- Email: [email protected]
Blunt cerebrovascular injuries in association with craniocervical distraction injuries: a retrospective review of consecutive cases
The Spine Journal 15 (2015) 499–505
Clinical Study
Blunt cerebrovascular injuries in association with craniocervical distraction injuries: a retrospective review of consecutive cases Marcelo D. Vilela, MDa,b, Louis J. Kim, MDb, Carlo Bellabarba, MDc, Richard J. Bransford, MDc,* a Department of Neurological Surgery, Mater Dei Hospital, Belo Hoisonte, Brazil Department of Neurological Surgery, Harborview Medical Center/University of Washington, Box 359798, 325 9th Ave, Seattle, WA 98104, USA c Department of Orthopaedics and Sports Medicine, Harborview Medical Center/University of Washington, Box 359798, 325 9th Ave, Seattle, WA 98104, USA b
Received 26 March 2014; revised 13 September 2014; accepted 7 October 2014
Abstract
BACKGROUND CONTEXT: Blunt cerebrovascular injuries (BCVIs) have the potential to cause brain, cerebellar, and/or spinal cord ischemia. Certain subtypes of spine fractures, such as vertebral subluxation, fractures through the foramen transversarium, and C1–C3 fractures have been linked to a higher incidence of BCVI. On the other hand, BCVI in association with craniocervical distraction injuries (CCDs) have been only anecdotally reported. PURPOSE: We hypothesized that because CCD is also caused by a high-energy hyperflexion/ hyperextension distraction mechanism, it could also be associated with a high incidence of BCVI. STUDY DESIGN/SETTING: Retrospective chart review. PATIENT SAMPLE: Of 46 consecutive patients with unstable craniocervical dissociations treated operatively at a single Level I trauma center from January 1996 to December 2009, 29 of the 46 had vascular studies that comprised the study sample. OUTCOME MEASURES: Primary outcomes assessed were BCVI subdivided into blunt carotid artery injuries and/or blunt vertebral artery injuries and classified according to the Biffl criteria. Secondary measures included associated strokes and evidence of emboli on transcranial Doppler. METHODS: All consecutive patients diagnosed with unstable CCD injuries that were surgically treated at a single Level I trauma center during the period of 1996 to 2009 were identified. Those who were adequately screened with a catheter angiogram and/or computed tomography angiogram of the neck so as to rule out BCVI were included in this study. Electronic medical records were used to determine mechanism, demographics, clinical findings, and transcranial Doppler reports. Angiography and computed tomography angiograms were analyzed to assess for BCVI. If a BCVI was identified, these were classified using the Biffl criteria. RESULTS: Among the 29 screened patients, 30 BCVIs were identified in 15 patients. According to the Biffl criteria, there were 13 Grade I, eight Grade II, five Grade III, three Grade IV, and one
FDA device/drug status: Not applicable. Author disclosures: MDV: Nothing to disclose. LJK: Stock Ownership: Spi Surgical, Inc. (12.5% ownership); Consulting: Microvention, Inc. (B, DSMB chair for clinical trial); Scientific Advisory Board/Other Office: Aesculap, Inc. (B); Research support (A, Investigator Salary, Staff/Materials): Volcano, Inc. (C, Paid directly to institution); Grants: NIH/NINDS, 1R01NS088072-01A1 (H, Paid directly to institution), DoD STTR grant, W81XWH-09-C-0159 (G, Paid directly to institution). CB: Speaking and/or Teaching Arrangements: AOSpine (C); Grants: DePuy Synthes (B for support of UW Spine Grant Rounds teaching program, Paid directly to institution); Fellowship Support: AOSpine North America (E for UW Spine fellowship support, Paid directly to institution). RJB: Speaking and/or Teaching Arrangements: AOSpine North America (B per year); Grants: DePuy Synthes (B per year in Sponsorship of Spine Grant Rounds, Paid directly to institution); Fellowship Support: DePuy Synthes (E per year, Paid directly to institution), AOSpine North America (E per year, Paid directly to institution). http://dx.doi.org/10.1016/j.spinee.2014.10.012 1529-9430/Ó 2015 Elsevier Inc. All rights reserved.
The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. Confirmation of data control: I, RJB, the corresponding author, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis as well as the decision to submit for publication. Funding: No funding or material support was received for the completion of this work. Conflict of interest: The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this article. This work has not been published previously. * Corresponding author. Department of Orthopaedics and Sports Medicine, Harborview Medical Center/University of Washington, Box 359798, 325 9th Ave, Seattle, WA 98104, USA. Tel.: (206) 744-3298; fax: (206) 744-3227. E-mail address: [email protected] (R.J. Bransford)
500
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
Grade V injuries. Three major strokes were diagnosed in those 15 patients with BCVI, as opposed to none among the other 14 patients without BCVI. CONCLUSIONS: Blunt cerebrovascular injuries were seen in more than 50% of the patients screened, with major strokes occurring in 20% of the patients. We suggest screening for BCVI in all patients presenting with C0–C1 and/or C1–C2 distraction injuries. Ó 2015 Elsevier Inc. All rights reserved. Keywords:
Blunt vessel injury; Carotid artery; Craniocervical distraction; Atlanto-occipital dissociation; Stroke; Vertebral artery
Introduction The occurrence of blunt cerebrovascular injuries (BCVIs) is particularly concerning because of the potential deleterious disabilities related to stroke. Recently, screening of certain trauma subpopulations with computed tomography angiograms (CTAs), magnetic resonance angiograms, or angiography has led to early diagnosis and initiation of treatment, with reported decreases in stroke rates [1]. Certain subtypes of traumatic spinal injuries, such as subaxial cervical spine subluxations, C1–C3 fractures, and fractures through the vertebral artery foramen transversarium, have been linked to both vertebral and carotid injuries [2]. With advancements in prehospital and hospital care, the number of surviving patients with craniocervical dissociation (CCD) injuries has been increasing in the past two decades [3]. Although BCVI has been reported anecdotally in patients with craniocervical injuries [4,5], the incidence of blunt vascular injuries in such instances is unknown because of the limited number of cases. Because it is well known that BCVI occurs in association with high-energy hyperextension/flexion injuries [6], we hypothesized that patients with CCD injuries would have a high incidence of BCVI because hyperextension/distraction is the primary force responsible for the CCD. Our goal was to investigate the incidence of BCVI in a series of patients treated in a single Level I trauma institution who presented with unstable craniocervical distraction injuries.
fracture that merited surgical fixation. Distraction injuries at C1–C2 were also included given the injury of the stabilizing ligaments between the dens and cranium as these too are typically included in CCD definitions [3]. Injuries of the craniocervical junction thought to be stable and treated nonoperatively were excluded. Among the patients with unstable CCDs, all who underwent screening neck CTA and/or a catheter angiography performed to rule out BCVI were included in this study. The current indications for screening for BCVI in our institution are outlined in Table 1. The official neuroradiology readings on the neck CTA and/or catheter angiograms were subdivided into blunt carotid artery injuries (BCIs) and/or blunt vertebral artery injuries (BVIs) caused by a blunt trauma mechanism, and classified according to the Biffl grading (Table 2) [7]. For the purpose of this article, a carotid cavernous fistulae was classified as Grade V. The presence of additional cervical spine, facial fractures, mandible fractures, skull base fracture, imaging features characteristic of stroke on head computed tomography (CT) and/or magnetic resonance scans, transcranial Doppler (TCD) emboli monitoring recordings, and/or any traumatic brain injury (TBI) resulting in intracranial hemorrhage were also entered in the data set. Because the investigational purpose of the study was purely identification of the incidence of BCVI among patients with craniocervical distraction injuries, no efforts at identifying treatment effectiveness and its relation to outcomes were undertaken.
Materials and methods
Results
After obtaining institutional review board approval, all patients diagnosed with unstable CCD injuries that were operatively managed at a single Level I trauma institute from January 1996 to December 2009 were retrospectively identified from a prospectively maintained database. Data on demographics, mechanism of injury, clinical examinations, imaging features, treatment, and outcomes were reviewed and collected from the patient medical records and review of radiographic imaging studies. Unstable CCDs were defined as injuries with clear disruption of the facet capsules and surrounding stabilizing ligaments and radiographic distraction and/or translation at the craniocervical junction with or without associated fractures of the occipital condyles, C1 ring, or Type I dens
We identified 46 patients with unstable CCD injuries that were treated surgically at our Level I trauma center during the period of January 1996 through December 2009. There were three patients with primarily C1–C2 distraction injuries and 43 patients with C0–C1 distraction injuries. Twenty-nine of the 46 patients were adequately screened with either neck CTA (28 of 29) and/or catheter angiogram (6 of 29) to assess for BCVIs. Fifteen of the 29 patients (52%) who underwent screening had positive imaging findings for BCVI. Mechanism of injury was a motor vehicle accident in 24 patients, car versus pedestrian in three patients, snowmobile versus car accident in one patient, and a fall off a two-story height in one patient.
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
Mean age was 28.8 years (9–62) in the 15 patients with BCVI. Associated injuries consisted of mandible fractures in four, skull base fractures in four, facial fractures in four, TBI in nine, and additional spine fractures in eight. In the 14 patients with a negative screening CTA and/or angiogram, mean age was 31.1 years (12– 58). Associated injuries in this group consisted of no mandible nor skull base fractures, facial fracture in one, TBI in seven, and additional cervical spine fractures in four patients (Table 3). We identified a total of 14 BCIs and 16 BVIs, totaling 30 BCVIs in 15 patients. There were 13 Grade I, eight Grade II, five Grade III, three Grade IV, and one Grade V injuries, according to the Biffl classification. In the seven patients with BCVI and without additional cervical spine injuries, there were 11 BCIs and four BVIs. In the eight patients with BCVIs and with additional cervical spine injuries, there were 12 BVIs and three BCIs (Fig. 1). Three major associated strokes were diagnosed in 15 patients with BCVI, as opposed to none among 14 patients without BCVI. 1. One patient was a 25-year-old male with a significant CCD (Fig. 2) who had bilateral cerebellar strokes (Fig. 3) leading to tonsillar herniation (Fig. 4) requiring a posterior fossa decompression in addition to an occipital-cervical fusion related to bilateral Grade II BVI (Fig. 5), with a good outcome. The patient was treated with 81 mg of acetylsalicylic acid (ASA) daily for 2 months. At 6 months, he had returned to independent living and work with no incapacitating deficits. 2. One patient was a 19-year-old male who had a cerebellar stroke related to a left vertebral artery occlusion and a Grade I right vertebral artery injury but remained a C4 quadriplegic related to the CCD. He was treated with 325 mg of ASA daily for 3 months. He remained a ventilator-dependent quadraplegic and died 14 months after his injury. 3. One patient was a 24-year-old male who had a left basal ganglia and internal capsule strokes with hemiparesis, as well as blindness, related to a Grade II left BCI and was left with moderate disability. He was treated initially with 325 mg of ASA per day and transitioned to 81 mg per day, which he took for 1 year. At 1-year follow-up, the patient was ambulatory with a prosthesis related to a lower extremity amputation but had full upper extremity strength. His vision did not improve. All three strokes occurred acutely. Among the 15 patients with BCVI, 14 patients had TCD emboli monitoring study during the evaluation, with only two patients having more than 20 emboli/h in at least one examination. Among the three patients with strokes, in two patients TCD examinations demonstrated no significant emboli in the intracranial vasculature.
501
Context Blunt cerebrovascular injury (BCVI) and resultant strokes are potential sequelae following craniocervical distraction injuries. The incidence and epidemiology of BCVI following craniocervical distraction has not been thoroughly investigated, however. Contribution The authors performed a retrospective review of 29 patients with unstable craniocervical distraction injuries who also had appropriate vascular screening. Approximately half the patients sustained one or more BCVIs. Three major strokes occurred among the patients with BCVIs while none were documented in patients without this concomitant injury. The authors advocate for screening for BCVI in all patients with craniocervical distraction injuries. Implications This retrospective study is confounded by selection and indication bias, thus limiting the capacity to determine the true incidence of BCVI after craniocervical trauma. Patients who did not receive appropriate vascular screening may have also had these injuries, but could not be included in this analysis. Furthermore, from a Bayesian perspective, the mere fact that the cohort under consideration had vascular studies performed indicates that they probably were at elevated risk of vascular trauma. Nonetheless, given the rarity of craniocervical dissociation and the severity of vascular sequelae if undetected, the authors’ recommendations for universal screening for BCVI are likely well founded. —The Editors Seven patients received no specific anticoagulation treatment for the BCVI (four patients had only Grade I injuries, two had intracranial hemorrhages, and one patient had severe pelvic bleeding). Seven patients were treated with antiplatelet agents (ASA) and one patient, with a diagnosis of pulmonary embolism, received warfarin. Of the 15 patients with BCVI, one was lost to follow-up, one died as an inpatient, one died 24 months postoperatively while quadriplegic, one remained ASIA (American Spinal Injury Association) A quadriplegic, one was triplegic and wheelchair bound, four were severely disabled/vegetative because of TBI, and six patients were independent with minimal and/or mild disabilities.
Discussion Although the incidence of BCVI in the general trauma population is in the range of 0.1 to slightly more than 1% [8,9], it
502
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
Table 1 Current screening criteria for patients with indications for blunt carotid and/or vertebral artery injury
Table 3 Associated injuries in those who underwent screening for BCVI
Symptoms or signs of BCVI 1. Massive epistaxis 2. Central or lateralizing neurological deficit unexplained by findings on head CT scan 3. Expanding cervical hematoma 4. Acute or subacute infarction on head CT scan 5. TIA or stroke after neck trauma 6. Horner’s syndrome 7. Cervical bruit in a patient younger than 50 years Injury mechanisms known as risk factors for BCVI 1. Midface fracture with cervical hyperextension/rotation/flexion injury 2. Complex mandible fracture and cervical hyperextension/rotation/ flexion injury 3. Closed head injury with DAI and cervical hyperextension/rotation/ flexion injury 4. Displaced midface fracture (Lefort II or III) 5. Skull base fracture involving foramen lacerum, sphenoid, mastoid, or petrous bones 6. Complex mandible fracture because of high-speed mechanism 7. Seatbelt abrasion, hanging bruise, contusion, or hematoma of the neck with cervical swelling 8. Cervical spine fracture with cervical hyperextension/rotation/flexion injury 9. Vertebral artery foramen transversarium fracture 10. Cervical vertebral subluxation 11. Cervical fracture C1 through C3
Associated injuries
BCVI present (15), n (%)
BCVI absent (14), n (%)
Mandible fractures Facial fractures Traumatic brain injury Additional spine fractures Skull base fractures
4 4 9 8 4
0 1 7 4 0
BCVI, blunt cerebrovascular injury; CT, computed tomography; TIA, transient ischemic attack; DAI, diffuse axonal injury.
can be a cause of severe disability and/or mortality among trauma patients [1]. During the past two decades, efforts have been put into diagnosing BCVIs more efficiently in a timely fashion. This has been related to improvements in the quality, noninvasiveness, and speed of CT scans and the demonstration that rapid diagnosis and initiation of therapy for BCVI can prevent or diminish brain ischemia and neurological deficits [1,10–13]. Although ideally the trauma surgeon would like to appropriately identify all patients, screening of the entire trauma population is not cost effective. Therefore, several studies attempting to identify subpopulations that would have a greater likelihood of having suffered a BCVI have been done [6,8,14]. Blunt cerebrovascular injury has been correlated with a high-speed trauma mechanism associated with hyperextension/flexion, especially in those patients who have suffered one of the following: (1) TBI with a Glascow Coma Scale !6, (2) Lefort facial fractures, (3) basilar skull fracture with carotid canal involvement, and (4) cervical
(27) (27) (64) (57) (27)
(0) (7) (50) (29) (0)
BCVI, blunt cerebrovascular injury.
spine fractures at C1–C3, foramen transversarium fractures, and/or subluxations of the cervical spine [6,8,14,15]. Craniocervical dissociation is a severe injury that commonly results in death because of hyperextension of the brain stem at or near the foramen magnum, with severe injury to the medulla oblongata and/or medullary-spinal cord junction [16–19]. A major distractive/hyperextension and high-speed mechanism has been implicated as the usual cause for CCD [17]. The resulting separation of the skull from the upper cervical spine can be seen in imaging studies as an increase in the distance between the basion and either the tip of the dens or the posterior axial line and increased distance between occipital condyles and the lateral masses of C1 on CT scans or widening of the C1–C2 joints [1,9,13]. Magnetic resonance imaging commonly shows disruption of the apical, alar ligaments, tectorial membrane, and Occ-C1 and/or C1–C2 joint capsules, along with a varied spectrum of cord abnormalities [3,20–23]. At Harborview Medical Center, Seattle, Washington Medical Center, a Level I trauma center covering a fourstate area, we have in the past two decades treated a relatively large number of patients with craniocervical distraction injuries. A concerted effort has been made to diagnose these patients as early as possible through recognition of this unique injury through rigorous education of emergency room radiologists and traumatologists as this has
Table 2 Biffl grading [6] for blunt cerebrovascular injuries Grade
Finding
1 2
Luminal irregularity or dissection with !25% stenosis Dissections with O25% luminal narrowing or a raised intimal flap Pseudoaneurysm Complete occlusion Transection of the artery, with free extravasation of contrast or significant arteriovenous fistula
3 4 5
Fig. 1. Breakdown of patients with pure ligamentous injuries in the setting of CCD versus those with associated craniocervical fractures in the setting of CCD and comparison of BCI versus BVI. CCD, craniocervical distraction injury; BCI, blunt carotid artery injury; BVI, blunt vertebral artery injury; BCVI, blunt cerebrovascular injury.
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
503
Fig. 2. (Left) Preoperative sagittal CT and (Right) parasagittal CT scan demonstrating wide separation of the occipital C1 joint and a basion-dens interval of almost 23 mm with an associated C1 anterior ring transverse fracture. CT, computed tomography.
been shown to be critical to optimizing neurological outcome in these patients [3]. As we continue to try to understand this subset of trauma patients, we have also attempted to understand associated injuries and findings and how to best work these up to optimize their care. Because CCD is usually caused by a high-speed mechanism, patients usually suffer other severe injuries that also may lead to severe disability and/or death. A high incidence
of severe TBIs, craniofacial fractures, skull base fractures, thoracic injuries, and other cervical spine injuries have been reported in patients with CCD, both in survivors and in autopsy series, which is consistent with our findings [3,17,18]. On the other hand, the occurrence of BCVI in survivors with CCD has been reported in only a few anecdotal cases [4,5], whereas tears of the carotid and basilar arteries
Fig. 3. Preoperative axial T2 image of the brain demonstrating bilateral cerebellar strokes.
Fig. 4. Preoperative sagittal T1 image of the brain demonstrating tonsillar herniation. Line shows the tonsillar hernation through the foramen magnum.
504
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
Fig. 5. Preoperative coronal computed tomography angiogram demonstrating bilateral vertebral artery injuries. The carotids and vertebral arteries were normal below the level of C1. The arrows demonstrate the locaton of the vertebral artery injuries in this patient.
have been described in postmortem series in patients with CCD related to a hyperextension mechanism [18]. We postulated that CCD could be associated with BCVI because of several reasons. First, both are usually caused by the same mechanism, that is, high speed with hyperextension/ distraction acting on the neck. Second, both CCD and BCVI are commonly associated with severe injuries of the craniofacial region, that is, head injuries, mandible fractures, facial fractures, and other cervical spine fractures. Third, the carotid arteries are intimately related to the occipital condyles and C1 vertebrae, whereas the vertebral arteries pass behind the C0–C1 joints and curve anteriorly to enter the foramen magnum, making them highly susceptible to the same distractive/hyperextension forces that cause the CCD. The importance of properly diagnosing BCVI in trauma patients is of utmost importance because it can lead to severe neurological deficits and even death related to a major stroke. Given the unstable nature of the CCD and the susceptibility for progressive neurological deficits if not operatively stabilized, it is also critical to know if a BCVI has occurred, even if a stroke has not, as this may play a role in the type of fixation used. If a patient has a vertebral artery injury on one side, the surgeon may elect to use a C2 translaminar screw on the contralateral side or a short pars screw as opposed to a pedicle screw of transarticular screw that may place the one functional vertebral artery at risk and thus contribute to a devastating stroke. Therefore, appropriate screening of patients at risk is necessary so proper initiation of treatment and identification of possible complications can be done. In our study, 15 of 29 screened patients had evidence of BCVI, with a total of 30 vessel injuries. Though a small
number of patients because of the relatively rare nature of this injury, this is the largest series of patients with CCD and screening CTA/angiogram to rule out BCVI reported to date. Our findings that more than 50% of our screened patients had a diagnosis of BCVI support our hypothesis. One could postulate that the cause for the BCVI in our patients was the presence of associated cervical spine fractures because cervical spine fractures have been linked to BCVI in more than 30% of cases and have been found as the main risk factor for BVI [2,8,24,25]. Although we found that half of our patients studied had associated cervical spine fractures, the incidence of BCVI was not affected by the presence of additional spine fractures. The finding that the patients in our series with additional cervical fractures had more BVI, as opposed to a higher occurrence of BCI in those without additional cervical fractures, is consistent with other series on BCVI [8]. Transcranial Doppler with emboli monitoring is usually performed during 3 consecutive days in our patients with diagnosed BCVI so as to evaluate the number of emboli/hour in the intracranial vasculature and also guide therapy effectiveness. Fourteen patients had TCD studies done, with only two examinations showing more than 20 emboli/h, one of them being on the patient who suffered an internal capsule/ basal ganglia stroke. In the two patients with cerebellar strokes, TCD examinations were negative for emboli. The literature lacks evidence of TCD as a guide for the management of BCVI, and more studies are clearly needed to ascertain its value in such a setting. Although BCVI can be a significant cause of morbidity/ mortality, our study was neither performed nor powered to evaluate treatment/outcomes. Therefore, no conclusions can be drawn either related to the impact of BCVI on the outcomes or its treatment effectiveness. We could, on the other hand, observe that most patients also sustained other severe injuries, such as spinal cord and/or severe head injury, with resultant disability that ultimately contributed to an unfavorable outcome. The incidence of stroke in patients with BCVI has been reported to be as high as 50% [26,27]. In our study, we found three patients (20%), among 15 with BCVI, who suffered a stroke. Because our numbers are small, we did not attempt to correlate the incidence of stroke in our series to other series. All strokes in this series were related to Grade II to IV injuries. Although mean age, number of head injuries, skull base fractures, and additional spine fractures were similar between the two groups, there were more facial fractures in those with BCVI (four mandibles, four facial fractures in 15 patients) when compared with patients without BCVI (no mandible, one facial fracture). This suggests that higher energy/forces were transferred to the craniofacial region and ultimately to the cerebrovascular structures at the neck in the group with BCVI. It is important to mention that we included in the data set any midface fracture, mandible fracture, skull base fracture, and/or TBI with intracranial hemorrhage as an
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
indicator of injury severity to the craniofacial region. We are aware that including any facial, mandible, skull base, and head injury with intracranial hemorrhage encompasses more than the suggested indications proposed by some investigators for the screening of BCVI [6]. On the other hand, others have reported that using coexisting diagnosis does not yield a reliable prediction algorithm for screening trauma patients for BCVI, and the integration of clinical suspicion based on the mechanism and associated injuries may be essential to appropriately screen patients and minimize the number of missing injuries [9], a statement that corroborates the methods and screening criteria used in this study. Certainly, there are some major weaknesses to this study. Our numbers are small with a total 29 patients with CCD and CTA and/or angiography. Placed in perspective though, this is an uncommon injury with the largest series of patients with CCD reported to date at 17 patients [3] although the largest abstract to date reports on 48 patients [28]. Second, this is a retrospective study with the shortcomings of such a study. Third, although we had 46 patients with CCD, only 29 of them had CTAs and/or angiography. Our date dates back to 1996, and thus not all patients underwent imaging as this predated our current protocol. Many of our patients, in the more modern era, did not receive emergent CTA as they were too medically unstable and were emergently taken to the intensive care unit or to the operating room for management of associated injuries. In an ideal situation, each of these patients would have undergone CTA. We do recognize the deficiencies within this study but believe that this study also has important findings and is a stepping-stone toward further studies in the management of patients with CCD. Conclusions We found that more than 50% of our patients with CCD who were properly screened had BCVI, and 20% of those with BCVI had a major stroke. We suggest screening all patients with CCD with a neck CTA and/or catheter angiogram to assess for BCVI. References [1] Eastman AL, Muraliraj V, Sperry JL, Minei JP. CTA-based screening reduces time to diagnosis and stroke rate in blunt cervical vascular injury. J Trauma 2009;67:551–6. [2] Cothren CC, Moore EE, Biffl WL, Ciesla DJ, Ray CE Jr, Johnson JL, et al. Cervical spine fracture patterns predictive of blunt vertebral artery injury. J Trauma 2003;55:811–3. [3] Bellabarba C, Mirza SK, West GA, Mann FA, Dailey AT, Newell DW, et al. Diagnosis and treatment of craniocervical dislocation in a series of 17 consecutive survivors during an 8-year period. J Neurosurg Spine 2006;4:429–40. [4] Brinkman W, Cohen W, Manning T. Posterior fossa subarachnoid hemorrhage due to an atlantooccipital dislocation. AJR Am J Roentgenol 2003;180:1476. [5] Lee C, Woodring JH, Walsh JW. Carotid and vertebral artery injury in survivors of atlanto-occipital dislocation: case reports and literature review. J Trauma 1991;31:401–7.
505
[6] Biffl WL, Cothren CC, Moore EE, Kozar R, Cocanour C, Davis JW, et al. Western Trauma Association critical decisions in trauma: screening for and treatment of blunt cerebrovascular injuries. J Trauma 2009;67:1150–3. [7] Biffl WL, Moore EE, Offner PJ, Brega KE, Franciose RJ, Burch JM. Blunt carotid arterial injuries: implications of a new grading scale. J Trauma 1999;47:845–53. [8] Berne JD, Cook A, Rowe SA, Norwood SH. A multivariate logistic regression analysis of risk factors for blunt cerebrovascular injury. J Vasc Surg 2010;51:57–64. [9] Cook A, Osler T, Gaudet M, Berne J, Norwood S. Blunt cerebrovascular injury is poorly predicted by modeling with other injuries: analysis of NTDB data. J Trauma 2011;71:114–9. [10] Biffl WL, Egglin T, Benedetto B, Gibbs F, Cioffi WG. Sixteen-slice computed tomographic angiography is a reliable noninvasive screening test for clinically significant blunt cerebrovascular injuries. J Trauma 2006;60:745–51. [11] Sliker CW. Blunt cerebrovascular injuries: imaging with multidetector CT angiography. Radiographics 2008;28:1689–708. [12] Stengel D, Rademacher G, Hanson B, Ekkernkamp A, Mutze S. Screening for blunt cerebrovascular injuries: the essential role of computed tomography angiography. Semin Ultrasound CT MR 2007;28:101–8. [13] Tikkakoski T. Screening for blunt cerebrovascular injury in trauma patients. Acta Radiol 2007;48:707. [14] Ringer AJ, Matern E, Parikh S, Levine NB. Screening for blunt cerebrovascular injury: selection criteria for use of angiography. J Neurosurg 2010;112:1146–9. [15] Cothren CC, Moore EE, Ray CE Jr, Johnson JL, Moore JB, Burch JM. Cervical spine fracture patterns mandating screening to rule out blunt cerebrovascular injury. Surgery 2007;141:76–82. [16] Gross A. Traumatic basal subarachnoid hemorrhages: autopsy material analysis. Forensic Sci Int 1990;45:53–61. [17] Imaizumi T, Sohma T, Hotta H, Teto I, Imaizumi H, Kaneko M. Associated injuries and mechanism of atlanto-occipital dislocation caused by trauma. Neurol Med Chir (Tokyo) 1995;35:385–91. [18] Ohshima T, Kondo T. Forensic pathological observations on fatal injuries to the brain stem and/or upper cervical spinal cord in traffic accidents. J Clin Forensic Med 1998;5:129–34. [19] Tepper SL, Fligner CL, Reay DT. Atlanto-occipital disarticulation. Accident characteristics. Am J Forensic Med Pathol 1990;11:193–7. [20] Chaput CD, Walgama J, Torres E, Dominguez D, Hanson J, Song J, et al. Defining and detecting missed ligamentous injuries of the occipitocervical complex. Spine 2011;36:709–14. [21] Deliganis AV, Baxter AB, Hanson JA, Fisher DJ, Cohen WA, Wilson AJ, et al. Radiologic spectrum of craniocervical distraction injuries. Radiographics 2000;20 Spec No.:S237–50. [22] Pang D, Nemzek WR, Zovickian J. Atlanto-occipital dislocation— part 2: the clinical use of (occipital) condyle-C1 interval, comparison with other diagnostic methods, and the manifestation, management, and outcome of atlanto-occipital dislocation in children. Neurosurgery 2007;61:995–1015. [23] Pang D, Nemzek WR, Zovickian J. Atlanto-occipital dislocation: part 1—normal occipital condyle-C1 interval in 89 children. Neurosurgery 2007;61:514–21. [24] Biffl WL, Moore EE, Elliott JP, Ray C, Offner PJ, Franciose RJ, et al. The devastating potential of blunt vertebral arterial injuries. Ann Surg 2000;231:672–81. [25] Miller PR, Fabian TC, Croce MA, Cagiannos C, Williams JS, Vang M, et al. Prospective screening for blunt cerebrovascular injuries: analysis of diagnostic modalities and outcomes. Ann Surg 2002;236:386–93. [26] Burlew CC, Biffl WL. Blunt cerebrovascular trauma. Curr Opin Crit Care 2010;16:587–95. [27] DiCocco JM, Fabian TC, Emmett KP, Magnotti LJ, Zarzaur BL, Bate BG, et al. Optimal outcomes for patients with blunt cerebrovascular injury (BCVI): tailoring treatment to the lesion. JAm Coll Surg 2011;212:549–57. [28] Bellabarba C, Bransford RJ, Chapman JR. Timing to diagnosis and neurological outcomes in 48 consecutive craniocervical dissociation patients. Spine J 2011;11:S57.
Clinical Study
Blunt cerebrovascular injuries in association with craniocervical distraction injuries: a retrospective review of consecutive cases Marcelo D. Vilela, MDa,b, Louis J. Kim, MDb, Carlo Bellabarba, MDc, Richard J. Bransford, MDc,* a Department of Neurological Surgery, Mater Dei Hospital, Belo Hoisonte, Brazil Department of Neurological Surgery, Harborview Medical Center/University of Washington, Box 359798, 325 9th Ave, Seattle, WA 98104, USA c Department of Orthopaedics and Sports Medicine, Harborview Medical Center/University of Washington, Box 359798, 325 9th Ave, Seattle, WA 98104, USA b
Received 26 March 2014; revised 13 September 2014; accepted 7 October 2014
Abstract
BACKGROUND CONTEXT: Blunt cerebrovascular injuries (BCVIs) have the potential to cause brain, cerebellar, and/or spinal cord ischemia. Certain subtypes of spine fractures, such as vertebral subluxation, fractures through the foramen transversarium, and C1–C3 fractures have been linked to a higher incidence of BCVI. On the other hand, BCVI in association with craniocervical distraction injuries (CCDs) have been only anecdotally reported. PURPOSE: We hypothesized that because CCD is also caused by a high-energy hyperflexion/ hyperextension distraction mechanism, it could also be associated with a high incidence of BCVI. STUDY DESIGN/SETTING: Retrospective chart review. PATIENT SAMPLE: Of 46 consecutive patients with unstable craniocervical dissociations treated operatively at a single Level I trauma center from January 1996 to December 2009, 29 of the 46 had vascular studies that comprised the study sample. OUTCOME MEASURES: Primary outcomes assessed were BCVI subdivided into blunt carotid artery injuries and/or blunt vertebral artery injuries and classified according to the Biffl criteria. Secondary measures included associated strokes and evidence of emboli on transcranial Doppler. METHODS: All consecutive patients diagnosed with unstable CCD injuries that were surgically treated at a single Level I trauma center during the period of 1996 to 2009 were identified. Those who were adequately screened with a catheter angiogram and/or computed tomography angiogram of the neck so as to rule out BCVI were included in this study. Electronic medical records were used to determine mechanism, demographics, clinical findings, and transcranial Doppler reports. Angiography and computed tomography angiograms were analyzed to assess for BCVI. If a BCVI was identified, these were classified using the Biffl criteria. RESULTS: Among the 29 screened patients, 30 BCVIs were identified in 15 patients. According to the Biffl criteria, there were 13 Grade I, eight Grade II, five Grade III, three Grade IV, and one
FDA device/drug status: Not applicable. Author disclosures: MDV: Nothing to disclose. LJK: Stock Ownership: Spi Surgical, Inc. (12.5% ownership); Consulting: Microvention, Inc. (B, DSMB chair for clinical trial); Scientific Advisory Board/Other Office: Aesculap, Inc. (B); Research support (A, Investigator Salary, Staff/Materials): Volcano, Inc. (C, Paid directly to institution); Grants: NIH/NINDS, 1R01NS088072-01A1 (H, Paid directly to institution), DoD STTR grant, W81XWH-09-C-0159 (G, Paid directly to institution). CB: Speaking and/or Teaching Arrangements: AOSpine (C); Grants: DePuy Synthes (B for support of UW Spine Grant Rounds teaching program, Paid directly to institution); Fellowship Support: AOSpine North America (E for UW Spine fellowship support, Paid directly to institution). RJB: Speaking and/or Teaching Arrangements: AOSpine North America (B per year); Grants: DePuy Synthes (B per year in Sponsorship of Spine Grant Rounds, Paid directly to institution); Fellowship Support: DePuy Synthes (E per year, Paid directly to institution), AOSpine North America (E per year, Paid directly to institution). http://dx.doi.org/10.1016/j.spinee.2014.10.012 1529-9430/Ó 2015 Elsevier Inc. All rights reserved.
The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. Confirmation of data control: I, RJB, the corresponding author, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis as well as the decision to submit for publication. Funding: No funding or material support was received for the completion of this work. Conflict of interest: The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this article. This work has not been published previously. * Corresponding author. Department of Orthopaedics and Sports Medicine, Harborview Medical Center/University of Washington, Box 359798, 325 9th Ave, Seattle, WA 98104, USA. Tel.: (206) 744-3298; fax: (206) 744-3227. E-mail address: [email protected] (R.J. Bransford)
500
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
Grade V injuries. Three major strokes were diagnosed in those 15 patients with BCVI, as opposed to none among the other 14 patients without BCVI. CONCLUSIONS: Blunt cerebrovascular injuries were seen in more than 50% of the patients screened, with major strokes occurring in 20% of the patients. We suggest screening for BCVI in all patients presenting with C0–C1 and/or C1–C2 distraction injuries. Ó 2015 Elsevier Inc. All rights reserved. Keywords:
Blunt vessel injury; Carotid artery; Craniocervical distraction; Atlanto-occipital dissociation; Stroke; Vertebral artery
Introduction The occurrence of blunt cerebrovascular injuries (BCVIs) is particularly concerning because of the potential deleterious disabilities related to stroke. Recently, screening of certain trauma subpopulations with computed tomography angiograms (CTAs), magnetic resonance angiograms, or angiography has led to early diagnosis and initiation of treatment, with reported decreases in stroke rates [1]. Certain subtypes of traumatic spinal injuries, such as subaxial cervical spine subluxations, C1–C3 fractures, and fractures through the vertebral artery foramen transversarium, have been linked to both vertebral and carotid injuries [2]. With advancements in prehospital and hospital care, the number of surviving patients with craniocervical dissociation (CCD) injuries has been increasing in the past two decades [3]. Although BCVI has been reported anecdotally in patients with craniocervical injuries [4,5], the incidence of blunt vascular injuries in such instances is unknown because of the limited number of cases. Because it is well known that BCVI occurs in association with high-energy hyperextension/flexion injuries [6], we hypothesized that patients with CCD injuries would have a high incidence of BCVI because hyperextension/distraction is the primary force responsible for the CCD. Our goal was to investigate the incidence of BCVI in a series of patients treated in a single Level I trauma institution who presented with unstable craniocervical distraction injuries.
fracture that merited surgical fixation. Distraction injuries at C1–C2 were also included given the injury of the stabilizing ligaments between the dens and cranium as these too are typically included in CCD definitions [3]. Injuries of the craniocervical junction thought to be stable and treated nonoperatively were excluded. Among the patients with unstable CCDs, all who underwent screening neck CTA and/or a catheter angiography performed to rule out BCVI were included in this study. The current indications for screening for BCVI in our institution are outlined in Table 1. The official neuroradiology readings on the neck CTA and/or catheter angiograms were subdivided into blunt carotid artery injuries (BCIs) and/or blunt vertebral artery injuries (BVIs) caused by a blunt trauma mechanism, and classified according to the Biffl grading (Table 2) [7]. For the purpose of this article, a carotid cavernous fistulae was classified as Grade V. The presence of additional cervical spine, facial fractures, mandible fractures, skull base fracture, imaging features characteristic of stroke on head computed tomography (CT) and/or magnetic resonance scans, transcranial Doppler (TCD) emboli monitoring recordings, and/or any traumatic brain injury (TBI) resulting in intracranial hemorrhage were also entered in the data set. Because the investigational purpose of the study was purely identification of the incidence of BCVI among patients with craniocervical distraction injuries, no efforts at identifying treatment effectiveness and its relation to outcomes were undertaken.
Materials and methods
Results
After obtaining institutional review board approval, all patients diagnosed with unstable CCD injuries that were operatively managed at a single Level I trauma institute from January 1996 to December 2009 were retrospectively identified from a prospectively maintained database. Data on demographics, mechanism of injury, clinical examinations, imaging features, treatment, and outcomes were reviewed and collected from the patient medical records and review of radiographic imaging studies. Unstable CCDs were defined as injuries with clear disruption of the facet capsules and surrounding stabilizing ligaments and radiographic distraction and/or translation at the craniocervical junction with or without associated fractures of the occipital condyles, C1 ring, or Type I dens
We identified 46 patients with unstable CCD injuries that were treated surgically at our Level I trauma center during the period of January 1996 through December 2009. There were three patients with primarily C1–C2 distraction injuries and 43 patients with C0–C1 distraction injuries. Twenty-nine of the 46 patients were adequately screened with either neck CTA (28 of 29) and/or catheter angiogram (6 of 29) to assess for BCVIs. Fifteen of the 29 patients (52%) who underwent screening had positive imaging findings for BCVI. Mechanism of injury was a motor vehicle accident in 24 patients, car versus pedestrian in three patients, snowmobile versus car accident in one patient, and a fall off a two-story height in one patient.
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
Mean age was 28.8 years (9–62) in the 15 patients with BCVI. Associated injuries consisted of mandible fractures in four, skull base fractures in four, facial fractures in four, TBI in nine, and additional spine fractures in eight. In the 14 patients with a negative screening CTA and/or angiogram, mean age was 31.1 years (12– 58). Associated injuries in this group consisted of no mandible nor skull base fractures, facial fracture in one, TBI in seven, and additional cervical spine fractures in four patients (Table 3). We identified a total of 14 BCIs and 16 BVIs, totaling 30 BCVIs in 15 patients. There were 13 Grade I, eight Grade II, five Grade III, three Grade IV, and one Grade V injuries, according to the Biffl classification. In the seven patients with BCVI and without additional cervical spine injuries, there were 11 BCIs and four BVIs. In the eight patients with BCVIs and with additional cervical spine injuries, there were 12 BVIs and three BCIs (Fig. 1). Three major associated strokes were diagnosed in 15 patients with BCVI, as opposed to none among 14 patients without BCVI. 1. One patient was a 25-year-old male with a significant CCD (Fig. 2) who had bilateral cerebellar strokes (Fig. 3) leading to tonsillar herniation (Fig. 4) requiring a posterior fossa decompression in addition to an occipital-cervical fusion related to bilateral Grade II BVI (Fig. 5), with a good outcome. The patient was treated with 81 mg of acetylsalicylic acid (ASA) daily for 2 months. At 6 months, he had returned to independent living and work with no incapacitating deficits. 2. One patient was a 19-year-old male who had a cerebellar stroke related to a left vertebral artery occlusion and a Grade I right vertebral artery injury but remained a C4 quadriplegic related to the CCD. He was treated with 325 mg of ASA daily for 3 months. He remained a ventilator-dependent quadraplegic and died 14 months after his injury. 3. One patient was a 24-year-old male who had a left basal ganglia and internal capsule strokes with hemiparesis, as well as blindness, related to a Grade II left BCI and was left with moderate disability. He was treated initially with 325 mg of ASA per day and transitioned to 81 mg per day, which he took for 1 year. At 1-year follow-up, the patient was ambulatory with a prosthesis related to a lower extremity amputation but had full upper extremity strength. His vision did not improve. All three strokes occurred acutely. Among the 15 patients with BCVI, 14 patients had TCD emboli monitoring study during the evaluation, with only two patients having more than 20 emboli/h in at least one examination. Among the three patients with strokes, in two patients TCD examinations demonstrated no significant emboli in the intracranial vasculature.
501
Context Blunt cerebrovascular injury (BCVI) and resultant strokes are potential sequelae following craniocervical distraction injuries. The incidence and epidemiology of BCVI following craniocervical distraction has not been thoroughly investigated, however. Contribution The authors performed a retrospective review of 29 patients with unstable craniocervical distraction injuries who also had appropriate vascular screening. Approximately half the patients sustained one or more BCVIs. Three major strokes occurred among the patients with BCVIs while none were documented in patients without this concomitant injury. The authors advocate for screening for BCVI in all patients with craniocervical distraction injuries. Implications This retrospective study is confounded by selection and indication bias, thus limiting the capacity to determine the true incidence of BCVI after craniocervical trauma. Patients who did not receive appropriate vascular screening may have also had these injuries, but could not be included in this analysis. Furthermore, from a Bayesian perspective, the mere fact that the cohort under consideration had vascular studies performed indicates that they probably were at elevated risk of vascular trauma. Nonetheless, given the rarity of craniocervical dissociation and the severity of vascular sequelae if undetected, the authors’ recommendations for universal screening for BCVI are likely well founded. —The Editors Seven patients received no specific anticoagulation treatment for the BCVI (four patients had only Grade I injuries, two had intracranial hemorrhages, and one patient had severe pelvic bleeding). Seven patients were treated with antiplatelet agents (ASA) and one patient, with a diagnosis of pulmonary embolism, received warfarin. Of the 15 patients with BCVI, one was lost to follow-up, one died as an inpatient, one died 24 months postoperatively while quadriplegic, one remained ASIA (American Spinal Injury Association) A quadriplegic, one was triplegic and wheelchair bound, four were severely disabled/vegetative because of TBI, and six patients were independent with minimal and/or mild disabilities.
Discussion Although the incidence of BCVI in the general trauma population is in the range of 0.1 to slightly more than 1% [8,9], it
502
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
Table 1 Current screening criteria for patients with indications for blunt carotid and/or vertebral artery injury
Table 3 Associated injuries in those who underwent screening for BCVI
Symptoms or signs of BCVI 1. Massive epistaxis 2. Central or lateralizing neurological deficit unexplained by findings on head CT scan 3. Expanding cervical hematoma 4. Acute or subacute infarction on head CT scan 5. TIA or stroke after neck trauma 6. Horner’s syndrome 7. Cervical bruit in a patient younger than 50 years Injury mechanisms known as risk factors for BCVI 1. Midface fracture with cervical hyperextension/rotation/flexion injury 2. Complex mandible fracture and cervical hyperextension/rotation/ flexion injury 3. Closed head injury with DAI and cervical hyperextension/rotation/ flexion injury 4. Displaced midface fracture (Lefort II or III) 5. Skull base fracture involving foramen lacerum, sphenoid, mastoid, or petrous bones 6. Complex mandible fracture because of high-speed mechanism 7. Seatbelt abrasion, hanging bruise, contusion, or hematoma of the neck with cervical swelling 8. Cervical spine fracture with cervical hyperextension/rotation/flexion injury 9. Vertebral artery foramen transversarium fracture 10. Cervical vertebral subluxation 11. Cervical fracture C1 through C3
Associated injuries
BCVI present (15), n (%)
BCVI absent (14), n (%)
Mandible fractures Facial fractures Traumatic brain injury Additional spine fractures Skull base fractures
4 4 9 8 4
0 1 7 4 0
BCVI, blunt cerebrovascular injury; CT, computed tomography; TIA, transient ischemic attack; DAI, diffuse axonal injury.
can be a cause of severe disability and/or mortality among trauma patients [1]. During the past two decades, efforts have been put into diagnosing BCVIs more efficiently in a timely fashion. This has been related to improvements in the quality, noninvasiveness, and speed of CT scans and the demonstration that rapid diagnosis and initiation of therapy for BCVI can prevent or diminish brain ischemia and neurological deficits [1,10–13]. Although ideally the trauma surgeon would like to appropriately identify all patients, screening of the entire trauma population is not cost effective. Therefore, several studies attempting to identify subpopulations that would have a greater likelihood of having suffered a BCVI have been done [6,8,14]. Blunt cerebrovascular injury has been correlated with a high-speed trauma mechanism associated with hyperextension/flexion, especially in those patients who have suffered one of the following: (1) TBI with a Glascow Coma Scale !6, (2) Lefort facial fractures, (3) basilar skull fracture with carotid canal involvement, and (4) cervical
(27) (27) (64) (57) (27)
(0) (7) (50) (29) (0)
BCVI, blunt cerebrovascular injury.
spine fractures at C1–C3, foramen transversarium fractures, and/or subluxations of the cervical spine [6,8,14,15]. Craniocervical dissociation is a severe injury that commonly results in death because of hyperextension of the brain stem at or near the foramen magnum, with severe injury to the medulla oblongata and/or medullary-spinal cord junction [16–19]. A major distractive/hyperextension and high-speed mechanism has been implicated as the usual cause for CCD [17]. The resulting separation of the skull from the upper cervical spine can be seen in imaging studies as an increase in the distance between the basion and either the tip of the dens or the posterior axial line and increased distance between occipital condyles and the lateral masses of C1 on CT scans or widening of the C1–C2 joints [1,9,13]. Magnetic resonance imaging commonly shows disruption of the apical, alar ligaments, tectorial membrane, and Occ-C1 and/or C1–C2 joint capsules, along with a varied spectrum of cord abnormalities [3,20–23]. At Harborview Medical Center, Seattle, Washington Medical Center, a Level I trauma center covering a fourstate area, we have in the past two decades treated a relatively large number of patients with craniocervical distraction injuries. A concerted effort has been made to diagnose these patients as early as possible through recognition of this unique injury through rigorous education of emergency room radiologists and traumatologists as this has
Table 2 Biffl grading [6] for blunt cerebrovascular injuries Grade
Finding
1 2
Luminal irregularity or dissection with !25% stenosis Dissections with O25% luminal narrowing or a raised intimal flap Pseudoaneurysm Complete occlusion Transection of the artery, with free extravasation of contrast or significant arteriovenous fistula
3 4 5
Fig. 1. Breakdown of patients with pure ligamentous injuries in the setting of CCD versus those with associated craniocervical fractures in the setting of CCD and comparison of BCI versus BVI. CCD, craniocervical distraction injury; BCI, blunt carotid artery injury; BVI, blunt vertebral artery injury; BCVI, blunt cerebrovascular injury.
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
503
Fig. 2. (Left) Preoperative sagittal CT and (Right) parasagittal CT scan demonstrating wide separation of the occipital C1 joint and a basion-dens interval of almost 23 mm with an associated C1 anterior ring transverse fracture. CT, computed tomography.
been shown to be critical to optimizing neurological outcome in these patients [3]. As we continue to try to understand this subset of trauma patients, we have also attempted to understand associated injuries and findings and how to best work these up to optimize their care. Because CCD is usually caused by a high-speed mechanism, patients usually suffer other severe injuries that also may lead to severe disability and/or death. A high incidence
of severe TBIs, craniofacial fractures, skull base fractures, thoracic injuries, and other cervical spine injuries have been reported in patients with CCD, both in survivors and in autopsy series, which is consistent with our findings [3,17,18]. On the other hand, the occurrence of BCVI in survivors with CCD has been reported in only a few anecdotal cases [4,5], whereas tears of the carotid and basilar arteries
Fig. 3. Preoperative axial T2 image of the brain demonstrating bilateral cerebellar strokes.
Fig. 4. Preoperative sagittal T1 image of the brain demonstrating tonsillar herniation. Line shows the tonsillar hernation through the foramen magnum.
504
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
Fig. 5. Preoperative coronal computed tomography angiogram demonstrating bilateral vertebral artery injuries. The carotids and vertebral arteries were normal below the level of C1. The arrows demonstrate the locaton of the vertebral artery injuries in this patient.
have been described in postmortem series in patients with CCD related to a hyperextension mechanism [18]. We postulated that CCD could be associated with BCVI because of several reasons. First, both are usually caused by the same mechanism, that is, high speed with hyperextension/ distraction acting on the neck. Second, both CCD and BCVI are commonly associated with severe injuries of the craniofacial region, that is, head injuries, mandible fractures, facial fractures, and other cervical spine fractures. Third, the carotid arteries are intimately related to the occipital condyles and C1 vertebrae, whereas the vertebral arteries pass behind the C0–C1 joints and curve anteriorly to enter the foramen magnum, making them highly susceptible to the same distractive/hyperextension forces that cause the CCD. The importance of properly diagnosing BCVI in trauma patients is of utmost importance because it can lead to severe neurological deficits and even death related to a major stroke. Given the unstable nature of the CCD and the susceptibility for progressive neurological deficits if not operatively stabilized, it is also critical to know if a BCVI has occurred, even if a stroke has not, as this may play a role in the type of fixation used. If a patient has a vertebral artery injury on one side, the surgeon may elect to use a C2 translaminar screw on the contralateral side or a short pars screw as opposed to a pedicle screw of transarticular screw that may place the one functional vertebral artery at risk and thus contribute to a devastating stroke. Therefore, appropriate screening of patients at risk is necessary so proper initiation of treatment and identification of possible complications can be done. In our study, 15 of 29 screened patients had evidence of BCVI, with a total of 30 vessel injuries. Though a small
number of patients because of the relatively rare nature of this injury, this is the largest series of patients with CCD and screening CTA/angiogram to rule out BCVI reported to date. Our findings that more than 50% of our screened patients had a diagnosis of BCVI support our hypothesis. One could postulate that the cause for the BCVI in our patients was the presence of associated cervical spine fractures because cervical spine fractures have been linked to BCVI in more than 30% of cases and have been found as the main risk factor for BVI [2,8,24,25]. Although we found that half of our patients studied had associated cervical spine fractures, the incidence of BCVI was not affected by the presence of additional spine fractures. The finding that the patients in our series with additional cervical fractures had more BVI, as opposed to a higher occurrence of BCI in those without additional cervical fractures, is consistent with other series on BCVI [8]. Transcranial Doppler with emboli monitoring is usually performed during 3 consecutive days in our patients with diagnosed BCVI so as to evaluate the number of emboli/hour in the intracranial vasculature and also guide therapy effectiveness. Fourteen patients had TCD studies done, with only two examinations showing more than 20 emboli/h, one of them being on the patient who suffered an internal capsule/ basal ganglia stroke. In the two patients with cerebellar strokes, TCD examinations were negative for emboli. The literature lacks evidence of TCD as a guide for the management of BCVI, and more studies are clearly needed to ascertain its value in such a setting. Although BCVI can be a significant cause of morbidity/ mortality, our study was neither performed nor powered to evaluate treatment/outcomes. Therefore, no conclusions can be drawn either related to the impact of BCVI on the outcomes or its treatment effectiveness. We could, on the other hand, observe that most patients also sustained other severe injuries, such as spinal cord and/or severe head injury, with resultant disability that ultimately contributed to an unfavorable outcome. The incidence of stroke in patients with BCVI has been reported to be as high as 50% [26,27]. In our study, we found three patients (20%), among 15 with BCVI, who suffered a stroke. Because our numbers are small, we did not attempt to correlate the incidence of stroke in our series to other series. All strokes in this series were related to Grade II to IV injuries. Although mean age, number of head injuries, skull base fractures, and additional spine fractures were similar between the two groups, there were more facial fractures in those with BCVI (four mandibles, four facial fractures in 15 patients) when compared with patients without BCVI (no mandible, one facial fracture). This suggests that higher energy/forces were transferred to the craniofacial region and ultimately to the cerebrovascular structures at the neck in the group with BCVI. It is important to mention that we included in the data set any midface fracture, mandible fracture, skull base fracture, and/or TBI with intracranial hemorrhage as an
M.D. Vilela et al. / The Spine Journal 15 (2015) 499–505
indicator of injury severity to the craniofacial region. We are aware that including any facial, mandible, skull base, and head injury with intracranial hemorrhage encompasses more than the suggested indications proposed by some investigators for the screening of BCVI [6]. On the other hand, others have reported that using coexisting diagnosis does not yield a reliable prediction algorithm for screening trauma patients for BCVI, and the integration of clinical suspicion based on the mechanism and associated injuries may be essential to appropriately screen patients and minimize the number of missing injuries [9], a statement that corroborates the methods and screening criteria used in this study. Certainly, there are some major weaknesses to this study. Our numbers are small with a total 29 patients with CCD and CTA and/or angiography. Placed in perspective though, this is an uncommon injury with the largest series of patients with CCD reported to date at 17 patients [3] although the largest abstract to date reports on 48 patients [28]. Second, this is a retrospective study with the shortcomings of such a study. Third, although we had 46 patients with CCD, only 29 of them had CTAs and/or angiography. Our date dates back to 1996, and thus not all patients underwent imaging as this predated our current protocol. Many of our patients, in the more modern era, did not receive emergent CTA as they were too medically unstable and were emergently taken to the intensive care unit or to the operating room for management of associated injuries. In an ideal situation, each of these patients would have undergone CTA. We do recognize the deficiencies within this study but believe that this study also has important findings and is a stepping-stone toward further studies in the management of patients with CCD. Conclusions We found that more than 50% of our patients with CCD who were properly screened had BCVI, and 20% of those with BCVI had a major stroke. We suggest screening all patients with CCD with a neck CTA and/or catheter angiogram to assess for BCVI. References [1] Eastman AL, Muraliraj V, Sperry JL, Minei JP. CTA-based screening reduces time to diagnosis and stroke rate in blunt cervical vascular injury. J Trauma 2009;67:551–6. [2] Cothren CC, Moore EE, Biffl WL, Ciesla DJ, Ray CE Jr, Johnson JL, et al. Cervical spine fracture patterns predictive of blunt vertebral artery injury. J Trauma 2003;55:811–3. [3] Bellabarba C, Mirza SK, West GA, Mann FA, Dailey AT, Newell DW, et al. Diagnosis and treatment of craniocervical dislocation in a series of 17 consecutive survivors during an 8-year period. J Neurosurg Spine 2006;4:429–40. [4] Brinkman W, Cohen W, Manning T. Posterior fossa subarachnoid hemorrhage due to an atlantooccipital dislocation. AJR Am J Roentgenol 2003;180:1476. [5] Lee C, Woodring JH, Walsh JW. Carotid and vertebral artery injury in survivors of atlanto-occipital dislocation: case reports and literature review. J Trauma 1991;31:401–7.
505
[6] Biffl WL, Cothren CC, Moore EE, Kozar R, Cocanour C, Davis JW, et al. Western Trauma Association critical decisions in trauma: screening for and treatment of blunt cerebrovascular injuries. J Trauma 2009;67:1150–3. [7] Biffl WL, Moore EE, Offner PJ, Brega KE, Franciose RJ, Burch JM. Blunt carotid arterial injuries: implications of a new grading scale. J Trauma 1999;47:845–53. [8] Berne JD, Cook A, Rowe SA, Norwood SH. A multivariate logistic regression analysis of risk factors for blunt cerebrovascular injury. J Vasc Surg 2010;51:57–64. [9] Cook A, Osler T, Gaudet M, Berne J, Norwood S. Blunt cerebrovascular injury is poorly predicted by modeling with other injuries: analysis of NTDB data. J Trauma 2011;71:114–9. [10] Biffl WL, Egglin T, Benedetto B, Gibbs F, Cioffi WG. Sixteen-slice computed tomographic angiography is a reliable noninvasive screening test for clinically significant blunt cerebrovascular injuries. J Trauma 2006;60:745–51. [11] Sliker CW. Blunt cerebrovascular injuries: imaging with multidetector CT angiography. Radiographics 2008;28:1689–708. [12] Stengel D, Rademacher G, Hanson B, Ekkernkamp A, Mutze S. Screening for blunt cerebrovascular injuries: the essential role of computed tomography angiography. Semin Ultrasound CT MR 2007;28:101–8. [13] Tikkakoski T. Screening for blunt cerebrovascular injury in trauma patients. Acta Radiol 2007;48:707. [14] Ringer AJ, Matern E, Parikh S, Levine NB. Screening for blunt cerebrovascular injury: selection criteria for use of angiography. J Neurosurg 2010;112:1146–9. [15] Cothren CC, Moore EE, Ray CE Jr, Johnson JL, Moore JB, Burch JM. Cervical spine fracture patterns mandating screening to rule out blunt cerebrovascular injury. Surgery 2007;141:76–82. [16] Gross A. Traumatic basal subarachnoid hemorrhages: autopsy material analysis. Forensic Sci Int 1990;45:53–61. [17] Imaizumi T, Sohma T, Hotta H, Teto I, Imaizumi H, Kaneko M. Associated injuries and mechanism of atlanto-occipital dislocation caused by trauma. Neurol Med Chir (Tokyo) 1995;35:385–91. [18] Ohshima T, Kondo T. Forensic pathological observations on fatal injuries to the brain stem and/or upper cervical spinal cord in traffic accidents. J Clin Forensic Med 1998;5:129–34. [19] Tepper SL, Fligner CL, Reay DT. Atlanto-occipital disarticulation. Accident characteristics. Am J Forensic Med Pathol 1990;11:193–7. [20] Chaput CD, Walgama J, Torres E, Dominguez D, Hanson J, Song J, et al. Defining and detecting missed ligamentous injuries of the occipitocervical complex. Spine 2011;36:709–14. [21] Deliganis AV, Baxter AB, Hanson JA, Fisher DJ, Cohen WA, Wilson AJ, et al. Radiologic spectrum of craniocervical distraction injuries. Radiographics 2000;20 Spec No.:S237–50. [22] Pang D, Nemzek WR, Zovickian J. Atlanto-occipital dislocation— part 2: the clinical use of (occipital) condyle-C1 interval, comparison with other diagnostic methods, and the manifestation, management, and outcome of atlanto-occipital dislocation in children. Neurosurgery 2007;61:995–1015. [23] Pang D, Nemzek WR, Zovickian J. Atlanto-occipital dislocation: part 1—normal occipital condyle-C1 interval in 89 children. Neurosurgery 2007;61:514–21. [24] Biffl WL, Moore EE, Elliott JP, Ray C, Offner PJ, Franciose RJ, et al. The devastating potential of blunt vertebral arterial injuries. Ann Surg 2000;231:672–81. [25] Miller PR, Fabian TC, Croce MA, Cagiannos C, Williams JS, Vang M, et al. Prospective screening for blunt cerebrovascular injuries: analysis of diagnostic modalities and outcomes. Ann Surg 2002;236:386–93. [26] Burlew CC, Biffl WL. Blunt cerebrovascular trauma. Curr Opin Crit Care 2010;16:587–95. [27] DiCocco JM, Fabian TC, Emmett KP, Magnotti LJ, Zarzaur BL, Bate BG, et al. Optimal outcomes for patients with blunt cerebrovascular injury (BCVI): tailoring treatment to the lesion. JAm Coll Surg 2011;212:549–57. [28] Bellabarba C, Bransford RJ, Chapman JR. Timing to diagnosis and neurological outcomes in 48 consecutive craniocervical dissociation patients. Spine J 2011;11:S57.