the course of both vertebral arteries. The tests for cerebrovascular risk factors and cardiovascular diseases and the serological tests for vasculitis were negative in these patients. Six patients were managed by stabilizing their neck movements using a hard cervical collar until they underwent posterior occipito-C2 or C1 through C2 stabilization using Jain’s or modified Brooks’ technique, respectively [2,5,13]. In patients with mobile AAD, reducibility
299
was determined by neck flexion and extension only and not cervical traction. The patient with fixed AAD underwent Crutchfield’s cervical traction for a period of 2 weeks. When the AAD was found irreducible even on traction, the patient underwent transoral decompression of odontoid before his posterior stabilization [2,5,13,20,23] under the same anesthesia. In 1 patient, surgery was deferred in view of his poor neurologic status (case 3). The course of normal VA was also studied in 5 control patients without AAD or VBI in whom a cranial DSA was being performed for a supratentorial pathology. 3. Results These patients (male/female ratio, 5:2; age range, 20-62 years; mean age, 33 years) presented with vertigo (n = 6), ataxia (n = 5), spastic quadriparesis (n = 4), dysarthria (n = 4), nystagmus and diplopia (n = 4), and transient loss of consciousness (n = 2). A short neck was seen in 3 patients and a low hairline in 1 patient. Only 2 of these had preexisting myelopathy and long tract signs before the development of VBI. The AAD was reducible in 6 patients and fixed in 1 patient. Occipitalization of atlas was present in 6 of the cases and C2 through C3 fusion in 3 of the cases. The third segment of the left VA was occluded in 4 and the right VA in 3 patients. In all these patients, the third segment of the contralateral VA showed a short and stretched loop at
Fig. 1. Case 1: Axial CT scan A: shows bilateral occipital lobe and B: cerebellar infarction. Left vertebral angiogram lateral view C: shows VA occlusion at the C1 through C2 level (arrow). Right vertebral angiogram anteroposterior D: view shows opening of the distal loop of the third segment (arrow) of VA as it emerges from the foramen transversarium of atlas and traverses on the dorsum of C1 posterior arch.
300
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
the C1 through C2 level. On DSA, the latter was evident by (a) opening of the distal loop of the VA as it emerges from the foramen transversarium of the atlas and traverses on the dorsum of the posterior arch of atlas (n = 3), (b) shortened and stretched VA that runs laterally and posteriorly forming the proximal loop after emerging from the foramen transversarium of axis (n = 2), or (c) both (n = 2). In all the 5 control patients, the third segment of bilateral vertebral arteries showed long and redundant distal and proximal loops. There was no evidence of either opening of the distal loop or shortening and stretching of the proximal loop of the third segment of VA. 3.1. Case 1: KD A 62-year-old woman presented with right-sided weakness and incoordination after an episode of sudden vertigo with a momentary lapse of consciousness 2 weeks before admission. She also had transient dysarthria that improved in a week’s time. There was no previous history of transient ischemic attacks, migraine, diabetes, hypertension, or cardiac disease. The neurologic examination revealed gaze-evoked nystagmus and right-sided ataxic hemiparesis. The CT revealed bilateral occipital lobe and cerebellar infarcts (Fig. 1A and B). The lateral radiographs of the cervical spine showed occipitalization of the atlas with a 10-mm AAD on flexion. The DSA showed a left VA occlusion at the C1 through C2 level (Fig. 1C). The right vertebral angiogram anteroposterior view (Fig. 1D) showed straightening and stretching of the third segment of VA at
the C1 through C2 level manifested as the opening of the distal loop of VA as it emerged from the foramen transversarium of atlas and traversed on the dorsum of the posterior arch of atlas. There have been no further attacks during a follow-up of 2 years after C1 through C2 stabilization. 3.2. Case 2: PBS A 35-year old man developed a sudden episode of headache, vomiting, gait ataxia, and nasal intonation after a jerk to the neck 6 months ago. For the past 1 month before admission, he developed progressive spastic quadriparesis. On examination, he had IX and X cranial nerve paresis with cerebellar signs and spastic quadriparesis. His dynamic trans-table radiograph of the CVJ revealed a mobile AAD of 7 mm with an assimilated posterior arch of atlas. The CT scan of the head revealed infarcts in the occipital and cerebellar regions (Fig. 2A and B). The dynamic MRI of the CVJ in flexed position revealed AAD causing compression of cervicomedullary junction and, in extension, showed reduction of AAD (Fig. 2C and D). The DSA showed occlusion of right VA at the level of axis (Fig. 2E and F). The third segment of left VA was short and straight. This was evident both by the opening of the distal loop of VA as it emerged from the foramen transversarium of atlas and traversed on the dorsum of posterior arch of atlas and by a shortened and stretched portion of VA that ran laterally and posteriorly forming the proximal loop after emerging from the foramen transversarium of axis
Fig. 2. Case 2: Axial CT scan A: shows occipital lobe and B: cerebellar infarction. Dynamic T2-weighted sagittal image in flexion C: shows AAD causing obliteration of subarachnoid space and compression of cervicomedullary junction, and in extension D: shows reduction of AAD. Right vertebral angiogram unsubtracted lateral E: and subtracted anteroposterior F: views show occlusion of VA at the C1 through C2 level (arrow). Left vertebral angiogram, unsubtracted lateral G: and subtracted anteroposterior H: views show straightening and shortening of the third segment of VA manifested both by the opening of the distal loop of VA as it emerges from the C1 foramen transversarium and traverses on the dorsum of C1 posterior arch (straight arrow), and by a shortened and stretched portion of VA that runs laterally and posteriorly forming the proximal loop after emerging from the C2 foramen transversarium (curved arrow).
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
(Fig. 2G and H). He underwent an occipitocervical fusion using sublaminar wires and the placement of strut and onlay bone grafts [2,13]. The patient was kept on a cervical collar for 3 months. There was a significant improvement in spasticity, and his power normalized. However, a mild gait ataxia persisted. 3.3. Case 3: RB This 35-year-old man had sudden onset of severe headache and recurrent vomiting associated with vertigo, scanning speech, and left-sided appendicular ataxia 1 month before admission. After admission, while he was being investigated, he developed a sudden loss of consciousness with absent doll’s eye movements and a decerebrate posturing. His MRI showed multiple brainstem and cerebellar infarcts (Fig. 3A). His dynamic (in flexion and extension) lateral cervical radiographs and the intrathecal dynamic contrast CT of the CVJ revealed a mobile AAD with an occipitalized posterior arch of atlas and C2 through C3 fusion (Fig. 3B and C). There was significant thecal compression in flexion. The left vertebral angiogram showed occlusion of VA at the level of C1 through C2 (Fig. 3D). The right vertebral DSA showed a shortened and stretched portion of VA that ran laterally and posteriorly forming the proximal loop after emerging from the foramen
301
transversarium of axis. There was complete filling of right VA until the origin of the BA with reflux into the left VA until its horizontal segment on the posterior arch of atlas (Fig. 3E). There was an abrupt cutoff of the dye column at the lower one-third of the BA and no forward filling of the BA. The right internal carotid artery injection showed filling of distal BA and the posterior cerebral arteries through the posterior communicating arteries. However, the proximal and mid-segmental basilar trunk showed no filling (Fig. 3F). The angiographic impression was that of VA thrombosis with migration of clot to the basilar trunk. In view of a bad prognosis related to his poor neurologic condition and the presence of established multiple brainstem and cerebellar infarcts, surgery for an occipitocervical posterior stabilization was refused by the patient’s relatives.
4. Discussion 4.1. Anatomical considerations Vertebral artery has traditionally been described in 4 segments. The first part constitutes the segment from its origin to entry into the foramen transversarium of the C6 vertebra; the second part courses vertically through the foramina transversarium of C6 to C2 vertebrae; the third
Fig. 3. Case 3: T2-weighted axial magnetic resonance image A: shows cerebellar infarction. Intrathecal contrast CT scan in flexion B: and extension C: shows mobile AAD with occipitalized posterior arch of atlas and C2 through C3 fusion. There is significant thecal compression in flexed position of the neck. Unsubtracted left vertebral angiogram D: shows occlusion of VA at the level of C1 through C2 (arrow). Subtracted right vertebral angiogram E: shows a shortened and stretched portion of VA that runs laterally and posteriorly forming the proximal loop after emerging from the foramen transversarium of the axis (arrow). There is a complete filling of the right VA until the origin of the BA with reflux into the left VA until its horizontal segment on the posterior arch of atlas, and an abrupt cutoff of the dye column at the lower one-third of BA (arrow). The right internal carotid artery injection F: shows reflux filling of distal BA and posterior cerebral arteries through the posterior communicating arteries. The proximal and mid-segmental basilar trunk shows no filling (arrow).
302
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
part extends from the foramen transversarium of C2 vertebra to its entry into the subarachnoid space; and, the fourth part courses within the subarachnoid space until it joins the opposite VA to form BA at the lower border of pons [7,12,31]. The third segment has a very tortuous course with 2 loops (Fig. 4A-C). After exiting from C2 transverse foramen, this segment of VA runs laterally and posteriorly over a short distance, forming the proximal loop, and then cranially to enter the transverse foramen of atlas. It then runs an oblique course on the dorsum of the posterior arch of atlas forming the distal loop before penetrating the posterior atlantoaxial membrane and the dura to enter the subarachnoid space [12,31]. At the C1 through C2 level, VA is particularly prone to mechanical compression during head and neck rotation. When the head is rotated, the atlantoaxial joint on the side to which the head is turned is fixed. On the opposite side, the C1 arch moves forward on the axis. Thus, the segment of VA between C1 and C2 gets stretched and may become compromised [1,6,18,33,34,39]. Under normal circumstances, the contralateral VA gets compressed at 308 rotation of the neck; with rotation beyond 458, the ipsilateral VA also begins to get compressed [27,38]. The long and redundant
course of the third segment of VA is a physiologic adaptation that is essential to accommodate for the stretching of VA on turning the neck to the opposite side. The laxity present in this segment of VA compensates for the excessive movements of VA between the fixed points provided by the C1 and C2 foramen transversarium as well as the fibrous dural entry of the VA. 4.2. Review of literature Craniovertebral junction anomalies have often been associated with anomalies of VA. In isolated basilar impression, the terminal part of VA, including the first part of BA, may be displaced dorsally [16]; a high incidence of supernumerary coiling of VA has also been reported [24]. In cases with assimilation of the posterior arch of atlas, VA may enter the dura more anteriorly and occasionally through a separate bony canal [25]. Permanent changes in the caliber of the vertebrobasilar vasculature have been noted, leading to obstruction or dissection. The changes range from complex interruption of vascular filling of BA to significant reduction in the lumen of VA with compensatory hypertrophic changes in the other VA [9]. Klausberger and Samec [17] found 30% or greater reduction in the diameter of VA in patients with an occipitalized atlas. Bernini et al [3] found
Fig. 4. Unsubtracted normal anteroposterior A: and lateral B: angiogram showing the third segment of VA with its tortuous course and its proximal (curved arrow) and distal (straight arrow) loops. Schematic diagrams showing C: normal C1 through C2 orientation and the third segment of VA with its normal proximal (curved arrow) and distal (straight arrow) loops, and D: showing AAD with opening of the distal loop of VA as it emerges from the C1 foramen transversarium and traverses on the dorsum of C1 posterior arch (straight arrow) and by a shortened and stretched portion of VA forming the proximal loop after emerging from the C2 foramen transversarium (curved arrow).
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304
VA to be anomalous in 24% of their patients with CVJ anomalies. These VA anomalies included narrowing and anomalous ending in posterior inferior cerebellar artery or the presence of a filiform artery. Vertebral artery abnormalities in the form of dolichoectasia, focal narrowing, elongation, arterial looping, or abnormal entry into an occipital artery in cases of CVJ anomalies associated with posterior circulation strokes have been noted. It is postulated that these changes made VA more susceptible to low-grade trauma and slowing of circulation [24]. Thus, CVJ anomalies may often be associated with dissection or occlusion of VA [4,8,10,14,22,29,32]. The vulnerability of VA to occlude at the C1 through C2 level has also been noted [1,15,39]. Menezes et al [20], in their study on CVJ anomalies, performed angiogram in 9 of their symptomatic patients and demonstrated complete block in 2 patients in the third segment. Vertebral artery obstruction at the C1 through C2 level resulting in VBI has also been studied by Sarathchandra et al [24]. Rotational C1 through C2 VA occlusion causing VBI (known as the bBow Hunter’sQ stroke) is a well-described entity [19,21,26, 28,30,36]. 4.3. Vertebral artery obstruction associated with AAD in the present study In our patients with AAD, VA obstruction was also seen in this third segment. The patient’s contralateral VA showed a short and stretched third segment (called the stretched loop sign) of VA at the C1 through C2 level (Fig. 4D). This was evident on angiograms by (a) opening of the distal loop of the VA as it emerges from the foramen transversarium of the atlas and traverses on the dorsum of the posterior arch of atlas, (b) shortened and stretched VA that runs laterally and posteriorly, forming the proximal loop after emerging from the foramen transversarium of the axis, or (c) both. Because the course of bilateral vertebral arteries is generally similar, perhaps this stretching predisposed to thrombosis and occlusion of VA on one side. The occipitoatlantoaxial complex forms the most mobile segment of the spine [38]. When AAD is present, the atlas slips forward, relative to the axis. Thus, an increase in the distance and the obliquity of the respective foramen transversarium of C1 and C2 also occurs. Because the artery is firmly anchored in the transverse foramina of the axis and atlas, an atlas that intermittently slides over the axis brings about stretch and trauma to the small segment of the artery that traverses between these 2 fixed points. A high incidence of occipitalized atlas and C2 through C3 fusion was also noted in our series. In the latter patients, the portions of VA between C2 and C3 vertebral bodies and that beyond the foramen transversarium of atlas are enclosed in separate bony tunnels, leaving a completely unprotected segment between C1 and C2 where all movements of the upper cervical spine are taking place. This probably exaggerates the trauma occurring at this part of the VA.
303
Repeated motion, kinking, and stretching of an already shortened VA may cause intimal tears that serve as a nidus for platelet aggregation and thrombus formation [24]. The thrombosis may extend along distal branches but may often not be detected because an adequate circle of Willis may prevent the development of VBI, and milder symptoms due to posterior circulation infarction are often not investigated [11,35,37]. Multiple, bilateral, posterior circulation infarcts occur mainly because of embolization from the thrombosed artery. In 1 of our patients, the thromboembolic episode also caused a simultaneous proximal basilar block (case 3). In the present study, occipito-C1 through C2 posterior fusion restricted the normal C1 through C2 rotation and the exaggerated translational movement (because of AAD), thus preventing stretching and further injury to the anomalous VA, but at the cost of significant restriction of neck movements [2,5,13,20,23]. 5. Conclusions Vertebrobasilar territory infarction may be the sole manifestation of AAD, as seen in the present series. Atlantoaxial dislocation should be looked for whenever the cause of posterior circulation stroke remains elusive. In this clinical situation, a coexisting AAD may often be missed unless dynamic lateral radiographs of the CVJ are undertaken. The complete diagnostic protocol for evaluation of AAD should include a DSA or MRA. The presence of a bstraight loop signQ of VA may be a valuable pointer toward an impending VBI. C1 through C2 stabilization in these patients may prevent further thromboembolic episodes. References [1] Barton JW, Margolis MT. Rotational obstruction of the vertebral artery at the atlanto-axial joint. Neuroradiology 1975;9:117 - 20. [2] Behari S, Bhargava V, Nayak S, Kirankumar MV, Banerji D, Chhabra DK, Jain VK. Congenital reducible atlantoaxial dislocation: classification and surgical considerations. Acta Neurochir (Wien) 2002;144:1165 - 77. [3] Bernini FP, Elefante R, Smatino F, Tedeschi G. Angiographic study on the vertebral artery in cases of deformities in craniocervical joints. AJR Am J Roentgenol 1969;107:526 - 9. [4] Bhatnagar M, Sponseller PD, Carol C, Tolo VT. Pediatric atlanto axial instability presenting as cerebellar and cerebral infarct. J Pediatr Orthop 1991;11:103 - 7. [5] Brooks AL, Jenkins EB. Atlantoaxial arthrodesis by the wedge compression method. J Bone Joint Surg (Am) 1978;60:279 - 84. [6] Brown BSJ, Tatlow WF. Radiographic studies of the vertebral arteries in cadavers: effect of position and traction on the head. Radiology 1963;81:80 - 8. [7] Cacciola F, Phalke U, Goel A. Vertebral artery in relationship to C1C2 vertebrae: an anatomical study. Neurol India 2004;52:178 - 84. [8] Chakera TMH, Anderson JEM, Edis RH. Atlanto axial dislocations and vertebral artery aneurysm. Br J Radiol 1982;55:863 - 4. [9] Djinjian R, Pansini A. L’arteriographie vertebrale dans les cervicarthrosis et les malformations de la charneiere cervico-occipitale. Rev Neurol 1962;106:691 - 8. [10] Dumas JL, Salama J, Dreyfus P, Thoreux P, Goldlust D, Chevrel JP. Magnetic resonance angiographic analysis of atlanto axial rotation:
304
[11]
[12]
[13]
[14] [15]
[16] [17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
V. Sawlani et al. / Surgical Neurology 66 (2006) 298–304 anatomic bases of compression of the vertebral arteries. Surg Radiol Anat 1996;18:303 - 13. Faris AA, Poser CM, Wilmore DW, Agnew CH. Radiologic visualization of the neck vessels in healthy men. Neurology 1963;13:386 - 96. Hodak JA, Mamourian A, Dean BL. Radiologic evaluation of the craniovertebral junction. In: Dickman CA, Spetzler RF, Sonntag VKH, editors. Surgery of the Craniovertebral Junction. New York7 Thieme; 1998. p. 81 - 102. Jain VK, Behari S. Posterior occipitoaxial fusion for atlantoaxial dislocation associated with occipitalized atlas. In: Wilkins R, Rengachary S, editors. Neurosurgical Operative Atlas, vol. 7. Illinois7 American Association of Neurological Surgeons; 1997. p. 249 - 56. Janeway R, Toole JF, Lainbach LB, Miller HS. Vertebral artery obstruction with basilar impression. Arch Neurol 1966;15:211 - 6. Jones MW, Kaufman JC. Vertebrobasilar artery insufficiency in rheumatoid atlanto axial subluxation. J Neurol Neurosurg Psychiatry 1976;39:122 - 8. Klaus E. Vertebral artery findings in basilar impressions. Radiologe 1969;9:293 - 9. Klausberger E, Samec P. Foramen retroarticulare atlantis and the vertebral angiogram. MMW Munch Med Wochenschr 1975;117: 483 - 6. Kuether TA, Nesbit GM, Clark GM, Barnwell SL. Rotational vertebral artery occlusion: a mechanism of vertebrobasilar insufficiency. Neurosurgery 1997;41:427 - 33. Matsuyama T, Morimoto T, Sakaki T. Comparison of C1-2 posterior fusion and decompression of the vertebral artery in the treatment of bow hunter’s stroke. J Neurosurg 1997;86:619 - 23. Menezes AH, VanGilder JC, Graf CJ, McDonnell DE. Craniocervical anomalies. A comprehensive surgical approach. J Neurosurg 1980;53:444 - 55. Pang D, Li V. Atlantoaxial rotatory fixation: Part 1—Biomechanics of normal rotation at the atlantoaxial joint in children. Neurosurgery 2004;55:614 - 26. Puca A, Scogna A, Rollo M. Craniovertebral junction malformation and rotational occlusion of the vertebral artery. Br J Neurosurg 2000;14:361 - 4. Ransford AO, Crockard HA, Pozo JL, Thomas NP, Nelson IW. Craniocervical instability treated by contoured loop fixation. J Bone Joint Surg (Br) 1986;68:173 - 7. Sarathchandra P, Ramdas GV, Gaikwad SB, Kale SS. Craniovertebral junction anomalies with posterior circulation strokes. Proceedings of the Annual Conference of the Neurological Society of India, Mumbai, 2001. p. 23 - 32 [December 15-18]. Schmidt H, Sartor K, Heckl RW. Bone malformation of the craniocervical junction. In: Vinken PJ, Bruyn GW, editors. Handbook of Clinical Neurology. Amsterdam7 North-Holland; 1978. p. 1 - 84. Seki T, Hida K, Akino M, Iwasaki Y. Anterior decompression of the atlantoaxial vertebral artery to treat bow hunter’s stroke: technical case report. Neurosurgery 2001;49:1474 - 6. Selecki BR. The effects of rotation of the atlas on the axis. Med J Aust 1969;1:1012 - 5.
[28] Shimazu T, Waga S, Kojima T, Niwa S. Decompression of the vertebral artery for bow hunter’s stroke: case report. J Neurosurg 1988;69:127 - 31. [29] Singer WD, Haler JS, Wolpart SM. Occlusive vertebro-basilar disease associated with cervical spine anomaly. Am J Dis Child 1975; 129:492 - 5. [30] Sorenson BF. Bow hunter’s stroke: case report. Neurosurgery 1978;2:259 - 61. [31] Taveras JM, Wood EH. Vertebral angiography. In: Taveras JM, Wood EH, editors. 2nd ed. Diagnostic Neuroradiology, vol. 2. Baltimore7 Williams and Wilkins; 1977. p. 778 - 80. [32] Tetsuya T, Satoru H, Haruhiko H, Koichiro H, Shigiyatsu E, Tadashi S. Os odontoideum with vertebral artery occlusion. Spine 1994; 19:460 - 2. [33] Toole JF, Tucker SH. Influence of head position upon cerebral circulation. Arch Neurol 1960;2:616 - 23. [34] Traflet RF, Babaria AR, Bell RD, Sollenberger SE, Doan HT, Gonzalez CF, Mishkin MM. Vertebral artery dissection after rapid head turning. AJNR Am J Neuroradiol 1989;10:650 - 1. [35] Troost BT. Dizziness and vertigo in vertebrobasilar disease. Part II: Central causes and vertebrobasilar disease. Stroke 1980;11:413 - 5. [36] Vilela MD, Goodkin R, Lundin DA, Newell DW. Rotational vertebrobasilar ischaemia: hemodynamic assessment and surgical treatment. Neurosurgery 2005;56:36 - 45. [37] Vinchon M, Assaker R, Leclerc X, Lejeune JP. Vertebrobasilar insufficiency resulting from traumatic atlantoaxial instability: case report. Neurosurgery 1955;36:839 - 41. [38] White III AA, Panjabi MM. The clinical biomechanics of the occipitoatlantoaxial complex. Orthop Clin North Am 1978;9:867 - 78. [39] Yang PJ, Latack JT, Gabrielson TO. Rotational vertebral artery occlusion at C1-C2. AJNR Am J Neuroradiol 1985;6:96 - 100.
Commentary The authors present 7 individuals who have radiographic evidence of vertebrobasilar abnormality as a result of atlantoaxial craniovertebral abnormality. The routine use of CT angiography and MRA in the evaluation of craniovertebral abnormalities is pertinent, especially with the findings of the authors. The senior author has significant experience in craniovertebral abnormalities, and the findings here reflect their experience. Arnold H. Menezes, MD Department of Neurosurgery University of Iowa Hospital Iowa City, IA 52242, USA