Introduction

In modern reports, blunt cerebrovascular injury (BCVI) is diagnosed in 1–3% of patients who suffer from blunt trauma [1,2,3]. Subsequent stroke can result in debilitating morbidity and mortality. Patients with untreated BCVIs have a stroke risk of up to 40%, with variable morbidity based on the vessel injured and grade of injury [4]. Over the past several decades, regionalization of trauma care and advances in screening technology have increased recognition of BCVI, resulting in earlier treatment with medical therapy and decreased stroke rates.

Early screening was initially performed using diagnostic angiography. Today, computed tomography angiography (CTA) is the primary diagnostic modality with angiography reserved for select cases [5]. Systematic screening for BCVI has resulted in earlier diagnosis. Several screening guidelines have been proposed, all of which include patients considered high risk by injury pattern [6, 7]. Selective use of screening will underestimate the true incidence of BCVI and with the ever expanding “criteria” included in protocols, universal screening is becoming a more common alternative [8•, 9••].

The mainstay of BCVI treatment is medical therapy with anticoagulation or antiplatelet agents. No dedicated study has evaluated the use of pharmacologic agents compared to one another, however either therapy appears acceptable [10, 11••]. More importantly, it appears to be the early initiation of medical therapy that decreases stroke rate [2, 4]. Use of endovascular interventions was initially explored for higher grade injuries, but recent literature advises against routine use of stents with stroke rates remaining low with medical therapy alone [11••, 12, 13]. The aim of this review is to describe the evolution of diagnostic and management strategies in BCVI in the literature.

Diagnostic Modalities

BCVI was initially diagnosed after patients demonstrated neurologic sequelae following blunt trauma. Typically, the neurologic symptoms presented after a latent period and injury was confirmed using angiography. Fabian et al. identified 21 patients with blunt carotid artery dissections on angiography after they presented with a change in neurologic examination, or an initial neurologic deficit not explained by computed tomography (CT) imaging [14]. This prompted more liberal use of diagnostic angiography in blunt trauma patients. Over an 8-year period Cothren et al. screened 727 patients with diagnostic angiography, of which 244 were diagnosed with BCVI. The authors concluded comprehensive screening with diagnostic angiography prevented strokes and was cost-effective [15].

While angiography was the early diagnostic modality of choice, less invasive and less expensive modalities were explored as an alternative for screening. Berne et al. used 4-slice CTA to screen 486 patients and found 19 patients with BCVIs, which were subsequently confirmed with angiography. The authors stated there were no patients with a negative CTA that had a missed injury [16]. However, others found CTA inadequate compared to angiography with sensitivity for BCVI diagnosis ranging from 29 to 68% [17,18,19]. Magnetic resonance angiography (MRA) and duplex doppler ultrasound were also found to be insufficient with sensitivities of 50% to 75% and 38.5%, respectively [17, 18, 20].

CT imaging has improved with advancement in technology; however, angiography remains the gold standard in diagnosis. Despite superior sensitivity with angiography, CTA has emerged as the principal modality for screening likely due to its use in combination with liberal screening criteria, ease, cost, and non-invasive nature [11••, 21•]. At our institution, we screen patients with CTA and when BCVI is identified or imaging is equivocal, angiography is subsequently performed. While CTA is widely accepted, its diagnostic ability is still limited compared to angiography.

Screening Criteria Guidelines

The first reports of BCVI occurred after patients experienced neurologic sequelae followed by confirmatory imaging. Understanding the pathophysiology and common symptomatic presentation, led to the development of screening guidelines based on associated injuries, mechanism of injury, and symptoms to identify patients with potential BCVI.

Biffl et al. demonstrated patients with Glasgow coma score ≤ 6, petrous bone fracture, diffuse axonal injury, or Lefort II or III fracture, in the setting of high-risk mechanisms (hyperextension, hyperflexion, or direct blow) had a 41% risk of carotid injury. 39% of patients with cervical spine fractures were found to have vertebral artery injuries. Importantly, 20% of patients diagnosed with BCVI did not have any of those risk factors, which highlights the need for broad screening to prevent missed injuries [22]. Further investigation resulted in the more liberalized Memphis Criteria which included cervical spine fractures, Lefort II or III fractures, Horner’s syndrome, skull base fractures involving the foramen lacerum, neck soft tissue injury, or neurological abnormalities unexplained by CT head and yielded a 29% BCVI diagnosis when implemented as a screening protocol. After implementation of the screening criteria, 91% of BCVIs were detected prior to the development of neurologic sequelae [17]. Cervical spine fractures involving subluxation, extension into the foramen transversarium, and C1 to C3 fractures were injury patterns specifically associated with vertebral artery injuries [23].

The Denver Criteria incorporated signs and symptoms of BCVIs and risk factors associated with injury patterns and mechanism. Of 727 patients screened, 244 (34%) patients were diagnosed with BCVI [15]. When the Denver Criteria were incorporated into a screening protocol, a significant increase in BCVI diagnosis (0.52% vs 1.06%), yet a decreased rate of stroke was reported (40% vs 27%). Since the development of the Memphis and Denver screening criteria in the early 2000’s, the criteria have been refined and expanded [3, 4, 6, 24]. The American College of Surgeons Trauma Quality Improvement Program (ACS TQIP) Best Practices Guidelines in Imaging incorporates 15 signs and findings based on physical exam, mechanism, and injuries into a BCVI screening protocol [25].

Leichtle et al. implemented universal screening at their institution and found almost 20% of patients with BCVI would not have been diagnosed by using the expanded Denver criteria or the ACS TQIP Best Practices Guidelines in Imaging [25]. Most recently, Black et al. employed universal screening and concluded 25.3% to 52.7% of patients diagnosed with BCVI would not have been screened for BCVI using the Memphis, Denver, or expanded Denver criteria. 7.6% of patients who suffered from blunt trauma were diagnosed with BCVI [9••].

Early screening protocols were developed to identify high-risk patients without over utilizing resources [22]. Expansion of criteria and now universal screening have captured the true incidence of BCVI, decreasing the devastating morbidity and mortality from neurological sequelae. In our institution, we universally screen blunt trauma patients with CTA for BCVI with a protocol developed in conjunction with the radiology department to minimize contrast load.

Medical Therapy

BCVIs are difficult to operate on due to their inaccessible location, therefore treatment is with either systemic anticoagulation or antiplatelet therapy. Fabian et al. demonstrated heparin therapy to be independently associated with improvement in neurologic outcome and survival. Heparin therapy started before the onset of symptoms prevented neurologic deterioration [26]. Further studies found anticoagulation with heparin, antiplatelet therapy, or a combination of both demonstrated decreased incidence of stroke in patients with BCVIs [2, 4, 10].

In many studies, a combination of anticoagulation and antiplatelet therapy is used in treatment, making it difficult to determine the superiority of one therapy compared to another. Cothren et al. identified 282 asymptomatic BCVIs and initiated treatment with either heparin, aspirin, or aspirin and/or clopidogrel. Only one patient developed a stroke, injury healing rates were similar (39% vs 43% vs 46%) among treatment groups and injury progression rates were also similar (12% vs 10% vs 15%) among treatment groups [10]. Esposito et al. published a large multicenter trial evaluating factors associated with stroke formation. There were 777 BCVIs included in the study with a stroke rate of 8.9%. The authors found use of antiplatelet therapy, specifically aspirin, during the hospitalization was associated with lower stroke rates [27•].

At our institution, treatment with systemic heparin is initiated as soon as possible. Questionable findings or injuries identified on CTA are confirmed using angiography. Although we usually repeat imaging within 1 week after diagnosis to evaluate for resolution of injury, work from Wagenaar et al. showed early follow up imaging of high-grade injuries may not be warranted because high-grade injuries did not resolve [28]. This area requires further study. Long-term treatment is then transitioned to single or dual-antiplatelet therapy.

Timing of Medical Therapy

In early studies, BCVIs were diagnosed after neurologic symptoms presented in patients, and subsequent treatment with medical therapy was started reactively [14, 26]. The authors concluded that since patients improved with medical therapy, earlier diagnosis and treatment would improve outcomes. Management shifted to aggressive screening focusing on early identification and treatment of BCVIs during their “latent” period, optimizing outcomes. Several studies demonstrated decreased stroke rate when BCVIs were treated before patients developed neurologic sequalae [2, 4, 15].

A more challenging scenario is when to start treatment in patients who have concomitant traumatic brain injuries (TBI) and/or solid organ injuries (SOI). Management strategies in this patient population are controversial and not well-described. Shahan et al. evaluated 119 patients, with BCVI and either TBI or SOI or both, treated with anticoagulation or antiplatelet therapy. The mean time to therapeutic active partial thromboplastic time (aPTT) from diagnosis was 7 h. Medical therapy did not increase risk of worsening TBI or SOI [29].

However, a recent multi-institutional trial found treatment delay may not necessarily result in stroke. In the 16-center trial, 636 BCVIs were identified, and treatment was considered delayed if initiation was > 24 h. Median time to medical therapy was 62 h in the delayed group compared to 11 h in the non-delayed group. ISS was greater in the delayed group, but there was no overall difference in stroke rate between groups. The authors conclude necessary delays in treatment in patients with concomitant injuries may not result in increased stroke rate [30•].

Modern treatment of BCVI hinges on early diagnosis and initiation of medical therapy. At our institution we initiate medical therapy as soon as possible, including all patients with SOI, TBI and spine injuries (unless patients go directly to the operating room). Heparin infusion is initiated when patients with TBI demonstrate hemorrhage stability on CT imaging. This is usually achieved 6 h after their initial CT scan, however in cases when TBIs worsen on repeat imaging, time to heparin infusion can be delayed by 24 to 36 h. Patients are closely monitored in an intensive care unit setting.

Endovascular Interventions

Injuries with luminal narrowing and pseudoaneurysms were initially thought to benefit from therapeutic endovascular stenting. Early studies demonstrated feasibility and successful outcomes by decreasing the risk of embolism and rupture with stent placement [31, 32]. However, the role of endovascular stenting has evolved over that past few decades. After a significant increase in use, further studies reported risk of stent thrombosis and neurologic deficits after stent placement [33, 34]. Burlew et al. reported stent use had significantly decreased over the study period. The authors found antithrombotic treatment was effective in high grade BCVIs and routine stenting did not provide additional benefit [12]. Shahan et al. reported similar findings, with a decrease in stent use (34% vs 8.9%) over a 5-year period, yet no change in stroke rate [13]. Recently published practice management guidelines recommended against routine use of endovascular stenting [11••].

At our institution, the use of endovascular stents has dramatically decreased over the years. Occasional use in an enlarging carotid pseudoaneurysm or dissection with significant stenosis may be warranted. Since treatment with anticoagulation or antiplatelet therapy alone has demonstrated decreased risk of stroke and mortality, the role for endovascular stent placement is limited. Possible procedural complications, need for long-term antiplatelet therapy, and need for repeat imaging and/or procedures must be considered when determining the treatment plan for the individual patient.

Conclusion

Diagnosis and management of BCVI has evolved over the past several decades. CTA is a noninvasive diagnostic modality that has increased in use and is considered standard in screening. Its ease and cost make liberal screening possible which is critical to early diagnosis and treatment to prevent the devastating morbidity and mortality associated with stroke events. Universal screening results in BCVI diagnosis in a substantial number of patients who would not usually be identified using even the most aggressive screening criteria. CTA used in combination with universal screening helps overcome the diagnostic limitations of CTA. Medical therapy with anticoagulation or antiplatelet agents has been the mainstay of treatment and can safely be used in patients with TBI and SOI. The use of endovascular stents has decreased over time due to risk of complications and should only be considered in select cases.