Incidence and outcomes of posterior circulation stroke following traumatic vertebral artery injury: A 6-year single-center retrospective cohort study

ABSTRACT

Objectives:

Vertebral artery injury (VAI) is an uncommon, but potentially devastating, complication following cervical spine trauma. Posterior circulation ischemia (POCS) can result from a disruption to the vertebral arteries and potentially lead to long-term morbidity and mortality for these patients. This study aimed to describe the outcomes for patients with POCS as a result of VAI.

Methods:

Six hundred and twenty patients who sustained a cervical spine fracture with or without dislocation were identified from the years 2011 to 2017. Demographic data, injury details, and imaging results were collected with inpatient and outpatient records on the regional electronic care record. Patients with VAI were identified (n = 20) and screened for subsequent POCS. Their outcome was recorded from inpatient and outpatient records.

Results:

POCS was identified in 6 patients who had sustained a VAI (30%). Of those 6 patients, 5 had long-term neurological deficits following their injury and POCS. Two patients died within 30 days and 3 patients had died by 60 days following injury.

Conclusions:

POCS occurred in 30% of patients who sustained a VAI in this study. It carried significant morbidity and mortality for the patients who suffered this potentially devastating complication. Clinicians involved in the care of patients with cervical spine trauma should be aware of the risk factors for VAI and ensure to be suspicious of POCS in patients who have proven VAI.

INTRODUCTION

Vertebral artery injury (VAI) is an uncommon but important complication following cervical spine trauma. A recent systematic review of the available literature by Temperley et al. reported an overall incidence of 5.19%,[1] however, the reported incidence is widely varying depending on the study, screening tests, and injury mechanism.[2,3,4,5]

The vertebral arteries are paired, each arising from the first part of the ipsilateral subclavian artery. Each has 4 “segments” (V1-4) defined by its anatomical course. Of note, it is well reported that the left vertebral artery is often dominant and individuals may have unilateral hypoplasia in up to 10%.[6]

• V1: Preforaminal-from origin to the transverse foramen of C6

• V2: Foraminal segment-artery ascends within transverse foramina of C6-C2

• V3: Extradural-Passage from C2 through C1 before ascending to pierce the dura

• V4: Intradural-From the dura to their confluence to form the basilar artery.

The vertebral arteries and basilar arteries provide blood supply to the posterior brain, cerebellum, brainstem, and upper spinal cord. The course of the vertebral artery through the transverse foramen results in an increased susceptibility to injury in cervical spine trauma. Given the regions supplied by the vertebral arteries, a potentially devastating complication of traumatic VAI is posterior circulation ischemia. Studies suggest that risk factors include upper cervical spine fracture (C1–3), involvement of the foramen transversarium (TF), and subluxation/dislocation injuries.[7,8,9] Although it is uncommon, posterior circulation stroke (POCS) as a result of VAI is a potential cause of long-term morbidity and mortality. Reported POCS rate of VAI differs in the literature with one study by Miller reporting 0% incidence of stroke and Biffl et al. reporting 24%.[7,10,11]

Clinical signs/symptoms

The often asymptomatic nature of traumatic VAI necessitates a level of clinical suspicion in those with injuries susceptible to VAI. Due to adequate collateral circulation supplied by the contralateral artery, traumatic VAI is thought to be often asymptomatic.[10,12,13] However, in those that are symptomatic, symptoms and signs are most often due to ischemia of the brainstem, visual cortex, and cerebellum. These include head-and-neck pain, gait disturbance, dizziness, nausea, reduced consciousness, speech abnormalities, and visual disturbance.[14]

Diagnosis and treatment

Imaging modalities used in the diagnosis of VAI include computed tomography and magnetic resonance angiography (CTA and MRA), digital subtraction angiography, and duplex ultrasonography (USS).[15] CTA is widely available and appears to have good sensitivity and specificity when compared with other imaging modalities.[16,17,18] Modern CT scanners have also continued to improve the imaging quality and diagnostic accuracy.[19] A benefit compared to MRA is also the speed at which it can be undertaken, which is an important factor in the trauma patient. Digital subtraction angiography has been reported as the gold standard investigation, however, it itself is less safe in the head-and-neck trauma patients.[16] Duplex USS has a poorer reported sensitivity[20] than CTA, and in this setting, is not used in the authors’ institution.

The management of traumatic VAI has not been well reported on in the literature. Strategies include observation, antiplatelets, anticoagulation, and endovascular therapy. A single large study published in 2009 suggests the early use of treatment in patients with blunt cervical spine injury may reduce the subsequent infarction risk, however, often stroke occurs early and is unpreventable.[21] Owing to associated injuries (including intracranial injuries), anticoagulant therapy may be contraindicated in many patients with VAI.

Objectives

Primarily, this study aimed to identify the outcomes of patients who sustained a posterior circulation stroke as a result of VAI in the setting of cervical spine trauma.

This study aimed also to assess the incidence of VAI in patients with cervical spine fracture, with particular interest paid to the presence of risk factors (TF involvement, subluxation/dislocation, and upper cervical spine fracture).

METHODS

This study was registered and approved by the Audit Department in the Royal Victoria Hospital (RVH), Belfast. De-identified data were provided, following approval, from the Fractures Outcomes Research Database (FORD). Due to the retrospective and anonymized nature of the study, informed consent was not required.

The authors interrogated the FORD for all cervical spine fractures presenting to the RVH, Belfast, between 2011 and 2017. A total of 620 patients were identified as having sustained a traumatic cervical spine fracture.

Information including age, sex, date of injury, mechanism of injury, TF involvement, head injury, and subsequent angiographic studies was obtained and recorded in a Microsoft Excel^® spreadsheet.

Imaging reports were accessed through the Northern Ireland Electronic Care Record (NIECR), ensuring the reporting was performed by a consultant radiologist.

The reports of discharge letters and outpatient clinics were obtained using NIECR. This allowed careful documentation of outcomes including permanent neurological sequelae and mortality.

RESULTS

Demographics

A total of 238 females and 382 males were included in this study. The mean age of the patient was 58.8 years (range: 15–104 years).

A total of 84 patients had associated head injuries defined as intracranial hemorrhage or skull fracture.

Of the 620 fractures, 120 (19.4%) were identified as having involved the TF and 45 (7.3%) were identified as having subluxation or dislocation of the facet joints. Three hundred twenty-two (51.9%) patients had fractures involving C1–C3 [Table 1].

Table 1.

Total number of patients with an identified risk factor for vertebral artery injury

Risk factor (identified on imaging)	N	VAI number
Foramen transversarium involvement	120	15
Subluxation/dislocation	45	7
C1–3	322	11
None of the above	210	0

Column 2 shows the number of VAI sustained with each risk factor. VAI - Vertebral artery injury; C1–3 - Cervical spine fracture

One hundred CT angiogram studies were performed, of which 5 were performed on a separate admission due to ongoing symptoms. MRA was utilized a total of 9 times for assessment of the vertebral artery.

Twenty vertebral artery injuries were identified in the group of 620 patients (3.2%). Two patients sustained bilateral VAI. Six patients (30%) sustained a VAI and had subsequent imaging that confirmed posterior circulation ischemia. A majority, 4 (66.6%), of those patients with VAI resulting in POCS had fracture-dislocation injuries. The other 2 patients suffered C2 fractures.

All patients with VAI had one of the identified risk factors in Table 1. Nine (45%) had a single risk factor, 8 (40%) had two risk factors, and 3 (15%) had all three risk factors. Of the patients who sustained a POCS, 4 (67%) had a single risk factor, 1 patient (17%) had two risk factors, and 1 patient had all three risk factors. The most common risk factor identified in patients with POCS following VAI was a dislocation (in 4 of 6 patients).

Table 2 outlines some important demographics and characteristics of each patient who suffered from POCS in this study. The average age of those involved was 66 years old and a majority suffered a dislocation type injury. Only 1 patient died within 30 days following POCS, however, a further 2 patients died within 60 days. Giving a 60-day mortality of 50% in this study, 2 further patients died at 1.3 and 5.5 years.

Table 2.

Demographics, injury patterns, management, and outcome for the patients who sustained posterior circulation ischemia in this study

Patient number	Age	Injury	Side of VAI	Management	POCS distribution	Other neurological deficit	Mortality days
1	82	C5/6 bi-facetal fracture	Right	Posterior stabilization	Cerebellar	Incomplete cord injury	33
2	44	C5/6 bi-facetal fracture	Bilateral	Posterior stabilization	Occipital/temporal	Complete cord injury	1996
3	80	C4/5 uni-facetal fracture	Bilateral	Posterior stabilization	Cerebellar/pons	Incomplete cord injury	9
4	78	C5–6 bi-facetal	Left	Posterior stabilization	Occipital/cerebellar	Incomplete cord Injury	52
5	78	Type 3 C2 fracture involving foramen	Right	Halo	Medullary infarct (bulbar palsy)	Hemiparesis	483
6	32	Type 3 C2	Left	Rigid collar (Aspen)	Cerebellar	Nil	Not deceased

POCS - Posterior circulation ischemia; VAI - Vertebral artery injury

DISCUSSION

As has previously been described, VAI and posterior circulation ischemia are a source of significant morbidity and mortality following cervical spine trauma. The aim of the study was to describe the outcomes of patients who sustained a POCS following traumatic VAI, as this information is not widely available in the literature.

For these patients, they suffered significant morbidity following their injury. Given the nature of the high cervical injury, attributing mortality solely to POCS can be difficult in these patients. A review of the patient’s notes available on their electronic care record suggests POCS contributed significantly to morbidity and eventual mortality in patients 1, 3, 4, and 5. Patient 2 had tetraplegia secondary to a high-level cord injury (C5/6) and as such, it is difficult to attribute symptoms solely to POCS. It clearly, however, added significant complexity to the case. The patient’s death was over 5 years from injury but ultimately was attributable to his disability.

Four of six patients with POCS had a cervical dislocation injury. Of these, all died with significant resultant neurological deficits relating to their injury. When compared with the other patients in the database with dislocation-type injuries (45), 8 (17.7%) died with ongoing disability. Dislocation pattern injuries are often disabling and appear to be devastating when they cause a VAI and subsequent POCS. It is suggested, therefore, that clinicians remain suspicious of VAI in the context of dislocation pattern injury and remain vigilant for signs of posterior circulation ischemia.

This study is one of the largest single-center reviews in the UK and Ireland to look at the incidence of VAI, posterior circulation ischemia, and death. The overall incidence of VAI in this study (3.2%) was largely comparable with a large multicenter systematic review.[1] The overall incidence of POCS following VAI in this study was 30%, and given the significant morbidity and mortality associated, the authors suggest a high index of suspicion for posterior circulation ischemia in any patient with VAI (bilateral or unilateral).

CONCLUSIONS

Posterior circulation ischemia occurred in 30% of patients in this study who had sustained VAI. There was significant associated morbidity and mortality, with 50% of patients dying within 60 days from injury. The authors suggest clinicians are suspicious of VAI when a patient presents with one or more of the suggested risk factors on imaging and are aware of the signs and symptoms of posterior circulation ischemia which can occur in these patients.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.