The Incidence, Characteristics and Outcomes of Vertebral Artery Injury Associated with Cervical Spine Trauma: A Systematic Review

Abstract

Study design

Systematic Review

Objectives

Vertebral Artery Injury (VAI) is a potentially serious complication of cervical spine fractures. As many patients can be asymptomatic at the time of injury, the identification and diagnosis of VAI can often prove difficult. Due to the high rates of morbidity and mortality associated with VAI, high clinical suspicion is paramount. The purpose of this review is to elucidate incidence, diagnosis, treatment and outcomes of VAI associated with cervical spine injuries.

Methods

A systematic search of electronic databases was performed using ‘PUBMED’, ‘EMBASE’,‘Medline (OVID)’, and ‘Web of Science, for articles pertaining to traumatic cervical fractures with associated VAI.

Results

24 studies were included in this systematic review. Data was included from 48 744 patients. In regards to the demographics of the focus groups that highlighted information on VAI, the mean average age was 46.6 (32.1-62.6). 75.1% (169/225) were male and 24.9% (56/225) were female. Overall incidence of VAI was 596/11 479 (5.19%). 190/420 (45.2%) of patients with VAI had fractures involving the transverse foramina. The right vertebral artery was the most commonly injured 114/234 (48.7%). V3 was the most common section injured (16/36 (44.4%)). Grade I was the most common (103/218 (47.2%)) injury noted. Collective acute hospital mortality rate was 32/226 (14.2%), ranging from 0-26.2% across studies.

Conclusion

VAI secondary to cervical spine trauma has a notable incidence and high associated mortality rates. The current available literature is limited by a low quality of evidence. In order to optimise diagnostic protocols and treatment strategies, in addition to reducing mortality rates associated with VAI, robust quantitative and qualitative studies are needed.

Introduction

Vertebral Artery Injury (VAI) is a potentially serious complication of cervical spine fractures. The vertebral arteries (VA) provide vital blood supply to the occipital lobe, brainstem and cerebellum. ¹ The left VA is dominant in 50% of the population, the right in 25%, and the arteries demonstrate equal calibre in 25% of the population, ² with respective diameter of approximately 3-5 mm. Each respective VA can be divided into 4 segments (V1-4). V1 (pre-foraminal) commences at the origin to the transverse foramen of cervical vertebra C6. V2 (foraminal) lies through the transverse foramen of C2-C6. V3 (extradural) starts from C2, where the artery loops and turns lateral to ascend into the transverse foramen and continues through C1 to pierce the dura. V4 (intracranial) runs from the dura at the lateral edge of the posterior atlanto-occipital membrane to their confluence on the medulla to form the basilar artery.

The incidence of VAI in patients with blunt cervical spine trauma ranges from .53% to 39% in the literature.^3-5 This wide variation in incidence is most likely due to differences in sample size and the diagnostic imaging modality (digital subtraction angiography (DSA), magnetic resonance angiography (MRA) or Computed Tomography Angiograms (CTA)) employed, as well as patient selection bias. Recently, there has been a higher incidence of VAI reported, most likely due to advances in imaging technology. ³ Nevertheless, fractures involving the transverse process, which can extend into the transverse foramina, and fractures with associated subluxation are particularly renowned for increased risk of VAI.^6,7

Broadly, VAI can be categorized into traumatic and spontaneous. Traumatic VAI can be further subdivided into blunt and penetrating injuries and can include dissection, intimal tears, stenosis or traumatic aneurysm. ⁸ VAI can be graded using the Denver scale, ⁶ illustrated in Figure 1. Although the majority of patients are initially asymptomatic, ⁷ believed to be due to delayed thrombus formation and development of the vascular injury, VAI can have devastating consequences including permanent neurological deficits and associated mortality. Therefore, high clinical suspicion is important regarding the possibility of VAI in patients presenting with cervical spine trauma. As such, the purpose of this review is to elucidate incidence, diagnosis, treatment and outcomes of VAI associated with cervical spine injuries denoted in academic literature. By doing so, efficacious, multi-disciplinary, timely treatment protocols may be improved and implemented for this vulnerable cohort.

Figure 1.

Denver Grading Scale for blunt cerebrovascular injuries. Grabowski G, Robertson RN, Barton BM, Cairns MA, Webb SW. Blunt Cerebrovascular Injury in Cervical Spine Fractures: Are More-Liberal Screening Criteria Warranted? Global Spine J. 2016 Nov;6 (7):679-685.

Methods

Search Strategy and Data Extraction

A systematic review was conducted in accordance to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A formal systematic search was performed of the PubMed, Embase, Medline (OVID) and Web of Science databases to identify relevant titles up to and including February 26^th 2022, for studies relating to traumatic cervical fractures with associated vertebral artery injury.

The search was performed by 2 independent reviewers (HCT and NOS), using a predetermined search strategy that was designed by the senior authors. Details in relation to the search strategy can be found in Supplementary Appendix 1. Retrieved studies were reviewed to ensure studies met the predefined inclusion and exclusion criteria (see below). Discordances in opinion were resolved through consultation with a third author (JM). Manual cross-referencing of reference lists from studies was undertaken to identify further potential articles for inclusion.

Data extraction was also performed by 2 independent reviewers (HCT and NOS), with study details, basic patient clinicopathological characteristics, management strategies and outcomes recorded. Furthermore, information extracted was based on the PICOTS framework (Population, Intervention, Comparator, Outcomes, Timing and Setting). GraphPad Prism (Version 8.3.0) was utilised for analysis and figures.

Eligibility Criteria

The inclusion criteria were as follows: 1) Published studies demonstrating vertebral artery injury in cervical spinal trauma. 2) Published in the English language. 3) Published after 2000.

The exclusion criteria were as follows: 1) Abstract only. 2) Studies failing to discuss or denote VAI. 3) Paediatric cohorts aged less than 18 years. 4) Case reports

Risk of Bias

Assessment of potential bias was assessed using Newcastle- Ottawa scale (NOS) risk of bias tool and the results tabulated. This assessment tool grades each study as being ‘satisfactory’ or ‘unsatisfactory’ across various categories. We assigned stars to evaluate study quality: 7 stars - “very good”, 5-6 stars “good”, 3-4 stars “satisfactory” and 0-2 stars “unsatisfactory”. The critical appraisal was completed by 2 reviewers independently (HCT and NOS), where once again a third reviewer (JM) was asked to arbitrate in cases of discrepancies in opinion.

Aims and Objectives

The overall aims and objectives of this study were as follows:

• 1. To elucidate the incidence of associated VAI in cervical spine trauma patients

• 2. To outline the characteristics of VAI and presentation in cervical spine trauma patients

• 3. To report on management strategies and outcomes of patients with VAI

Results

Study Selection/Included Studies

The systematic search strategy identified a total of 977 studies, of which 448 duplicate studies were manually removed. The remaining 529 studies were screened for relevance, before 54 full texts were reviewed. In total, 24 studies fulfilled our inclusion criteria and were included in this systematic review^4,5,9-30 (Figure 2). Due to heterogeneity in study method and results presented, a collative summary of findings was deemed appropriate.

Figure 2.

PRISMA flowchart outlining the systematic search process.

Baseline Characteristics

The data collected was highly heterogenous. The majority of studies had contrasting and inconsistent study population groups. In total, there was data included from 48 744 patients. All 24 studies were population-based studies concerned with cervical fractures, 17/24 of which had sub-cohort analysis for patients with confirmed VAI and 2/24 for cervical spine fracture patients suspicious for VAI

In regards to the demographics of the focus groups that highlighted information on VAI, the mean average age 46.6 (range of reported means: 32.1-62.6), and 66.3% (available gender reported data:169/225) were male. The study design was a retrospective review in 18/24 studies, with the remaining 6/24 being prospective reviews. Of the 24 studies included in this analysis, 58.3% were conducted in United States surgical research institutions (14/24). Other countries of publication included; Japan, UK, Germany, Spain, South Korea, Saudi Arabia and India. Publication dates ranged from 2000-2022 (Table 1). A summary of study characteristics is presented in Table 1.

Table 1.

Baseline study characteristics.

Author	Year	Country	Study Period	Study Design	Study Group	Sample Size of Study Group	Focus Group	Sample Size of Main Focus Group	Focus group Age: Mean (Range)	Focus group Sex (Male/Female)	MOI of VAI Group	MOI in Other Group
Veras	2000	Spain	1992-1997	Retrospective review	Cervical spine fractures	N/A	VAI	6	32.1 (18-49)	5/1	MVA: 5/6 Dive into swimming pool: 1/6	N/A
Biffl	2000	USA	1996-1999	Prospective	Blunt trauma	7205	VAI	38	38.9 (16-77)	Na	MVA in 22 (58%) fall in 5 (13%) pedestrian struck in 4 (11%); other mechanisms were uncommon	N/A
Miller	2002	USA	1995-1999	Prospective	Patients who met BCVI screening criteria	241	Those with blunt cervical trauma undergoing arteriography	216	37.6	115/101	N/A	Those with blunt cervical trauma undergoing arteriography MVA (78%) fall (8%), pedestrian struck (4%) assault (4%). Remaining patients had a combination of less-common mechanisms
Kral	2002	Germany	1998-2001	Prospective	Those with blunt cervical trauma suspicious for VAI	119	Those with potential VAI	31	44.2 (17-89)	19/12	N/A	N/A
Cothren	2003	USA	1996-2002	Prospective	Blunt trauma	12 552	VAI	92	38	58/34	MVA 58 Auto-pedestrian 5 MCC 4 Fall 9 Hanging 2 Ski/snowboard accident 5 Other 9	N/A
McKinney	2007	USA	2003-2004	Retrospective review	Blunt trauma	2073	Angiograms performed in patients who had fractures with foraminal involvement	71	N/A	N/A	N/A	Angiograms performed in patients who had fractures with foraminal involvement MVA: 43/71 Falls: 13/71 Assaults, crush injuries, and falling objects; 15/71
Kopelman	2011	USA	2003-2009	Retrospective review	Cervical spine injury	496	Cervical spine fracture and underwent evaluation of the cerebrovasculature	260	N/A	N/A	N/A	Cervical spine fracture and underwent evaluation of the cerebrovasculature divided into 2 groups based on the presence/absence of risk factors MVA: (Group A, 74%; group B,66%) Falls (group A, 11%; group B, 13%), Pedestrian struck (group A,7%; group B, 8%) other (group A, 8%; group B, 13%)
Mueller	2011	Germany	1996-2010	Prospective observational study	Cervical spine trauma	599	Cervical spine injury suspicious for VAI due to facet luxation and/or fractures extending into the transverse foramen	69	43	44/25	N/A	Cervical spine injury suspicious for VAI due to facet luxation and/or fractures extending into the transverse foramen. 36/69 (52.17%) high velocity accidents (motor vehicle and bicycle accidents) 33/69 (47.83%) low velocity (falls) accidents
Jang	2011	South Korea	2005-2007	Prospective	Cervical spine fractures and/or dislocations	99	VAI	7	43 (33-55)	4/3	MVA: 3/7 Falls: 3/7 Unknown aetiology: 1/7 patient
Emmett	2011	USA	2007-2009	Retrospective review	Blunt trauma	20 049	Trauma admissions screened for BCVI who underwent Screening Cerebral Angiography	748	39	515/233	N/A	BCVI (not specifically related to VAI group) MVA: 97 (83%) Fall: 4 (3%) Assault: 4 (3%) Motorcycle collision: 3 (3%) Pedestrian struck: 7 (6%) Hanging: 1 (1%) Crush: 1 (1%)
Chung	2012	South Korea	2006-2010	Retrospective review	Traumatically destabilised subaxial cervical spine	91	VAI	18	50 (27-77)	16/2	N/A	N/A
Lebl	2013	USA	2007-2010	Retrospective review	Cervical injuries who underwent screening for VAI by MDCTA	253	VAI	42	56.43	23/19	MVA: 22/42 Fall: 19/42 Jumped 3 stories: 1/42	N/A
Payabvash	2014	USA	2006-2011	Retrospective review	All blunt trauma patients with C1 and/or C2 fractures	210	VAI	30	N/A	N/A	N/A	All blunt trauma patients with C1 and/or C2 fractures Fall 98 (46.7%) Motor vehicle accident 90 (42.9%) Motorcycle accident 9 (4.3%) Vehicle vs pedestrian accident 6 (2.9%) Assault 1 (.5%) Others 6 (2.9%)
Durand	2015	USA	2004-2014	Retrospective review	Axis fracture on noncontract cervical spine CT scan	2831	VAI	25	62.63	10/12	N/A	N/A
Lockwood	2016	USA	2002-2012	Retrospective review	Cervical spine fractures that underwent CTA for VAI screening	732	VAI	51	N/A	N/A	N/A	N/A
Sud	2016	USA	2002-2011	Retrospective cohort study	Traumatic cervical spine fractures meeting studies inclusion criteria	119	VAI	30	54.6	22/8	Fall 15/30 MVA 14/30 Other 1/30	Traumatic cervical spine fractures meeting studies inclusion criteria MVA: 73/116 (62.9%) Fall 43/116 (37.1%)
Nagata	2017	Japan	2008-2014	Retrospective review	Cervical spine dislocation	26	VAI	18	53.7	16/2	N/A	N/A
Bonney	2017	USA	2009-2014	Retrospective review	Isolated C-spine transverse process fractures	45	VAI	5	48.8 (21-88)	N/A	MVA: 4/5 Fall: 1/5	Isolated C-spine TP fracture patients MVA 32 (71.1%) MCC 6 (13.3%) Fall 3 (6.7%) Pedestrian 1 (2.2%) All-terrain vehicle 1 (2.2%) Other blunt 2 (4.4%) All other C-spine fracture patients MVA 204 (71.6%) MCC 25 (8.8%) Fall 36 (12.6%) Pedestrian 6 (2.1%) 1.00 ATV 3 (1.1%) Other blunt 11 (3.9%)
Chrowdhury	2020	Saudi Arabia	2018-2019	Retrospective cohort study	Blunt trauma patients undergoing screening CT neck angiography	120	VAI	10	N/A	N/A	N/A	Blunt trauma patients undergoing screening CT neck angiography with C-Spine fracture MVA: 73 (86.9%) Falls: 10 (11.9%) Assault: 1 (1.2%)
Sheppard	2020	UK	2018-2019	Retrospective cohort study	Cervical spine fracture	68	VAI	5	50	4/1	MVA: 4/5 Fall from horse: 1/5	Cervical spine fracture MVA 23 (33.8%) Assault 2 (2.9%) Go-karting accident 1 (1.5%) Rugby tackle 1 (1.5%)
Cloney	2021	USA	1999-2016	Retrospective review	Trauma atlas fractures	97	VAI	10	N/A	N/A	N/A	Trauma atlas fractures Fall 51.6% 31 MVA 19.6% 18 Diving 6.2% 8 Intoxication 6.2% 8 Violence 5.2% 5
Wynn	2021	USA	2014-2018	Retrospective cohort study	Blunt cervical spine injury and/or fracture	475	VAI	35	N/A	N/A	N/A	Blunt cervical spine injury and/or fracture MVA: 225/475 (47.4%)
Nakamura	2021	Japan	2015-2018	Retrospective review	Traumatic cervical spine injury	89	VAI	15	62.6	11/4	Fall: 8/15 MVA: 7/15	N/A
Gupta	2022	India	2014-2019	Retrospective review	Cervical vertebrae fracture/subluxation	149	Cervical vertebrae fracture/subluxation	149	N/A	127/22	N/A	Cervical vertebrae fracture/subluxation MVA: 73/149 Fall: 62/149 Other: 14/149

MOI: Mechanism of injury. VAI: Vertebral artery injury. N/A: not applicable. MVA: Motor vehicle accident. BCVI: Blunt cerebrovascular injury. MCC: Motorcycle crash. TP: Transverse process. ATV- All-terrain vehicle.

Risk of Bias

The risk of bias was assessed in included studies. One study was deemed ‘very good’, thirteen studies were rated as ‘good’. Seven studies were noted to be “satisfactory”, and 3 studies were found to be “unsatisfactory”. The risk of bias was assessed using Newcastle-Ottawa Scale (NOS) ³² and is presented in Supplementary Appendix 2.

Vertebral Artery Injury

Incidence of VAI

22/24 studies reported on the incidence of VAI. Study cohorts included: blunt trauma patients screening for blunt cerebrovascular injury (BCVI) with diagnostic angiography (11/24), cervical spine fractures and/or dislocation (7/24), blunt trauma patients (1/24), isolated axis fractures (1/24), isolated atlas fractures (1/24), isolated C-spine transverse process fractures (1/24), patients with cervical spine injury and VAI (1/24), angiograms performed in patients who had fractures with foraminal involvement (1/24). Figure 3 displays the reported VAI incident rates across the studies.

Figure 3.

Incidence of VAI. VAI: Vertebral artery injury.

It is challenging to calculate a true incidence of VAI because of the heterogenous study populations. Some studies gave incidence of VAI in all blunt trauma patients compared to others which only reported on VAI in those who had undergone screening for high risk of VAI. Amongst all study populations included in this study, the overall incidence is 596/11 479 (5.19%). To analyse further, the incidence of VAI associated with any patient presenting with cervical spine fractures and/or dislocation was 14/807 (1.7%), trauma atlas (C1) fractures 8/97 (8.2%), isolated axis fractures who had undergone subsequent CTA 25/67 (37.3%) and blunt trauma patients 38/7205 (.53%). 11/24 studies focused on blunt trauma patients screening for BCVI undergoing diagnostic angiography (CTA, DSA, MRA, angiography), with a collective incidence of 410/3490 (11.7%). One study focused on isolated C-spine transverse process fractures, reported an incidence of 5/45 (11.1%).

Mechanism of Injury (MOI)

Of the 17/24 studies which outlined VAI as their main focus group/cohort of interest, 9/17 described the MOI. Motor vehicle accidents (MVA) was the most common injury in 7/9 studies, with falls being the most common in 2/9. In total, of the VAI studies which analysed MOI 139/240 (57.9%) were MVA and 61/240 (25.4%) falls. Other injuries were uncommon but consisted of; pedestrians hit by vehicles, jumping from buildings, snow-sport accidents, horse riding accidents, hangings, diving into swimming pools and others. The remaining 7/24 studies denoting incidental cases of VAI did not report on MOI.

Location of VA Affected

Overall, 11/24 studies reported on whether the vertebral artery (VA) affected was right, left or bilateral. In total, 234 VAs were reported on across these studies. The right VA was the most commonly injured 114/234 (48.7%) followed by the left 97/234 (41.5%), with bilateral injury being the least common injury 23/234 (9.8%). Two studies reported on whether the injury was unilateral or bilateral. Both studies report unilateral injury to be more common than bilateral injury (53/71 (74.6%) vs 11/18 (61.1%)). Three studies report the particular section of the VA injured. V3 was the most common section injured (16/36 (44.4%)), followed by V1/2/3 (6/36 (16.7%)) and V1/2 (2/36 (5.6%)), (Figure 4).

Figure 4.

Location of VAI. VA: vertebral artery.

Type of Injury to the VA

Overall, 18/24 studies reported on the type of injury which occurred to the VA. In total, 11/18 of these used the Denver Scale ³³ (Biffl et al) to classify these injuries. Furthermore, 9/11 of those studies using the Denver scale reported comprehensive characteristics of the entire cohort, while 2/11 studies only reported which grade was the most common and did not give specific figures. From the 9/11 studies which did give figures for each grade, there were 218 patients analysed (Figure 5). Grade I was the most common (103/218 (47.2%)) injury noted. Others included; Grade II/(53/218 (24.3%)), Grade III (8/218 (3.7%)), Grade IV (53/218 (24.3%)), and Grade V (1/218 (.5%)). In total, 7/18 studies (82 patients with VAI) did not use the Denver scale to demonstrate the injuries. Overall characteristics of injury included occlusion (40/82 (48.8%)), dissection (34/82 (41.5%)), stenosis (4/82 (4.9%)), occlusive-dissection (2/82 (2.4%)), and compression (2/82 (2.4%)). One study was only concerned with VA dissections secondary to injury, Gupta et al ¹⁶ noted an incidence of VA dissection of 14.1% (21/149).

Figure 5.

Grades of VAI. VA: vertebral artery.

Presentation/Symptoms/Neurological Status

Overall, 7/24 studies reported on symptoms at presentation or during the admission. Given the paucity and heterogenous nature of this outcome we were only able to demonstrate a descriptive narrative. The most common presentation across multiple studies were headaches, clinical signs of vertebrobasilar ischemia, and vertigo. In total, 2/7 studies noted headaches as a presenting symptom, with a range of 5.2-60%. 3/7 studies noted clinical signs of vertebrobasilar ischemia at presentation, with a range of 0-19%. 2/7 studies reported asymptomatic patients at time of diagnosis, with a range of 60-66.7%. A full descriptive narrative of presentation and symptoms associated with VAI across studies is outlined in Table 2.

Table 2.

Vertebral Artery Injury Overview.

Author	Incidence of VAI	What is the Incidence of a Percentage of?	Symptoms	Diagnostic Modality	Management of Those with VAI	Outcomes in Those with VAI	Mortality with VAI (Average Follow up)
Veras	N/A	N/A	1 Blurred vision 1 Locked-in syndrome 3 No vertebral Ischemic Symptoms	MRI/MRA	- Embolization: N/A - Observation: 3/6 - Antiplatelet: N/A - Heparin: 3/6 - Surgery: 5/6 (reduction and anterior spinal fusion)	After a follow-up period of 15 months (6-24 months): - 5/6 patients: Free of ischemic symptoms, and neurologic damage had not progressed in the patient with vertebrobasilar territory infarctions. - 1 patient: Follow-up MRI demonstrated the persistence of vertebral artery occlusion	N/A
Biffl	38/7205 (.53%)	Blunt trauma patients	N/A	Four-vessel cerebral arteriography (DSA)	- Embolization: 1/38 - Observation: 6/38 - Antiplatelet: 6/38 - Heparin: 25/38 - Surgery: N/A	- 12/38 (32%) of the patients were neurologically normal at discharge. - 12/38 had mild deficits: Four were attributable to spinal cord injuries, 2 to brain injuries, 1 to blunt carotid injury, 1 to brachial plexus injury, and 4 to the BVI. - Cervical spine injuries were present in 71% of patients, but there was no predilection for cervical vertebral level or fracture pattern. - Incidence of posterior circulation stroke: 24% - BVI-attributable death rate was 8%. Stroke incidence and neurologic outcome were independent of BVI injury grade. - In patients treated with systemic heparin, fewer overall had a poor neurologic outcome, and fewer had a poor outcome after stroke. - Trends associated with heparin therapy included fewer injuries progressing to a higher injury grade, fewer patients in whom stroke developed, and fewer patients deteriorating neurologically from diagnosis to discharge	7/38 (18.4%) (During admission)
Miller	43/216 (19.9%)	Those with blunt cervical trauma undergoing arteriography	All of the VAIs were found before the development of ischemic symptoms	DSA and/or CTA and/or MRA	- Embolization: N/A - Observation: 3/43 - Antiplatelet: 32/43 (aspirin or aspirin/clopidogrel) - Heparin: 8/43 - Surgery: N/A	- No strokes were seen in the 43 patients with VAI - All VAIs were diagnosed before the development of ischemia	4/43 (9%) (N/A)
Kral	5/31 (15.1%)	Patients with potential VAI	2/5 headache 3/5 asymptomatic	DSA	- Embolization: N/A - Observation: N/A - Antiplatelet: 5/5 (aspirin for 6 months) - Heparin: 2/5 (for 2 weeks) - Surgery: 4/5	- All patients (5/5) had no neurological symptoms at 6-month follow-up. - DSA at 6 months showed permanent stenosis in 1/5	N/A
Cothren	92/605 (15.2%)	Blunt trauma patients screening for BCVI undergoing diagnostic angiography	N/A	‘Diagnostic angiography’	N/A	N/A	N/A
McKinney	N/A	N/A	N/A	Catheter, CT or MR angiography	N/A	N/A	N/A
Kopelman	14/260 (5.4%)	Cervical spine fractures	N/A	CTA and/or DSA	N/A	N/A	N/A
Mueller	19/69 (27.5%)	Cervical spine trauma suspicion for VAI and therefore underwent DSA	4 (21%) had clinical signs of vertebrobasilar ischemia 2 were initially comatose, 1 had vertigo and headache, and 1 patient had vertigo	DSA	- Embolization: N/A - Observation: N/A - Antiplatelet: 19/19 (aspirin for 6 months) - Heparin: 19/19 - Surgery: - Anterior cervical plate-screw fusion 9/19 - dorsal rod-screw fusion 1/19 - combined ventral–dorsal fusion 1/19	- Despite anticoagulation therapy, 5.8% became clinically symptomatic and 2.9% died due to cerebrovascular ischemia. - Of the 19 patients, 15 (79%) with VA injury were neurologically intact at the 3- and 6-month follow-up. 2 patients died and two patients reported persistent vertigo. - The final assessment was conducted at 12-168 months (mean follow up period 83.3 months, n = 17). None of the patients reported persistent headache, vertigo or other focal neurological deficits	2/19 (10.5%) (during admission)
Jang	7/99 (7.1%)	Cervical spine fractures and/or dislocation	No patient had clinical signs of vertebrobasilar ischemia	CTA	- Embolization: N/A - Observation: N/A - Antiplatelet: 5/7 (aspirin for 3 months) - Heparin: 0/7 - Surgery: 5/7 (2/7 conservative treatment with a soft collar brace)	- During the follow-up period, a mean duration of 23 months, vertebro-basilar insufficiency was not detected in any of our patients. - At follow-up 4/7 patients had an intact neurological exam. 3/7 had quadriparesis	N/A
Emmett	65/748 (8.7%)	Trauma admissions screened for BCVI who underwent Screening Cerebral Angiography	One patient developed neurologic symptoms subsequent to initial evaluation	CTA	N/A	N/A	N/A
Chung	18/91 (19.8%)	Traumatically destabilised subaxial cervical spine	N/A	CTA or MRA or DSA	N/A	N/A	N/A
Lebl	42/253 (17%)	Cervical injuries who underwent screening for VAI by MDCTA	N/A	MDCTA	- Embolization: 0/42 - Observation: 19/42 - Antiplatelet: 15/42 home medications ((aspirin, clopidogrel, or coumadin) - Heparin: 8/42 (19%) - Surgery: 0/42	- VAI were associated with a lower Glasgow coma scale (P < .001), a higher injury severity score (P < .01), and a higher mortality (P < .001). - Subgroup analyses of neurological events secondary to VAI occurred in 14% (6 of 42) and the stroke-related mortality rate was 4.8% (2 of 42)	11/42 (26.2%) (N/A)
Payabvash	30/126 (24%)	Patients with cervical spine trauma evaluated by CTA and/or DSA	N/A	CTA +/− DSA Or DSA only	- Embolization: N/A - Observation: 13/30 - Antiplatelet: 1 (3%) Clopidogrel, 8 (28%) Aspirin and 2 (7%) other - Heparin: 4 (13%) Enoxaparin - Warfarin: 2 (7%) Warfarin, - Surgery: N/A	- Posterior circulation stroke 3 (10%) - Cervical pain/radiculopathy 1 (3%) - Paraplegia/quadriplegia 2 (7%) - Traumatic brain injury 1 (3%)	Related to trauma 2/30 (6.7%) Related to co-morbidity 1/30 (3%) (during admission)
Durand	25/67 (37.3%)	Isolated axis fractures and had undergone subsequent CTA	N/A	CTA	N/A	N/A	N/A
Lockwood	51/732 (.7%)	Cervical spine fractures that underwent CTA for VAI screening	N/A	CTA or MRA or DSA	- Embolization: 1/51 - Observation: 11/51 - Antiplatelet: 35/51 (3 months) - Heparin: 5/51 (all 5 then commenced aspirin for 3 months) - Surgery: N/A	- Posterior circulation stroke 4/51 (7.8%). - 2 were discharged neurologically intact - 2 died secondary to polytrauma injuries	N/A
Sud	30/119 (25.2%)	Traumatic cervical spine fractures meeting studies inclusion criteria	N/A	CTA or MRA	N/A	N/A	N/A
Nagata	18/26 (69%)	Cervical spine dislocation who underwent CTA	N/A	CTA and/or DSA	- Embolization: 1/18 - Observation: N/A - Antiplatelet: 0/18 - Heparin: 1/18 - Surgery: For all patients Halo-ring applied as a closed reduction. If unable to stabilise the dislocations - surgical reduction	N/A	0/18 (0%) (during admission)
Bonney	5/45 (11.1%)	Isolated C-spine transverse process fractures	N/A	CTA	N/A	N/A	N/A
Chrowdhury	10/120 (8.3%)	Blunt trauma patients undergoing screening CT neck angiography	N/A	CTA	- Embolization: N/A - Observation: N/A - Antiplatelet: 1/6 (antiplatelet agent alone – Grade I) - Heparin: 5/6 (antiplatelet plus heparin) - Apixaban: 5/6 (3 months) - Surgery: 0/5	- In 3 months follow up at the outpatient department, no further complications or neurological deterioration were observed	2/10 (20%) (during admission)
Sheppard	5/68 (7.4%)	Cervical spine fracture	N/A	CTA	- Embolization: 0/5 - Observation 2/5 (acute intracerebral haemorrhage) - Antiplatelet: 3/5 - Heparin: 0/5 - Surgery: 0/5	/	1/5 (20%) (during admission)
Cloney	8/97 (8.2%)	Trauma atlas (C1) fractures	N/A	CTA or MRA	N/A	- 1/8 ischemic stroke (12.5%)	N/A
Wynn	35/264 (13.2%)	Blunt cervical spine injury and/or fracture who underwent CTA	N/A	CTA +/− MRA	- Embolization: N/A - Observation N/A - Antiplatelet: N/A - Heparin: N/A - Surgery: N/A antiplatelet/anticoagulation therapy. 48.5% were already receiving anticoagulation as outpatient medication	- No patients with VAI required secondary vascular interventions, and no secondary strokes were identified	N/A
Nakamura	15/89 (16.9%)	Traumatic cervical spine injury	N/A	CTA +/− DSA +/− MRA	- Embolization: 2/15 (for bilateral vertebral artery occlusion (2/5)) - Endovascular treatment: 9/15 - Observation N/A - Antiplatelet: 15/15 - Heparin: 15/15 - Surgery: N/A	- Follow-up at 29.4months: 2/15 have neurological deficits from cerebral infraction	4/15 (26.6%) (29.4months)
Gupta	21/149 (14.1%)	Cervical vertebrae fracture/subluxation	This study did not report any catastrophic symptoms of vertebrobasilar insufficiency	CT/CTA/MRA	N/A	- Significant posterior circulation infarcts 4/21	2/21 (9.5%) (during admission)

VAI: Vertebral artery injury. N/A: Not applicable. MRI: Magnetic resonance imaging. MRA: Magnetic resonance angiography. DSA: Digital subtraction angiography. BVI: Blunt vertebral injury. CTA: Computed topography angiography. BCVI: Blunt cerebrovascular injury. MDCTA: Multidetector computed topography angiography. C; cervical spine, V: vertebral artery, VAI: vertebral artery injury, TF: transverse foramen, BCVI: blunt cervical vascular injury, mm: millimetres, TSA: traumatic spondylolisthesis, N/a: not applicable.

Diagnostic Modality

All 24 studies reported on the diagnostic modality they utilised. The modalities used overall were CT, CTA, DSA, MRI, MRA and catheter angiography. In the majority of studies, a combination of 2 or more diagnostic modalities were used on the same patient during their admission/part of follow-up. However, no specific reasoning or rationale was provided as to why the studies employed 2 or more different diagnostics modalities. In total, 13/24 reported using CT/CTA, 10/24 reported using DSA, 10/24 used MRI/MRA and 2/24 used catheter angiography (Figure 6).

Figure 6.

Diagnostic Imaging Modality Utilised and corresponding publications. DSA: Digital Subtraction Angiography, CTA: Computed Topography Angiography. MRA: Magnetic Resonance Angiography, MDCTA: multidetector computed tomographic angiography.

Management

Overall, 13 studies (285 patients) provided comprehensive detail of management strategies. In the majority of studies, the mainstay of treatment was with antiplatelet or anticoagulation agents, or both. Overall, 147/285 (51.6%) received an antiplatelet agent, either as a monotherapy or additional to an anticoagulant after the acute period ranging for 3-6 months. In total, 57/285 (20%) were observed without any antiplatelet or anticoagulant. Authors reported that often this was due to contraindications or acute cerebral haemorrhage on imaging. In total, 95/285 (33.3%) received heparin in the acute period, in the majority of cases for 2 weeks, before either switching to antiplatelet monotherapy or discontinuing all antiplatelet/anticoagulation agents. 5/285 (1.8%) were commenced on Apixaban for 3 months after 2 weeks of intravenous Heparin. Warfarin was commenced in 2/285 (.7%) patients for an unknown amount of time. 5/285 (1.8%) patients underwent embolization of the VA, with 2/5 (40%) of these being for bilateral vertebral artery occlusion. Whether or not a patient underwent surgery for fracture management or to control the VAI is unclear and inconsistent, and therefore we cannot comment on the statistics in this regard. Mueller et al ⁴ was the only study to describe in detail which VAI patients underwent which surgery. Anterior cervical plate-screw fusion 9/19 (47.4%), dorsal rod-screw fusion 1/19 (5.3%) and combined ventral–dorsal fusion 1/19 (5.3%). See Table 2 for full descriptions of reported management.

Outcomes in Those with VAI

Overall, 10/24 studies reported on mortality. 9/10 reported on mortality during the initial admission (Figure 7). The mortality rate in this group was 32/226 (14.2%), ranging from 0-26.2% The reason for the high mortality in 1 study was related to the fact that its study population were those screened for being high risk of BCVI and for whom angiogram was indicated, however, the VAI-attributable mortality rate was only 4.8% (2 of 42). Only 1/10 study reported on a mortality rate at their follow-up (mean: 29.4months) - 4/15 (26.6%).

Figure 7.

In-Hospital Mortality Associated with VAI. VAI: vertebral artery injury.

Overall, 14/24 studies reported on outcomes of patients. The outcomes reported are highly heterogenous and therefore the outcomes are displayed as a descriptive narrative in Table 2. Furthermore, it is challenging to differentiate in the studies whether poor outcomes were a consequence of the cervical spinal injury, the VAI, or both. 6/24 studies (with a total of 62 patients) report on follow-up outcomes at outpatient clinic. The mean follow-up across these 6 studies was 26.6 months (3 - 83.3 months). 5/6 of these studies (52 patients) reported on residual neurological deficit at mean follow-up. 46/52 (88.5%) had no neurological symptoms or signs of vertebrobasilar infarct at mean follow-up. 2/6 studies reported on follow-up imaging at outpatient clinic. 1/62 (1.6%) patient’s follow-up MRI demonstrated the persistence of vertebral artery occlusion and 1/62 (1.6%) DSA at 6 months showed permanent stenosis. Full description of outcomes at follow-up are available in Table 2.

Fracture/Injury to Vertebrae and VA

Vertebrae Level

Overall, 12/24 studies reported on where the level of injury/fracture had occurred. There were 285 fractures/injuries in total amongst all patients. The most common level was C5 (71/285 (24.9%)), followed by C2 (56/285 (19.6%)) and C6 (56/285 (19.6%)). Others included; C1 (33/285 (11.6%)), C4 (31/285 (10.9%)), C3 (20/285 (7.0%)) and C7 (18/285 (6.3%)). 1/24 studies reported on the 3 most common levels (most common: C6, followed by C5, then C4 and C3. 1/24 studies were only concerned with fractures at C1. Similarly, 1/24 studies analysed only C2 fractures, and 1/24 analysed those with C1 and C2 injuries. 8/24 did not report any information on the level (Figures 8 and 9).

Figure 8.

Type 4 left vertebral artery injury with acute ischemic stroke following corrective spinal surgery. A) CTA demonstrating a cervical spine fracture/subluxation with occlusion of the left vertebral artery (arrow). B) CTA demonstrating occlusion of the left vertebral artery (arrow). C) DWI sequence obtained after corrective cervical spine surgery with evidence of an acute ischemic stroke. D) MRA demonstrating restored patency of the previously occluded vertebral artery (1).

Figure 9.

Type A dissecting aneurysm. A) DSA of the left vertebral artery showing a fusiform dilation. b, c, d Consecutive angiograms of the right vertebral artery during balloon test occlusion showing that the site of dissection is opacified distal to proximal (2). Type B dissecting aneurysm. A) DSA of the left vertebral artery showing a fusiform dilation. b, c, d Consecutive angiograms of the right vertebral artery during balloon test occlusion showing that the site of dissection is opacified proximal to distal (2).

Injury/Fracture Characteristics

Overall, the data collected on the fracture pattern was highly heterogenous and the majority of studies did not report on consistent characteristics. 17/24 studies reported on whether the transverse foramen (TF) was involved in the injury, with 420 patients reported on. 190/420 (45.2%) of patients with VAI had TF involvement. 8/24 studies reported on facet joint involvement (192 patients). 77/192 (40.1%) sustained a facet joint injury. In regard to the type of injury sustained to the facet joint: fracture (29/192 (15.1%)), dislocation (28/192 (14.6%)), subluxation (5/192 (2.6%)) and locked joints (2/192 (1.0%))were reported. 7/24 studies reported on vertebral subluxation (220 patients). 77/220 (35%) had a vertebral subluxation. 7/24 reported on vertebral/cranial dislocation (151 patients). 13/151 (8.6%) sustained a dislocation. Of those reported, 1/13 (7.7%) was occipitocervical, 1/13 (7.7%) craniocervical and 1/13 (7.7%) atlanto-occiptal dislocation. 4/24 reported on odontoid/dens fractures (133 patients). 33/133 (24.8%) patients sustained a dens/odontoid fracture. Other less commonly reported characteristics included: (specific fracture terminology) burst fracture (9/52 (17.3%)), Hangman (6/63 (9.5%)), Jefferson (3/15 (20%)), subaxial (21/51 (41.2%)), lateral mass (8/66 (12.1%)), lamina (9/81 (11.1%)), pedicle (5/38 (13.2%)), arch (11/130 (8.5%) and ligamentous injury (19/62 (30.6%)). See Table 3 for the characteristics and fracture patterns of the VAI.

Table 3.

Statistically significant factors associated with greater odds of having vertebral artery injury.

Author	Factor	Statistical Test	OR	95% Confidence Interval (CI)	P
Chung	Facet fracture	OR	34.708	4.358−276.453	.001
	Facetal dislocation	OR	4.067	1.331−12.425	.014
Lebl	Occipitocervical dissociation	Fisher exact test	/	/	<.001
	Basilar skull fracture	OR	5.18	2.07-12.98	<.001
	Mortality	Fisher exact test	/	/	<.001
	ISS	Fisher exact test	/	/	.027
	AS/DISH	OR	8.04	1.30-49.68	.034
Payabvash	Comminuted fracture of C3–C7	RR	13.2	1.5-111	.011
Durand	Dens fracture	Fisher exact test	/	/	.0264
Lockwood	Combined C-1 and C-2	OR	3.8	1.8-7.9	<.05
	Subluxation	OR	4.8	2.2-10.4	<.05
	High-risk cervical fracture per Denver Criteria	OR	19.2	2.6-139.9	<.05
Bonney	Number of fractures	Fisher’s exact test	/	/	<.0012
Cloney	Transverse ligament injury	OR	8.51	1.17-61.72	.034
	Facial injury	OR	7.77	1.05-57.51	.045
	Intoxication	OR	51.42	1.10-2408.82	.045
	Type 3 fracture	OR	.08	.01 .81	.033
Wynn	Subluxation/displacement/dislocation of fracture	OR	2.23	1.08-4.59	.03
	High energy mechanism	OR	2.38	1.14-4.96	.02
	SLIC score >4	OR	2.65	1.07-6.55	.04
	Concomitant lumbar spine injury	OR	4.37	1.13-16.9	.03
Gupta	Lamina fracture	OR	0.2	.07-0.6	.005
	Cerebral infarction	OR	10.5	1.9-59	.008

OR: Odds ratio. CI: confidence interval. P: P value. ISS: Injury severity score. AS: Ankylosing spondylitis. DISH: diffuse idiopathic skeletal hyperosteosis. RR: Relative ratio. SLIC: Subaxial-cervical spine injury classification.

Overall, 8/24 studies reported on the statistical association between transverse foramen injury and VAI (Table 4) 5/8 studies showed a statistically significant association (p<.05). Regarding statistically significant associations, Durand et al was unable to show an overall significant association between fractures involving the transverse foramen (p = .4439), however they demonstrated a significant association between transverse foramen intraforaminal fragment fracture and VAI (p = .0017), and transverse foramen comminuted and intraforaminal fragment and VAI (p = .0012). Nagata et al and Lebl et al found a statistically significant association between displacement into foramen>1 mm and VAI (p = .014 and p = .011 respectively). Other statistically significant characteristics association with VAI can be seen summarised in Table 4.

Table 4.

Significant and non-significant transverse foramen injuries and association to VAI.

		Statistical Test	OR	95% Confidence Interval (CI)	P
Bonney		Fisher’s Exact Test	/	/	.01
Chung		OR	5.598	1.867−16.783	.002
Durand	Fracture involves transverse foramen	Fisher’s exact test	/	/	.4439
	Transverse foramen intraforaminal fragment fracture	Fisher’s exact test	/	/	.0017
	Transverse foramen comminuted and intraforaminal fragment	Fisher’s exact test	/	/	.0012
Lockwood		OR	6.3	3.5-11.4	<.05
Nagata	Fracture involves transverse foramen	Chi-squared test	/	/	.019
	Displacement into foramen>1 mm	Chi-squared test	/	/	.014
	Presence of gross fracture (>2 mm displacement)	Chi-squared test	/	/	.028
Payabvash	C1–C2 transverse foramen fracture	RR	1.1	1.0-1.2)	>.05
	C3-7 transverse foramen fracture	RR	2.2	(1.0-4.8)	>.05
Lebl	Fracture involves transverse foramen	OR	1.31	.64-2.66	.458
	Displacement into foramen ≥1 mm	OR	2.37	1.21-4.64	.011
Wynn		OR	4.27	2.05-8.89	.0001

OR: Odds ratio. CI: Confidence interval. P: P value. Mm: millimetre. RR: Relative ratio.

Discussion

Injury to the cervical spine occurs in 2.4% of blunt trauma patients. ³⁴ The majority of the spinal injuries (60%) involve young healthy males between 15 and 35 years old with cervical spine injuries being the most common location of traumatic injury to the spine. ³⁵ The cervical spine remains the most common level for spinal cord injury (SCI), representing 55% of all SCIs. ³⁶ The first incidence of VAI associated with cervical spine fracture was described by Carpenter et al in 1961. ¹² Since then, a variation of VAI incidence (.53-39%) has been reported in the literature, with high rates of morbidity and mortality shown across studies. Certain studies have reported mortality rates as high as 31%. ³⁷ From our review, only 1 study examined VAI attributable mortality (4.8%). ²⁰ We found a collative mortality rate of 14.2% during initial admission, pooled with other risk factors for mortality in cervical trauma.

As such, identification and diagnosis is paramount to negating morbidity and mortality rates associated with VAI in cervical spine trauma patients. This can prove difficult in a significant proportion of patients as many present without symptoms. Therefore, practicing spine surgeons must maintain a high degree of clinical suspicion. Recent studies have sought to elucidate high risk characteristics predictive of VAI. Results have identified the following risk factors; upper cervical fracture (OR: 2.34 (95% CI 1.93, 2.84) p<.001), multiple levels fractured (1.25 (95% CI 1.04, 1.51) p = .02), involvement of the transverse foramen (OR: 11.31 (95% CI 9, 14.22) p<.001), high-energy injuries (OR: 1.79 (95% CI 1.43,2.23) p<.001), and subluxation or dislocation (OR: 1.96 (95% CI 1.43,2.63) p<.001).^38,39 The intimate relationship of the VA with the transverse process in the cervical spine requires close attention, especially when the transverse process fracture extends into the transverse foramen (TF).^29,35 Woodring et al ⁴⁰ performed arteriography on 8 patients who had cervical fracture with TF involvement - 7 (88%) had blunt VAI. This fragile relationship between TF fractures and VAI is corroborated by our findings, which found an overall incidence of VAI in TF fracture of 45.2%.

High clinical suspicion can be aided by modern advancements in diagnostic imaging. The main diagnostic techniques utilized include; CTA, DSA, MRA, catheter angiography and less notably, US doppler. Although DSA is traditionally the gold standard, our study suggests that CTA is the most commonly adopted in this setting (54.2% of studies included in the systematic review). Numerous studies have demonstrated high accuracy levels of CTA in detecting “clinically significant injuries”.^41,42 CTA is becoming the opportunistic screening test of choice under both urgent and emergency circumstances.

However, review of the literature shows discordance in its use as a first line diagnostic tool. Mitha et al report CTA has high sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of CT of 98%, 100%, 100%, 99%, and 99% respectively. ⁴³ In Mueller et al, ⁴ results show a high degree of concordance between DSA and CTA in the detection of a VAI. Nonetheless, prior studies have demonstrated false positive rates as high as 47% when used in the setting of blunt cervical injury for identification of VAI. ⁴⁴ Similarly, a systematic review in 2013, which highlighted 8 studies examining 5704 carotid or vertebral arteries in 1426 trauma patients, the sensitivity and specificity for BCVI detection with CTA vs DSA was 66% (95% CI, 49%-79%; I = 80.4%) and 97% (95% CI, 91%-99%; I = 94.6%), respectively. Novel CT models are capable of much faster acquisition times (multidetector CT) compared to conventional CT. Many centres believe this has improved accuracy of VAI, ⁴² yet there remains a lack of comparative robust evidence to date for identification accuracy of VAI. MRA, another non-invasive diagnostic procedure, has considerable advantages, including; the lack of iodinated contrast products, avoidance of bony artefact, and earlier diagnosis of cerebral infarction. ⁴⁵ However, screening studies have reported poor specificity at 67%, with sensitivities ranging from 50-75%. ⁴⁶ Similarly, a recent meta-analysis reported an overall sensitivity of 55%. ⁴⁷

Similar to VAI identification and diagnosis, it is vital that this devastating injury is managed in an organised, evidence-based manner in order to improve patient outcomes. Furthermore, spinal trauma centres should take a multidisciplinary approach, involving early neurological, orthopaedic and vascular input. The mainstay management strategy for surgically inaccessible injuries is antithrombotic therapy,^48,49 however, associated injuries often present a contraindication to antithrombotic therapy. The literature has shown up to 30% of trauma patients are contraindicated for antithrombotic therapy based on concomitant injuries.^31,50 Endovascular intervention is another option, however stents also require antithrombotic therapy post-placement. For patients with BCVI who do not have a persistent neurologic deficit in the territory of the injured artery and who do not have contraindications, antithrombotic therapy is recommended (unfractionated heparin and antiplatelet agents) over no such therapy.^51-53 Neurologic outcomes are improved in symptomatic patients, and fewer ischemic neurologic events (stroke) occur in asymptomatic patients who are treated with antithrombotic therapy.^54,55 The optimal medical treatment of VAI is still under debate, and an evidence based protocol on anticoagulation in patients with VAI does not exist. However, anticoagulative treatment appears standard therapy to date, as it is associated with reduced occurrence of secondary stroke.^56,57 Previous studies showed a decreased stroke rate from 30-50% to 2-10% using anticoagulation or antiplatelet therapy.^56,58 Regarding surgical intervention, early surgical fixation is considered to be indicated when concerned with an unstable cervical spine fracture in order to prevent further potential neurological and VAI damage. ⁴³ Although a consensus regarding VAI treatment is lacking, early diagnostic steps followed by a standardised treatment protocol is key. Intravenous anticoagulation, and if accessible and indicated – early stabilization, in the acute stage followed by an oral anticoagulant are seen as protection of the vertebrobasilar circulation, although more robust randomised trials are needed.

Delphi studies are an example of technique which can lead to improved treatment strategies and protocols. This model can improve efficient use of health care resources within the field of cervical trauma and has been utilized with success in other challenging pathologies ⁵⁹ and in the setting of acute cervical spine trauma. ⁶⁰

The review had several limitations. Firstly, the majority of the studies were retrospective reviews. Robust prospective, randomised controlled trials and qualitative Delphi studies are needed to define treatment and diagnostic protocols. Although the incidence of VAI in blunt cervical trauma remains low, the importance of deriving effective treatment protocols remain essential due to poor outcomes and high mortality rate. Secondly, there was a small number of directly relevant studies in the literature, and the quality of studies was low. Furthermore, there are no randomised controlled trials regarding management strategies to date. Thirdly, there was heterogeneity of study design, outcome measures and study populations between the included studies which precluded substantial collative meta-analysis. Finally, the true VAI attributable mortality rate was not calculated, given only 1 study commented on it. Furthermore, the majority of studies were retrospective in design and did not control for other confounding factors.

Nevertheless, our study provides an indication of incidence, characteristics, and effective management strategies that will aid spine surgeons in maintaining a high degree of clinical suspicion in order to attempt to minimise morbidity and mortality rates associated with VAI.

Conclusion

VAI secondary to cervical spine trauma has a notable incidence and high associated mortality rates. Ultimately, this is a rare but life-threatening injury. There is a high incidence of VAI when associated with transverse foramina injury. The current available literature is limited by a low quality of evidence. In order to optimise diagnostic protocols and treatment strategies, in addition to reducing mortality rates associated with VAI, robust quantitative and qualitative studies are needed. Utilizing evidence-based clinical parameters to predict chance of VAI may avoid unnecessary advanced imaging, alongside improving patient outcome. Nevertheless, practicing spine surgeons should reserve a high degree of clinical suspicion for associated VAI when dealing with traumatic cervical injuries.

ORCID iDs

Hugo C. Temperley https://orcid.org/0000-0001-9151-3431

Jake M. McDonnell https://orcid.org/0000-0002-8002-8024

Grainne Cunniffe https://orcid.org/0000-0001-7085-1706