Endovascular Stenting Is Rarely Necessary for the Management of Blunt Cerebrovascular Injuries Clay Cothren Burlew, MD, FACS, Walter L Biffl, MD, FACS, Ernest E Moore, MD, FACS, Fredric M Pieracci, MD, FACS, Kathryn M Beauchamp, MD, FACS, Robert Stovall, MD, Amy E Wagenaar, BA, Gregory J Jurkovich, MD, FACS The role of stenting for blunt cerebrovascular injuries (BCVI) continues to be debated, with a trend toward more endovascular stenting. With the recent intracranial stenting trial halted in favor of medical therapy, however, management of BCVI warrants reassessment. The study purpose was to determine if antithrombotic therapy, rather than stenting, was effective in post-injury patients with high-grade vascular dissections and pseudoaneurysms. STUDY DESIGN: In 1996, we began screening for BCVI. After the 2005 report on the risks of carotid stenting for BCVI, a virtual moratorium was placed on stenting at our institution; our primary therapy for BCVI has been antithrombotics. Patients with grade II (luminal narrowing >25%) and grade III (pseudoaneurysms) injuries were included in the analysis. RESULTS: Grade II or III BCVIs were diagnosed in 195 patients. Before 2005, 25% (21 of 86) of patients underwent stent placement, with 2 patients suffering stroke. Of patients treated with antithrombotics, 1 had a stroke. After 2005, only 2% (2 of 109) of patients with highgrade injuries had stents placed. After 2005, no patient treated with antithrombotics suffered a stroke and there was no rupture of a pseudoaneurysm. CONCLUSIONS: Antithrombotic treatment for BCVI is effective for stroke prevention. Routine stenting entails increased costs and potential risk for stroke, and does not appear to provide additional benefit. Intravascular stents should be reserved for the rare patient with symptomatology or a markedly enlarging pseudoaneurysm. (J Am Coll Surg 2014;218:1012e1017. 2014 by the American College of Surgeons)
BACKGROUND:
The role of endovascular stenting for blunt cerebrovascular injuries (BCVIs) continues to be debated in the literature. Despite our cautionary report in 2005, which highlights the risks associated with carotid artery stenting for BCVI,1 recent reports continue to advocate endovascular therapy for high-grade injuries.2 In 2011, the National Institutes of Health released a clinical alert reporting the halted trial of intracranial stenting in favor of medical therapy alone for high-grade carotid stenoses.3 That trial’s Data Safety Monitoring Board reported a 14% stroke/death rate with carotid stenting and medical
Disclosure Information: Nothing to disclose. Presented at the American College of Surgeons 99th Annual Clinical Congress, Washington, DC, October 2013. Received October 1, 2013; Revised January 21, 2014; Accepted January 22, 2014. From the Department of Surgery, Denver Health Medical Center and the University of Colorado School of Medicine, Denver, CO. Correspondence address: Clay Cothren Burlew, MD, FACS, Department of Surgery, Denver Health Medical Center and University of Colorado School of Medicine, 777 Bannock St, MC 0206, Denver, CO 80204. email: [email protected]
ª 2014 by the American College of Surgeons Published by Elsevier Inc.
therapy vs a 5.8% stroke/death rate in the medical treatment alone arm. In light of this information, the empiric use of vascular stents to manage BCVI warrants critical reassessment. The purpose of this study was to determine if antithrombotic therapy, rather than stenting, was effective in post-injury patients with high-grade vascular dissections (Denver grade II BCVI) and pseudoaneurysms (Denver grade III BCVI).
METHODS Denver Health Medical Center is a state-certified and American College of Surgeonsverified level I regional trauma center, and integral teaching facility of the University of Colorado Denver. In January 1996, we initiated comprehensive screening for BCVI based on injury mechanism, injury patterns, or symptoms. Screening criteria have been modified based on subsequent analyses, yielding our current management guideline (Fig. 1).4-6 The decision to screen a patient for BCVI outside the noted guidelines was at the discretion of the trauma attending. We currently use multi-slice (64-slice) CT angiography as our standard screening test for patients at risk for
1012
ISSN 1072-7515/14/$36.00 http://dx.doi.org/10.1016/j.jamcollsurg.2014.01.042
Vol. 218, No. 5, May 2014
Burlew et al
Stenting and Blunt Cerebrovascular Injuries
1013
Figure 1. Current screening criteria for blunt cerebrovascular injuries at Denver Health Medical Center. BCVI, blunt cerebrovascular injuries.
BCVI. During daytime hours, a neuroradiologist reads the CT angiography; at night, an attending radiologist reviews images in real time, with an over-read in the morning by a neuroradiologist. Digital subtraction 4-vessel angiography was used to evaluate patients for the first 10 years of the study period. Patients’ injuries were classified according to the Denver grading scale7 (Table 1). Before 2005, only patients with pseudoaneurysms were considered for endovascular stenting. After our 2005 report on the risks of endovascular therapy for BCVI, a virtual moratorium was Table 1. Denver Grading Scale for Blunt Cerebrovascular Injuries Grade I: irregularity of the vessel wall or a dissection/intramural hematoma with 25%) and grade III (pseudoaneurysms) injuries were included in the analysis due to reports advocating stenting for these injuries.2 To date, endovascular stenting for minimal luminal narrowing (grade I injury) or complete occlusion (grade IV injury) has not been advocated, so these patients were not included in the analysis. BCVI grades, treatment, indication for stenting, and inpatient outcomes during the study period were analyzed. Patients who presented with ischemic/stroke symptoms were excluded. Statistical analysis was performed using SAS software for Windows (SAS Institute). The Colorado Multi-Institutional Review Board, our University-based IRB, approved this study.
RESULTS During the 15-year study period, 195 patients were diagnosed with grade II or III BCVIs. The majority was men (67%), with a mean age of 39 years (range 19 months to 83 years). Before 2005, there were 86 asymptomatic patients who sustained a grade II/III BCVI; 39 patients had an isolated carotid artery injury (CAI), 32 patients had an isolated vertebral artery injury (VAI), and 15 patients had a combined CAI and VAI. Of the 86 asymptomatic patients, 21 (25%) patients underwent stent placement (Table 2). Endovascular therapy included 18 carotid stents and 3 vertebral stents. Two patients suffered a stroke due to their stent placement. The first was a 49-year-old man with a grade III CAI. Either during or immediately after stent placement, dense hemiplegia developed in the patient. Heparin was started after the major stroke; several days later the patient had an intracranial hemorrhage requiring craniotomy and evacuation. He was discharged to our rehabilitation unit; although he had overall improvement in his neurologic status, upper-extremity motor deficits persisted. The second patient was a 38-year-old man with a right CAI that had progressed from a grade II injury to grade III on repeat imaging at post-injury day 7. He suffered a middle cerebral artery stroke after stent placement. He was treated with heparin and discharged to rehabilitation with improving neurologic condition. Another patient suffered a subclavian dissection during attempted carotid stent placement. The patient underwent successful
stenting of the subclavian and carotid arteries. Of the 86 patients with asymptomatic injuries, 69 were treated with antithrombotics. One patient with a grade II VAI, who was appropriately treated with systemic heparin, suffered a stroke. After 2005, one hundred and nine asymptomatic patients were diagnosed with grade II/III BCVI; 53 patients had an isolated CAI, 35 patients had an isolated VAI, and 21 patients had a combined CAI and VAI. Only 2 (2%) patients had stents placed in this group. A 16-year-old with an occluded right CAI (grade IV) and a grade II left CAI suffered a transient ischemic attack when systemically heparinized. Imaging revealed a >80% stenosis of her remaining carotid artery, and she underwent endovascular therapy. After restoration of patency with the carotid stent in this remaining carotid artery, she had resolution of her symptoms. The second patient underwent carotid stenting for a pseudoaneurysm that increased in size from 5 mm to 8 mm. Stroke rate in the stent group vs the no-stent groups was significantly different by Fisher’s exact test; stroke in the stent group was 8.7% vs 0.06% in the no-stent group (p ¼ 0.04). Distribution of BCVI (CAI vs VAI), mean age, and mean Injury Severity Score were not significantly different between the stent and no-stent groups. During the latter study period (after 2005), repeat imaging results were tracked for BCVI progression vs resolution in the 109 patients; 33 (30%) patients did not have repeat imaging of their BCVI. Of the remaining 76 patients, 22 (20%) had no change in their injury, 17 (16%) resolved their injury, and 14 (13%) had improvement. The remaining 23 (21%) patients had progression of their injury: 8 had worsening from grade I to grade III, 5 had an increase in the size of their grade III pseudoaneurysm, 5 progressed from grade II to III, 4 had a worsening grade II injury, and 1 progressed from grade I to grade II. Of the 109 asymptomatic patients, 97 (89%) were treated with antithrombotics; no patient treated with antithrombotics suffered a stroke and there was no rupture of a pseudoaneurysm.
DISCUSSION In patients with grade III blunt CAI, endovascular stents have been advocated from a theoretical standpoint
Vol. 218, No. 5, May 2014
Burlew et al
because of the potential embolic source from stagnant blood within the pseudoaneurysm and fear of subsequent pseudoaneurysm rupture. Initial case reports documented successful endovascular treatment of traumatic cerebrovascular lesions in the blunt trauma patient.11-17 Based on these favorable results, stenting was more widely adopted and, subsequently, case series of individual center’s experience were published. We published one of the initial reports in 2000 demonstrating excellent results in 14 patients managed with endovascular therapy for their CAI pseudoaneurysm.18 Subsequent reports documented effective use of stents for both pseudoaneurysms and grade II narrowing (>25%) of the cerebrovasculature. In 2007, Edwards and colleagues reported use of endovascular treatment in 18 patients with 22 injuries; 4 of the stented injuries were for luminal compromise >70%, and the remainder were for pseudoaneurysms.19 A stroke occurred in 1 patient with an associated traumatic brain injury 5 days after stenting and the patient ultimately died. Concurrent with this publication, a group from South Korea described the successful use of endovascular stents for traumatic vertebral artery lesions in 5 patients, all with pseudoaneurysms.20 In 2008, Berne and colleagues reported their experience with 11 patients undergoing stenting for carotid pseudoaneurysms21; although they concluded carotid artery stenting is safe, it is unclear from their results whether 2 of the BCVI-related strokes were associated with stent placement. In 2009, the Maryland Shock Trauma group suggested they also had good outcomes with endovascular therapy for BCVI, reporting that only 2 of 33 patients sustaining a stroke after stenting for a 6% post-stent stroke rate.22 The Memphis group recently published the largest experience to date of endovascular therapy for BCVI.2 In their 4 ½-year study, stenting was considered for all grade II and III injuries, including both CAI and VAI. In their study, 80 patients underwent stenting; 50 patients had CAI and the remainder underwent stent placement for VAI. In-hospital, post-procedure stroke rate for the endovascular treatment group was 4%. If outpatient follow-up is included, the stroke rate for their endovascular treatment group is closer to 9% long-term. They concluded endovascular therapy results in the lowest stroke rate reported in the literature. In 2005, we issued a cautionary warning on the use of endovascular therapy for BCVI.1 In that report, there were 4 post-stent complications (2 in-hospital strokes, 1 outpatient stroke, and 1 subclavian dissection) and an alarming 45% carotid occlusion rate. We acknowledge that the medical management of patients who have stents placed has fundamentally changed since that report, with the importance of dual anti-platelet therapy for 6 to 12
Stenting and Blunt Cerebrovascular Injuries
1015
months post stent widely accepted. The more fundamental issue, however, remains. Just because we technically can place a stent into an injured carotid or vertebral artery, should we? Three components are fundamental to answer this question. First is stroke rate. Is the stroke rate of patients managed with endovascular therapy comparable, better, or worse than that of patients treated with antithrombotic therapy alone? Second, what is the long-term durability of endovascular stents? Do stented cerebrovascular arteries develop fibromuscular dysplasia and associated stenosis? Will the stent fracture and break? What is the repeat intervention rate? On the other side, what is the risk of flow-limiting luminal narrowing for a grade II injury or of pseudoaneurysm rupture for a grade III injury? To address the first question of stroke rate based on therapeutic intervention, we looked at the available literature. The 3 largest populations from these studies report in-hospital stroke rates of 4% to 5% with endovascular therapy.2,19,22 In this analysis, considering all asymptomatic grade II and III patients as candidates for endovascular therapy, much like the Memphis rationale, 25% of eligible patients underwent stenting, with a 9% stroke rate after stent deployment. Comparing these post-stent stroke rates with the available stroke rates for medical therapy alone, one notes a distinct difference. The lowest reported stroke rate in the literature is 0.3% for patients treated with antithrombotics alone in a large series.4 In that study, of the 307 asymptomatic patients with BCVI that received antithrombotic treatment, 1 patient had stroke (0.3%) and 1 patient a had a transient ischemic attack (0.3%). Others report stroke rate in medically treated injuries of 2.5%.2 Studies by Edwards and colleagues19 and Stein and colleagues22 do not subdivide patients as medical treatment alone vs endovascular therapy plus medical treatment in their reported stroke rates of 9% and 3.7%, respectively. Comparing stroke rates for endovascular therapy (4% to 9% combining the available series) vs antithrombotic therapy alone (0.3% to 2.5%), antithrombotic therapy appears to be as reasonable if not better than stenting for BCVI. Addressing the second questiondwhat is the long-term durability of endovascular stentsdis somewhat more problematic. The majority of stenting in the cerebrovasculature has been done for atherosclerotic disease. Reflective of the underlying cause requiring stenting, this patient population is much older than the multiply injured patient with a BCVI. How these stents will fare 30, 40, 50, or even 70 years down the road when placed in a 20-year-old trauma patient is unknown. Recent analysis comparing carotid artery stenting with endarterectomy, however, demonstrated stenting is associated with increased risks of long-term stroke and restenosis.23
1016
Burlew et al
Stenting and Blunt Cerebrovascular Injuries
Postulated mechanisms of restenosis after stent deployment include both axial and circumferential stress on the artery, which promotes arterial wall damage, cell migratory and proliferative responses, followed by platelet deposition and stenosis, in addition to geometric changes in the artery after stenting.24 Although endovascular therapy might still have a role in a subset of patients with vascular disease,25 criteria for stent placement continue to evolve. Finally, to address the third questiondWhat is the risk of pseudoaneurysm rupture? There are reported cases of intracranial pseudoaneurysm rupture, often in the petrous portion of the carotid canal, leading to epistaxis or subarachnoid hemorrhage26-30; these intracranial pseudoaneurysms are reported after blunt trauma, but also due to atherosclerosis, iatrogenic injury during mastoid surgery, chronic middle ear infections, congenital causes, and radiation injury. There is one report of a blunt traumarelated, extracranial, common carotid pseudoaneurysm measuring 3 3 4 cm in size that was managed using a carotid stent after rupture.17 One might question whether this case was, in fact, surgically accessible and should have been managed operatively at the time of diagnosis. With the limited information on the long-term sequela of these injuries, it is difficult to assess the true, although likely low, risk of rupture. This study has its limitations. This was not a randomized controlled trial. As such, the decision to stent a patient’s BCVI was at the discretion of the trauma attending. This is particularly notable in the second phase of the study, after 2005. Why was 1 patient with a worsening pseudoaneurysm stented and the other 22 patients with worsening injuries (4 grade II and 5 grade III BCVI) managed with medical therapy alone? Although enlarging pseudoaneurysms appears to be a viable intervention point for endovascular therapy, what should we consider “too big” for a pseudoaneurysm or “too narrow” for a dissection when evaluating these patients for stenting? Crescendo transient ischemic attacks in a patient who ultimately underwent stent placement with resolution of her symptoms suggest there are indications for selective stenting. However, should patients who have a completed stroke undergo stent placement? As is true of many controversies, this study has raised questions and elucidated current management options.
CONCLUSIONS The diagnosis and treatment of BCVI continues to evolve. With BCVI diagnosed in >1% of blunt trauma patients and the recognition of a clinically silent period, diagnosis of these injuries should be aggressively pursued. Stroke in asymptomatic patients can almost universally be
J Am Coll Surg
avoided with antithrombotic treatment. Routine stenting entails increased costs and potential risk for stroke, and does not appear to add benefit. Intravascular stents should be reserved for the rare patient with symptomatology due to narrowing or a markedly enlarging pseudoaneurysm. Author Contributions Study conception and design: Burlew Acquisition of data: Burlew, Biffl Analysis and interpretation of data: Burlew, Biffl, Moore, Pieracci Drafting of manuscript: Burlew Critical revision: Biffl, Moore, Pieracci, Beauchamp, Stovall, Wagenaar, Jurkovich REFERENCES 1. Cothren CC, Moore EE, Ray CE, et al. Carotid artery stents for blunt cerebrovascular injury: risks exceed benefits. Arch Surg 2005;140:480e486. 2. DiCocco JM, Fabian TC, Emmett KP, et al. Optimal outcomes for patients with blunt cerebrovascular injury (BCVI): tailoring treatment to the lesion. J Am Coll Surg 2011;212:549e559. 3. National Institutes of Health, National Institute of Neurological Disorders and Stroke. Clinical Alert 4/11/11. Available at: www.ninds.nih.gov. Accessed October 8, 2013. 4. Burlew CC, Biffl WL, Moore EE, et al. Blunt cerebrovascular injuries: redefining screening criteria in the era of non-invasive diagnosis. J Trauma 2012;72:330e337. 5. Cothren CC, Moore EE, Johnson JL, et al. Cervical spine fracture patterns mandating angiography to rule out blunt cerebrovascular injury. Surgery 2007;141:76e82. 6. Burlew CC, Biffl WL, Moore EE, et al. Blunt cerebrovascular injuries in children: broadened screening guidelines are warranted. J Trauma Acute Care Surg 2012;72:1120e1121. 7. Biffl WL, Moore EE, Offner PJ, et al. Blunt carotid arterial injuries: implications of a new grading scale. J Trauma 1999;47:845e853. 8. Cothren CC, Biffl WL, Moore EE, et al. Treatment for blunt cerebrovascular injuries: equivalence of anticoagulation and antiplatelet agents. Arch Surg 2009;44:685e690. 9. Biffl WL, Moore EE, Ryu RK, et al. The unrecognized epidemic of blunt carotid arterial injuries: early diagnosis improves neurologic outcome. Ann Surg 1998;228:462e470. 10. Biffl WL, Cothren CC, Moore EE, et al. Western Trauma Association critical decisions in trauma: screening for and treatment of blunt cerebrovascular injuries. J Trauma 2009;67:1150e1153. 11. Duane TM, Parker F, Stokes GK, et al. Endovascular carotid stenting after trauma. J Trauma 2002;52:149e153. 12. Kerby JD, May AK, Gomez CR, Rue LW 3rd. Treatment of bilateral blunt carotid injury using percutaneous angioplasty and stenting: case report and review of the literature. J Trauma 2000;49:784e787. 13. Shames ML, Davis JW, Evans AJ. Endoluminal stent placement for the treatment of traumatic carotid artery pseudoaneurysm: case report and review of the literature. J Trauma 1999;46:724e726. 14. Fanelli F, Salvatori FM, Ferrari R, et al. Stent repair of bilateral post-traumatic dissections of the internal carotid artery. J Endovasc Ther 2004;11:517e521.
Vol. 218, No. 5, May 2014
Burlew et al
15. Fusonie GE, Edwards JD, Reed AB. Covered stent exclusion of blunt traumatic carotid artery pseudoaneurysm: case report and review of the literature. Ann Vasc Surg 2004;18:376e379. 16. Malek AM, Halbach VV, Phatouros CC, et al. Endovascular treatment of a ruptured intracranial dissecting vertebral aneurysm in a kickboxer. J Trauma 2000;48:143e145. 17. Akiyama Y, Nakahara I, Tanaka M, et al. Urgent endovascular stent-graft placement for a ruptured traumatic pseudoaneurysm of the extracranial carotid artery. J Trauma 2005;58:624e627. 18. Coldwell DM, Novak Z, Ryu RK, et al. Treatment of posttraumatic internal carotid arterial pseudoaneurysms with endovascular stents. J Trauma 2000;48:470e472. 19. Edwards NM, Fabian TC, Claridge JA, et al. Antithrombotic therapy and endovascular stents are effective treatment for blunt carotid injuries: results from long-term follow-up. J Am Coll Surg 2007;5:1007e1014. 20. Lee YJ, Ahn JY, Han IB, et al. Therapeutic endovascular treatments for traumatic vertebral artery injuries. J Trauma 2007; 62:886e891. 21. Berne JD, Reuland KR, Villarreal DH, et al. Internal carotid artery stenting for blunt carotid artery injuries with an associated pseudoaneurysm. J Trauma 2008;64:398e405. 22. Stein DM, Boswell S, Sliker CW, et al. Blunt cerebrovascular injuries: does treatment always matter? J Trauma 2009;66: 132e143. 23. Arya S, Pipinos II, Garg N, et al. Carotid endarterectomy is superior to carotid angioplasty and stenting for perioperative
24. 25. 26.
27.
28.
29. 30.
Stenting and Blunt Cerebrovascular Injuries
1017
and long-term results. Vasc Endovascular Surg 2011;45: 490e498. Kamenskiy AV, Pipinos II, Dzenis YA, et al. Effects of carotid artery stenting on arterial geometry. J Am Coll Surg 2013;217: 251e262. Brott TG, Hobson RW 2nd, Howard G, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363:11e23. Han MH, Sung MW, Chang KH, et al. Traumatic pseudoaneurysm of the intracavernous ICA presenting with massive epistaxis: imaging diagnosis and endovascular treatment. Laryngoscope 1994;104[Pt 1]:370e377. Chen D, Concus AP, Halbach VV, Cheung SW. Epistaxis originating from traumatic pseudoaneurysm of the internal carotid artery: diagnosis and endovascular therapy. Laryngoscope 1998;108:326e331. Willinsky R, Lasjaunias P, Pruvost P, Boucherat M. Petrous internal carotid aneurysm causing epistaxis: balloon embolization with preservation of the parent vessel. Neuroradiology 1987;29:570e572. Costantino PD, Russell E, Reisch D, et al. Ruptured petrous carotid aneurysm presenting with otorrhagia and epistaxis. Am J Otol 1991;12:378e383. Auyeung KM, Lui WM, Chow LC, Chan FL. Massive epistaxis related to petrous carotid artery pseudoaneurysm after radiation therapy: emergency treatment with covered stent in two cases. AJNR Am J Neuroradiol 2003;24:1449e1452.
BACKGROUND:
The role of endovascular stenting for blunt cerebrovascular injuries (BCVIs) continues to be debated in the literature. Despite our cautionary report in 2005, which highlights the risks associated with carotid artery stenting for BCVI,1 recent reports continue to advocate endovascular therapy for high-grade injuries.2 In 2011, the National Institutes of Health released a clinical alert reporting the halted trial of intracranial stenting in favor of medical therapy alone for high-grade carotid stenoses.3 That trial’s Data Safety Monitoring Board reported a 14% stroke/death rate with carotid stenting and medical
Disclosure Information: Nothing to disclose. Presented at the American College of Surgeons 99th Annual Clinical Congress, Washington, DC, October 2013. Received October 1, 2013; Revised January 21, 2014; Accepted January 22, 2014. From the Department of Surgery, Denver Health Medical Center and the University of Colorado School of Medicine, Denver, CO. Correspondence address: Clay Cothren Burlew, MD, FACS, Department of Surgery, Denver Health Medical Center and University of Colorado School of Medicine, 777 Bannock St, MC 0206, Denver, CO 80204. email: [email protected]
ª 2014 by the American College of Surgeons Published by Elsevier Inc.
therapy vs a 5.8% stroke/death rate in the medical treatment alone arm. In light of this information, the empiric use of vascular stents to manage BCVI warrants critical reassessment. The purpose of this study was to determine if antithrombotic therapy, rather than stenting, was effective in post-injury patients with high-grade vascular dissections (Denver grade II BCVI) and pseudoaneurysms (Denver grade III BCVI).
METHODS Denver Health Medical Center is a state-certified and American College of Surgeonsverified level I regional trauma center, and integral teaching facility of the University of Colorado Denver. In January 1996, we initiated comprehensive screening for BCVI based on injury mechanism, injury patterns, or symptoms. Screening criteria have been modified based on subsequent analyses, yielding our current management guideline (Fig. 1).4-6 The decision to screen a patient for BCVI outside the noted guidelines was at the discretion of the trauma attending. We currently use multi-slice (64-slice) CT angiography as our standard screening test for patients at risk for
1012
ISSN 1072-7515/14/$36.00 http://dx.doi.org/10.1016/j.jamcollsurg.2014.01.042
Vol. 218, No. 5, May 2014
Burlew et al
Stenting and Blunt Cerebrovascular Injuries
1013
Figure 1. Current screening criteria for blunt cerebrovascular injuries at Denver Health Medical Center. BCVI, blunt cerebrovascular injuries.
BCVI. During daytime hours, a neuroradiologist reads the CT angiography; at night, an attending radiologist reviews images in real time, with an over-read in the morning by a neuroradiologist. Digital subtraction 4-vessel angiography was used to evaluate patients for the first 10 years of the study period. Patients’ injuries were classified according to the Denver grading scale7 (Table 1). Before 2005, only patients with pseudoaneurysms were considered for endovascular stenting. After our 2005 report on the risks of endovascular therapy for BCVI, a virtual moratorium was Table 1. Denver Grading Scale for Blunt Cerebrovascular Injuries Grade I: irregularity of the vessel wall or a dissection/intramural hematoma with 25%) and grade III (pseudoaneurysms) injuries were included in the analysis due to reports advocating stenting for these injuries.2 To date, endovascular stenting for minimal luminal narrowing (grade I injury) or complete occlusion (grade IV injury) has not been advocated, so these patients were not included in the analysis. BCVI grades, treatment, indication for stenting, and inpatient outcomes during the study period were analyzed. Patients who presented with ischemic/stroke symptoms were excluded. Statistical analysis was performed using SAS software for Windows (SAS Institute). The Colorado Multi-Institutional Review Board, our University-based IRB, approved this study.
RESULTS During the 15-year study period, 195 patients were diagnosed with grade II or III BCVIs. The majority was men (67%), with a mean age of 39 years (range 19 months to 83 years). Before 2005, there were 86 asymptomatic patients who sustained a grade II/III BCVI; 39 patients had an isolated carotid artery injury (CAI), 32 patients had an isolated vertebral artery injury (VAI), and 15 patients had a combined CAI and VAI. Of the 86 asymptomatic patients, 21 (25%) patients underwent stent placement (Table 2). Endovascular therapy included 18 carotid stents and 3 vertebral stents. Two patients suffered a stroke due to their stent placement. The first was a 49-year-old man with a grade III CAI. Either during or immediately after stent placement, dense hemiplegia developed in the patient. Heparin was started after the major stroke; several days later the patient had an intracranial hemorrhage requiring craniotomy and evacuation. He was discharged to our rehabilitation unit; although he had overall improvement in his neurologic status, upper-extremity motor deficits persisted. The second patient was a 38-year-old man with a right CAI that had progressed from a grade II injury to grade III on repeat imaging at post-injury day 7. He suffered a middle cerebral artery stroke after stent placement. He was treated with heparin and discharged to rehabilitation with improving neurologic condition. Another patient suffered a subclavian dissection during attempted carotid stent placement. The patient underwent successful
stenting of the subclavian and carotid arteries. Of the 86 patients with asymptomatic injuries, 69 were treated with antithrombotics. One patient with a grade II VAI, who was appropriately treated with systemic heparin, suffered a stroke. After 2005, one hundred and nine asymptomatic patients were diagnosed with grade II/III BCVI; 53 patients had an isolated CAI, 35 patients had an isolated VAI, and 21 patients had a combined CAI and VAI. Only 2 (2%) patients had stents placed in this group. A 16-year-old with an occluded right CAI (grade IV) and a grade II left CAI suffered a transient ischemic attack when systemically heparinized. Imaging revealed a >80% stenosis of her remaining carotid artery, and she underwent endovascular therapy. After restoration of patency with the carotid stent in this remaining carotid artery, she had resolution of her symptoms. The second patient underwent carotid stenting for a pseudoaneurysm that increased in size from 5 mm to 8 mm. Stroke rate in the stent group vs the no-stent groups was significantly different by Fisher’s exact test; stroke in the stent group was 8.7% vs 0.06% in the no-stent group (p ¼ 0.04). Distribution of BCVI (CAI vs VAI), mean age, and mean Injury Severity Score were not significantly different between the stent and no-stent groups. During the latter study period (after 2005), repeat imaging results were tracked for BCVI progression vs resolution in the 109 patients; 33 (30%) patients did not have repeat imaging of their BCVI. Of the remaining 76 patients, 22 (20%) had no change in their injury, 17 (16%) resolved their injury, and 14 (13%) had improvement. The remaining 23 (21%) patients had progression of their injury: 8 had worsening from grade I to grade III, 5 had an increase in the size of their grade III pseudoaneurysm, 5 progressed from grade II to III, 4 had a worsening grade II injury, and 1 progressed from grade I to grade II. Of the 109 asymptomatic patients, 97 (89%) were treated with antithrombotics; no patient treated with antithrombotics suffered a stroke and there was no rupture of a pseudoaneurysm.
DISCUSSION In patients with grade III blunt CAI, endovascular stents have been advocated from a theoretical standpoint
Vol. 218, No. 5, May 2014
Burlew et al
because of the potential embolic source from stagnant blood within the pseudoaneurysm and fear of subsequent pseudoaneurysm rupture. Initial case reports documented successful endovascular treatment of traumatic cerebrovascular lesions in the blunt trauma patient.11-17 Based on these favorable results, stenting was more widely adopted and, subsequently, case series of individual center’s experience were published. We published one of the initial reports in 2000 demonstrating excellent results in 14 patients managed with endovascular therapy for their CAI pseudoaneurysm.18 Subsequent reports documented effective use of stents for both pseudoaneurysms and grade II narrowing (>25%) of the cerebrovasculature. In 2007, Edwards and colleagues reported use of endovascular treatment in 18 patients with 22 injuries; 4 of the stented injuries were for luminal compromise >70%, and the remainder were for pseudoaneurysms.19 A stroke occurred in 1 patient with an associated traumatic brain injury 5 days after stenting and the patient ultimately died. Concurrent with this publication, a group from South Korea described the successful use of endovascular stents for traumatic vertebral artery lesions in 5 patients, all with pseudoaneurysms.20 In 2008, Berne and colleagues reported their experience with 11 patients undergoing stenting for carotid pseudoaneurysms21; although they concluded carotid artery stenting is safe, it is unclear from their results whether 2 of the BCVI-related strokes were associated with stent placement. In 2009, the Maryland Shock Trauma group suggested they also had good outcomes with endovascular therapy for BCVI, reporting that only 2 of 33 patients sustaining a stroke after stenting for a 6% post-stent stroke rate.22 The Memphis group recently published the largest experience to date of endovascular therapy for BCVI.2 In their 4 ½-year study, stenting was considered for all grade II and III injuries, including both CAI and VAI. In their study, 80 patients underwent stenting; 50 patients had CAI and the remainder underwent stent placement for VAI. In-hospital, post-procedure stroke rate for the endovascular treatment group was 4%. If outpatient follow-up is included, the stroke rate for their endovascular treatment group is closer to 9% long-term. They concluded endovascular therapy results in the lowest stroke rate reported in the literature. In 2005, we issued a cautionary warning on the use of endovascular therapy for BCVI.1 In that report, there were 4 post-stent complications (2 in-hospital strokes, 1 outpatient stroke, and 1 subclavian dissection) and an alarming 45% carotid occlusion rate. We acknowledge that the medical management of patients who have stents placed has fundamentally changed since that report, with the importance of dual anti-platelet therapy for 6 to 12
Stenting and Blunt Cerebrovascular Injuries
1015
months post stent widely accepted. The more fundamental issue, however, remains. Just because we technically can place a stent into an injured carotid or vertebral artery, should we? Three components are fundamental to answer this question. First is stroke rate. Is the stroke rate of patients managed with endovascular therapy comparable, better, or worse than that of patients treated with antithrombotic therapy alone? Second, what is the long-term durability of endovascular stents? Do stented cerebrovascular arteries develop fibromuscular dysplasia and associated stenosis? Will the stent fracture and break? What is the repeat intervention rate? On the other side, what is the risk of flow-limiting luminal narrowing for a grade II injury or of pseudoaneurysm rupture for a grade III injury? To address the first question of stroke rate based on therapeutic intervention, we looked at the available literature. The 3 largest populations from these studies report in-hospital stroke rates of 4% to 5% with endovascular therapy.2,19,22 In this analysis, considering all asymptomatic grade II and III patients as candidates for endovascular therapy, much like the Memphis rationale, 25% of eligible patients underwent stenting, with a 9% stroke rate after stent deployment. Comparing these post-stent stroke rates with the available stroke rates for medical therapy alone, one notes a distinct difference. The lowest reported stroke rate in the literature is 0.3% for patients treated with antithrombotics alone in a large series.4 In that study, of the 307 asymptomatic patients with BCVI that received antithrombotic treatment, 1 patient had stroke (0.3%) and 1 patient a had a transient ischemic attack (0.3%). Others report stroke rate in medically treated injuries of 2.5%.2 Studies by Edwards and colleagues19 and Stein and colleagues22 do not subdivide patients as medical treatment alone vs endovascular therapy plus medical treatment in their reported stroke rates of 9% and 3.7%, respectively. Comparing stroke rates for endovascular therapy (4% to 9% combining the available series) vs antithrombotic therapy alone (0.3% to 2.5%), antithrombotic therapy appears to be as reasonable if not better than stenting for BCVI. Addressing the second questiondwhat is the long-term durability of endovascular stentsdis somewhat more problematic. The majority of stenting in the cerebrovasculature has been done for atherosclerotic disease. Reflective of the underlying cause requiring stenting, this patient population is much older than the multiply injured patient with a BCVI. How these stents will fare 30, 40, 50, or even 70 years down the road when placed in a 20-year-old trauma patient is unknown. Recent analysis comparing carotid artery stenting with endarterectomy, however, demonstrated stenting is associated with increased risks of long-term stroke and restenosis.23
1016
Burlew et al
Stenting and Blunt Cerebrovascular Injuries
Postulated mechanisms of restenosis after stent deployment include both axial and circumferential stress on the artery, which promotes arterial wall damage, cell migratory and proliferative responses, followed by platelet deposition and stenosis, in addition to geometric changes in the artery after stenting.24 Although endovascular therapy might still have a role in a subset of patients with vascular disease,25 criteria for stent placement continue to evolve. Finally, to address the third questiondWhat is the risk of pseudoaneurysm rupture? There are reported cases of intracranial pseudoaneurysm rupture, often in the petrous portion of the carotid canal, leading to epistaxis or subarachnoid hemorrhage26-30; these intracranial pseudoaneurysms are reported after blunt trauma, but also due to atherosclerosis, iatrogenic injury during mastoid surgery, chronic middle ear infections, congenital causes, and radiation injury. There is one report of a blunt traumarelated, extracranial, common carotid pseudoaneurysm measuring 3 3 4 cm in size that was managed using a carotid stent after rupture.17 One might question whether this case was, in fact, surgically accessible and should have been managed operatively at the time of diagnosis. With the limited information on the long-term sequela of these injuries, it is difficult to assess the true, although likely low, risk of rupture. This study has its limitations. This was not a randomized controlled trial. As such, the decision to stent a patient’s BCVI was at the discretion of the trauma attending. This is particularly notable in the second phase of the study, after 2005. Why was 1 patient with a worsening pseudoaneurysm stented and the other 22 patients with worsening injuries (4 grade II and 5 grade III BCVI) managed with medical therapy alone? Although enlarging pseudoaneurysms appears to be a viable intervention point for endovascular therapy, what should we consider “too big” for a pseudoaneurysm or “too narrow” for a dissection when evaluating these patients for stenting? Crescendo transient ischemic attacks in a patient who ultimately underwent stent placement with resolution of her symptoms suggest there are indications for selective stenting. However, should patients who have a completed stroke undergo stent placement? As is true of many controversies, this study has raised questions and elucidated current management options.
CONCLUSIONS The diagnosis and treatment of BCVI continues to evolve. With BCVI diagnosed in >1% of blunt trauma patients and the recognition of a clinically silent period, diagnosis of these injuries should be aggressively pursued. Stroke in asymptomatic patients can almost universally be
J Am Coll Surg
avoided with antithrombotic treatment. Routine stenting entails increased costs and potential risk for stroke, and does not appear to add benefit. Intravascular stents should be reserved for the rare patient with symptomatology due to narrowing or a markedly enlarging pseudoaneurysm. Author Contributions Study conception and design: Burlew Acquisition of data: Burlew, Biffl Analysis and interpretation of data: Burlew, Biffl, Moore, Pieracci Drafting of manuscript: Burlew Critical revision: Biffl, Moore, Pieracci, Beauchamp, Stovall, Wagenaar, Jurkovich REFERENCES 1. Cothren CC, Moore EE, Ray CE, et al. Carotid artery stents for blunt cerebrovascular injury: risks exceed benefits. Arch Surg 2005;140:480e486. 2. DiCocco JM, Fabian TC, Emmett KP, et al. Optimal outcomes for patients with blunt cerebrovascular injury (BCVI): tailoring treatment to the lesion. J Am Coll Surg 2011;212:549e559. 3. National Institutes of Health, National Institute of Neurological Disorders and Stroke. Clinical Alert 4/11/11. Available at: www.ninds.nih.gov. Accessed October 8, 2013. 4. Burlew CC, Biffl WL, Moore EE, et al. Blunt cerebrovascular injuries: redefining screening criteria in the era of non-invasive diagnosis. J Trauma 2012;72:330e337. 5. Cothren CC, Moore EE, Johnson JL, et al. Cervical spine fracture patterns mandating angiography to rule out blunt cerebrovascular injury. Surgery 2007;141:76e82. 6. Burlew CC, Biffl WL, Moore EE, et al. Blunt cerebrovascular injuries in children: broadened screening guidelines are warranted. J Trauma Acute Care Surg 2012;72:1120e1121. 7. Biffl WL, Moore EE, Offner PJ, et al. Blunt carotid arterial injuries: implications of a new grading scale. J Trauma 1999;47:845e853. 8. Cothren CC, Biffl WL, Moore EE, et al. Treatment for blunt cerebrovascular injuries: equivalence of anticoagulation and antiplatelet agents. Arch Surg 2009;44:685e690. 9. Biffl WL, Moore EE, Ryu RK, et al. The unrecognized epidemic of blunt carotid arterial injuries: early diagnosis improves neurologic outcome. Ann Surg 1998;228:462e470. 10. Biffl WL, Cothren CC, Moore EE, et al. Western Trauma Association critical decisions in trauma: screening for and treatment of blunt cerebrovascular injuries. J Trauma 2009;67:1150e1153. 11. Duane TM, Parker F, Stokes GK, et al. Endovascular carotid stenting after trauma. J Trauma 2002;52:149e153. 12. Kerby JD, May AK, Gomez CR, Rue LW 3rd. Treatment of bilateral blunt carotid injury using percutaneous angioplasty and stenting: case report and review of the literature. J Trauma 2000;49:784e787. 13. Shames ML, Davis JW, Evans AJ. Endoluminal stent placement for the treatment of traumatic carotid artery pseudoaneurysm: case report and review of the literature. J Trauma 1999;46:724e726. 14. Fanelli F, Salvatori FM, Ferrari R, et al. Stent repair of bilateral post-traumatic dissections of the internal carotid artery. J Endovasc Ther 2004;11:517e521.
Vol. 218, No. 5, May 2014
Burlew et al
15. Fusonie GE, Edwards JD, Reed AB. Covered stent exclusion of blunt traumatic carotid artery pseudoaneurysm: case report and review of the literature. Ann Vasc Surg 2004;18:376e379. 16. Malek AM, Halbach VV, Phatouros CC, et al. Endovascular treatment of a ruptured intracranial dissecting vertebral aneurysm in a kickboxer. J Trauma 2000;48:143e145. 17. Akiyama Y, Nakahara I, Tanaka M, et al. Urgent endovascular stent-graft placement for a ruptured traumatic pseudoaneurysm of the extracranial carotid artery. J Trauma 2005;58:624e627. 18. Coldwell DM, Novak Z, Ryu RK, et al. Treatment of posttraumatic internal carotid arterial pseudoaneurysms with endovascular stents. J Trauma 2000;48:470e472. 19. Edwards NM, Fabian TC, Claridge JA, et al. Antithrombotic therapy and endovascular stents are effective treatment for blunt carotid injuries: results from long-term follow-up. J Am Coll Surg 2007;5:1007e1014. 20. Lee YJ, Ahn JY, Han IB, et al. Therapeutic endovascular treatments for traumatic vertebral artery injuries. J Trauma 2007; 62:886e891. 21. Berne JD, Reuland KR, Villarreal DH, et al. Internal carotid artery stenting for blunt carotid artery injuries with an associated pseudoaneurysm. J Trauma 2008;64:398e405. 22. Stein DM, Boswell S, Sliker CW, et al. Blunt cerebrovascular injuries: does treatment always matter? J Trauma 2009;66: 132e143. 23. Arya S, Pipinos II, Garg N, et al. Carotid endarterectomy is superior to carotid angioplasty and stenting for perioperative
24. 25. 26.
27.
28.
29. 30.
Stenting and Blunt Cerebrovascular Injuries
1017
and long-term results. Vasc Endovascular Surg 2011;45: 490e498. Kamenskiy AV, Pipinos II, Dzenis YA, et al. Effects of carotid artery stenting on arterial geometry. J Am Coll Surg 2013;217: 251e262. Brott TG, Hobson RW 2nd, Howard G, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363:11e23. Han MH, Sung MW, Chang KH, et al. Traumatic pseudoaneurysm of the intracavernous ICA presenting with massive epistaxis: imaging diagnosis and endovascular treatment. Laryngoscope 1994;104[Pt 1]:370e377. Chen D, Concus AP, Halbach VV, Cheung SW. Epistaxis originating from traumatic pseudoaneurysm of the internal carotid artery: diagnosis and endovascular therapy. Laryngoscope 1998;108:326e331. Willinsky R, Lasjaunias P, Pruvost P, Boucherat M. Petrous internal carotid aneurysm causing epistaxis: balloon embolization with preservation of the parent vessel. Neuroradiology 1987;29:570e572. Costantino PD, Russell E, Reisch D, et al. Ruptured petrous carotid aneurysm presenting with otorrhagia and epistaxis. Am J Otol 1991;12:378e383. Auyeung KM, Lui WM, Chow LC, Chan FL. Massive epistaxis related to petrous carotid artery pseudoaneurysm after radiation therapy: emergency treatment with covered stent in two cases. AJNR Am J Neuroradiol 2003;24:1449e1452.
