Of 66 grade I BCI, 41 cases (62%) were reimaged (Table 2). Grade I BCI healed in 70% of cases with heparin, and 63% healed without heparin; 26% in each treatment group were noted to persist. One grade I BCI (4%) progressed to pseudoaneurysm (grade III) formation despite heparin; in the subgroup without heparin, one case (6%) progressed to grade II and one progressed to grade III. There were no statistical differences between the treatment groups in terms of arteriographic outcome of grade I injuries.
Ten of 18 grade II BCI (56%) were studied by follow-up arteriography. All 10 of the lesions had been treated with heparin, so no comparison could be made with an untreated group. One of the lesions (10%) healed, one improved, and one persisted; seven injuries (70%) worsened: six cases formed pseudoaneurysms (grade III), and one case progressed to occlusion (grade IV).
Sixteen grade III injuries were studied by follow-up arteriography. Three were not treated, and all were found to persist. One of those treated with heparin (8%) healed; 11 cases (85%) persisted, and one case progressed to occlusion. Eight other lesions began as grade I or II and had progressed to grade III at follow-up; none of these cases healed with further anticoagulation, so the actual healing rate of grade III injuries was 4% (1 of 24). Endovascular stents were deployed to treat 18 of the persistent pseudoaneurysms. This method led to healing in 16 of the 18 lesions (89%). Two of the stents became occluded, with one resultant stroke.
Two of the grade IV injuries were reimaged; neither had changed. One of the patients was receiving heparin, the other was not. All four patients with grade V injuries succumbed to massive brain injury or hemorrhage within 12 hours of the event, and, thus, none had follow-up arteriography.
BCI Neurologic Outcome
The mean Glasgow Coma Scale (GCS) score on admission was 8.8 ± 0.6. Thirty-five patients (46%) had a GCS score less than or equal to 6, and 20 patients (26%) had GCS score equal to 3 on arrival. Severe head injuries were distributed fairly evenly across injury grades (Table 6). Consequently, no difference in overall neurologic outcome between the different injury grades could be found. In addition to the incidence of severe head injury (GCS score ≤ 6), Table 6 lists mortality and stroke rates associated with each injury grade. The highest injury grade recorded for an individual patient was used for this analysis (i.e., if bilateral injuries were present, or a lesion had worsened on follow-up arteriography, the highest injury grade for that patient was recorded). There was no difference in the presence of severe head injury or mortality rate among the different BCI grades, except that grade V BCI were associated with higher mortality than other grades. However, stroke rate increased with grade. The overall mortality in this series was 17% (13 deaths); 20 of the survivors (32%) suffered permanent severe neurologic disability.
BCI are associated with significant morbidity and mortality. 1–7 Unfortunately, a dearth of experience, even at busy trauma centers, has left a void in the literature, specifically, there are essentially no class I data on which to base BCI management decisions. Fabian and colleagues, 9 identifying a significantly higher incidence of BCI than previously reported, suggested that a higher index of suspicion be applied to the diagnosis of BCI, and concluded, “Liberal screening, leading to earlier diagnosis, would improve outcome.” Having recognized that at least half of the BCI at our institution were unsuspected on clinical grounds, 11 we initiated a screening protocol that considered mechanism of injury and injury patterns in addition to high-risk physical and radiologic findings. As a result, we created a veritable epidemic of BCI. 10 Although our experience has taught us many things about these injuries, one of the major questions that has remained unanswered is, what is the optimal management of BCI? Although Fabian and colleagues 9 demonstrated improved survival and neurologic outcome related to the treatment of BCI with of heparin, they and others 4,8 have suggested that discrete BCI types may behave differently and, therefore, may warrant individualized treatment. The absence of a formal BCI grading scale has been a major impediment to formulating sound practice guidelines. We hypothesized that different injury grades had distinct implications in terms of response to therapy and ultimate neurologic outcome. We reviewed the available literature and our experience over the past 9 years, and formulated a grading scale for BCI that offers prognostic and therapeutic implications.
Our grading scale broadly categorizes lesions as nonhemodynamically significant intimal injuries (grade I), potentially hemodynamically significant dissections and hematomas (grade II), pseudoaneurysms (grade III), occlusions (grade IV), and vessel transections (grade V). The use and improvement of this grading scale require careful anatomic description of the arterial lesions. Unfortunately, we have found noninvasive imaging modalities such as duplex ultrasonography 4 and computed tomographic angiography 12 to be inadequate in this regard. Magnetic resonance angiography holds promise in this area, 13 but until it has been rigorously evaluated, cerebral arteriography remains the diagnostic gold standard.
Autopsy data have been reviewed in an attempt to corroborate our arteriographic findings. Of the 13 patients who died, autopsy data were available on 10 patients. Four of the patients had arteriographic grade IV or V injuries, and these findings were confirmed at autopsy. Similarly, one patient with a grade II injury by arteriography had arterial disruption with dissection confirmed at autopsy. Five patients with arteriographic grade I injuries had autopsies. In two cases, there was no pathologic examination of the cervical ICA to delineate the injury, although one patient did have evidence of massive cerebral ischemic infarction ipsilateral to the arteriographic injury. Of the other three with grade I injuries by arteriography, one patient had free-floating thrombus within the vessel lumen, one patient had arterial occlusion secondary to extramural hemorrhage and compression, and one patient had an intramural hematoma. These latter three discrepancies between autopsy and arteriographic findings may represent progression of the injuries between BCI diagnosis and death but illustrate the potential for injuries to be worse than they appear on arteriography.
Grade I injuries seem to result from either stretching of the artery or a direct blow. Disruption of the intima exposes the thrombogenic subendothelial collagen, inviting platelet aggregation. There may not be sufficient thrombus to create a filling defect on the arteriogram, so the only visible abnormality on arteriography may be intimal irregularity or accompanying vasospasm. Clinically, continued platelet aggregation may result in partial or complete vessel thrombosis, or thromboembolization and stroke. It is likely that emboli pose a much greater threat to the brain than occlusion. Although only 20% of individuals have an intact and symmetric Circle of Willis, 14 single-vessel occlusion can be well tolerated in the absence of atherosclerotic disease. 15,16 This may partly explain the previous underrecognition of BCI, as focal deficits did not necessarily result from occlusive carotid arterial injuries. On the other hand, cerebral thromboembolism invariably results in cerebral ischemia, with functional consequences dependent solely on the affected area of brain and largely unaffected by collateral circulation.
In our series, 7% of grade I BCI progressed to higher grade lesions during the observation period. Although infrequent, the progression of these injuries was unpredictable. In addition, there was a 3% stroke risk associated with grade I lesions, suggesting that they should not be ignored. We could not identify a difference in arteriographic outcome of grade I BCI as related to treatment. However, it is important to recognize that this investigation was not a randomized study. Given the pathogenesis of these lesions, it makes intuitive sense that antiplatelet therapy would be advantageous to prevent ongoing platelet aggregation at sites of intimal disruption. Whether systemic heparin is necessary to achieve this protection is unknown. Antiplatelet agents, in particular the glycoprotein IIb-IIIa blockers (e.g., abciximab, eptifibatide, and tirofiban) and ADP receptor antagonists (e.g., clopidogrel), seem to offer reductions in thromboembolic events, while maintaining a favorable safety profile with respect to bleeding complications. 17,18 These characteristics are particularly desirable in a multisystem trauma patient. We would caution against the premature adoption of these agents, however. Although two recent reports 19,20 have concluded that no heparin, delayed heparin, or antiplatelet therapy are all viable alternatives to systemic heparinization, their conclusions were based on retrospective data in relatively small numbers of patients, without regard to BCI injury grade. We are presently enrolling patients in a prospective study in which full systemic heparinization is being compared with antiplatelet therapy in the treatment of grade I BCI.
In contrast to the relatively innocuous-appearing grade I injuries, grade II BCI appear to be higher risk lesions. Indeed, their predilection to progress to pseudoaneurysm formation or thrombotic occlusion (70% in our experience) support this clinical suspicion. We identified the visualization of a raised intimal flap, intraluminal thrombus, or a dissection with luminal stenosis as indicative of this increased risk. The stenotic lesions may potentially be subdivided into two groups: dissections and intramural hematomas. Although evident at autopsy, the difference is difficult to discern arteriographically, and, thus, no distinction is made in our grading scale. Either lesion may be potentially hemodynamically significant. Although a 50% reduction in diameter is generally believed to represent a hemodynamically significant lesion, we selected 25% stenosis as our threshold for two reasons. First, the diagnostic arteriogram represents a sampling of an evolving acute process, i.e., the fact that enough thrombus has accumulated to narrow the diameter by 25% suggests the potential to progress to a higher grade stenosis, or to embolize and cause stroke. Second, our intent was to limit grade I injuries to those without significant amounts of thrombus formation; the only narrowing associated with grade I injuries should be that attributable to vasospasm. The presence of a raised intimal flap allows the egress of a column of blood into the subintimal plane. This process further lifts the flap, narrowing the vessel lumen, and threatening its complete obliteration. Similarly, an intramural hematoma may narrow the lumen to the point of occlusion. In the case of dissections, there is the additional threat of platelet aggregation and thromboembolization. Intraluminal thrombus obviously indicates that this has already occurred. For these reasons, we believe grade II BCI put the patient at higher risk for adverse events, and, thus, we have had a lower threshold to administer anticoagulants to the patients despite multisystem trauma. Despite this aggressive stance, 70% of grade II injuries progressed to pseudoaneurysm formation or vessel occlusion while on heparin therapy, and only 10% healed. These observations attest to the inherent danger of grade II BCI. Fortunately, despite arteriographic progression, just one patient suffered a stroke, suggesting a benefit of anticoagulation. Thus, even in the absence of an untreated control group, we believe it is prudent to recommend anticoagulation for these lesions until prospective data become available. The optimal type, dose, and duration of anticoagulation remains unclear.
Grade III injuries deserve special consideration. Pseudoaneurysm formation indicates egress of blood into the subadventitial layer. These lesions not only put the patient at risk for thromboembolism or continued enlargement with vessel obliteration, but also for rupture with intracranial or extracranial hemorrhage. Surgical repair has been promoted for these injuries and should be considered first-line therapy. However, one must examine the arteriogram carefully to exclude distal injury. Attempts at surgical repair may uncover a friable artery poorly suited for repair with a distal extent that obviates secure control. In our experience, grade III lesions are typically not accessible. Anticoagulation is directed at the thrombus that may form inside the lumen and potentially embolize. However, in our experience and that of others, 9,21 pseudoaneurysms do not resolve on heparin therapy alone. The collective experience of our group, along with those from Memphis 9 and Rochester, Minnesota, 21 indicates that only 4% of pseudoaneurysms (2 of 56 cases) resolve with anticoagulation alone. This finding raises the intriguing possibility that anticoagulation may worsen the lesion. It also calls for the evaluation of alternative therapeutic interventions for surgically inaccessible pseudoaneurysms.
Angiographic embolization and balloon occlusion are not enthusiastically used in the ICA because of fear of stroke. An alternative endovascular technique that is gaining increasing favor in the treatment of vascular lesions is the deployment of endovascular stents. 22 Stenting of the carotid arteries for atherosclerotic disease has enjoyed increasing popularity of late, despite a lack of data to support its durability and overall efficacy. 23 Our group published two early reports describing the use of endovascular stents in the setting of BCI. 24,25 We have deployed stents to treat persistent pseudoaneurysms, in an attempt to tack down the intima and exclude the pseudoaneurysm from the circulation. We have found this treatment to be efficacious in 89% of cases, but have learned two lessons. First, the acutely injured ICA should not be manipulated within 48 to 72 hours of the event. In our early experience, patients were returned for early arteriographic follow-up of high-grade lesions; however, two patients suffered strokes that may have been related to manipulation of catheters in the acutely injured artery. Consequently, our current protocol is to wait 7 days before attempting to stent BCI. A second lesson was that endovascular stents placed in traumatized arteries should be treated adjunctively with full systemic anticoagulation. We have had two stents occlude, resulting in one transient ischemic attack and one stroke, and another patient suffered transient ischemic attacks when systemic heparin was stopped.
A recent randomized trial comparing angioplasty and stenting with carotid endarterectomy for symptomatic atherosclerotic carotid artery stenoses was terminated prematurely because of an excessive stroke incidence in the angioplasty/stent group. 26 Although the management of traumatic lesions should not necessarily be extrapolated from the treatment of atherosclerotic disease, it is clear that endovascular stents are not a panacea for surgically inaccessible pseudoaneurysms. This conclusion creates a clinical dilemma: observation is not an ideal strategy, because pseudoaneurysms have the potential for rupture or thromboembolization with fatal outcomes. 27,28 The remaining option would seem to be extracranial-intracranial bypass. This technique, which requires the participation of an experienced microvascular neurosurgeon, has not proven superior to medical therapy in the treatment of atherosclerotic cerebral ischemia. 29 On the other hand, there have been promising results of this technique applied to traumatic lesions. 30 Until more data become available, we recommend heparin for inaccessible pseudoaneurysms; if they persist for 7 days, we support the use of endovascular stents under controlled protocols.
Grade IV BCI, i.e., complete occlusions, have been reported to recanalize, but probably do not do so in the early postinjury period. In our limited experience, these injuries have been associated with a high stroke rate; however, it seems that heparin therapy is protective against stroke. Four patients with grade IV injuries suffered stroke in our series; all did so before the initiation of anticoagulation. Four patients with grade IV lesions were treated with heparin, and none of them suffered stroke. Given the potential to prevent stroke, we recommend heparin for grade IV injuries until evidence suggests otherwise.
Transections of the carotid artery (grade V injuries) are devastating. In our experience, four of four cases have been lethal. Surgical ligation of the proximal artery may attenuate the hemorrhage, but the distal artery must be controlled. Balloon occlusion or embolization can be accomplished by means of surgical or interventional radiologic techniques. As we have found, patient demise often preempts such attempts. Arteriovenous fistulae are a subgroup of transections with which we have had little experience. Small arteriovenous fistulae, diagnosed by early venous filling without demonstration of the actual fistula, seem to be innocuous lesions, and might be considered grade I injuries. Larger arteriovenous fistulae may contribute to neurologic symptoms and should probably be approached by balloon occlusion techniques. 4,9 We suggest classification of these more extensive injuries as grade III. Additional experience with arterial transections and arteriovenous fistulae may help refine the classification of these lesions and clarify their optimal treatment.
Herein, we have discussed our experience with 109 cases of BCI and presented a new grading scale for these injuries. Although the presence of head injuries confounded the analysis of neurologic outcomes related to injury grades, there is clearly an association between increasing grade and stroke risk. We believe additional experience from multiple institutions will be instructive in continuing to refine our management of these challenging lesions.
1. Yamada S, Kindt GW, Youmans JR. Carotid artery occlusion due to nonpenetrating injury. J Trauma. 1967; 7: 333–342.
2. Krajewski LP, Hertzer NR. Blunt carotid artery trauma: report of two cases and review of the literature. Ann Surg. 1980; 191: 341–346.
3. Perry MO, Snyder WH, Thal ER. Carotid artery injuries caused by blunt trauma. Ann Surg. 1980; 192: 74–77.
4. Cogbill TH, Moore EE, Meissner M, et al. The spectrum of blunt injury to the carotid artery: a multicenter perspective. J Trauma. 1994; 37: 473–479.
5. Davis JW, Holbrook TL, Hoyt DB, et al. Blunt carotid artery dissection: incidence, associated injuries, screening, and treatment. J Trauma. 1990; 30: 1514–1517.
6. Martin RF, Eldrup-Jorgensen J, Clark DE, Bredenberg CE. Blunt trauma to the carotid arteries. J Vasc Surg. 1991; 14: 789–795.
7. Ramadan F, Rutledge R, Oller D, et al. Carotid artery trauma: a review of contemporary trauma center experiences. J Vasc Surg. 1995; 21: 46–56.
8. Fakhry SM, Jaques PF, Proctor HJ. Cervical vessel injury after blunt trauma. J Vasc Surg. 1988; 8: 501–508.
9. Fabian TC, Patton JH Jr, Croce MA, et al. Blunt carotid injury: importance of early diagnosis and anticoagulant therapy. Ann Surg. 1996; 223: 513–525.
10. 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: 462–470.
11. Prall JA, Brega KE, Coldwell DM, Breeze RE. Incidence of unsuspected blunt carotid artery injury. Neurosurgery. 1998; 42: 495–499.
12. Biffl WL, Moore EE, Mestek M. Computed tomographic angiography as a screening modality for blunt cervical arterial injuries: A cautionary note [letter]. J Trauma. 1999; 47: 438–439.
13. Bok APL, Peter JC. Carotid and vertebral artery occlusion after blunt cervical injury: the role of MR angiography in early diagnosis. J Trauma. 1996; 40: 968–972.
14. Deutsch LS. Anatomy and angiographic diagnosis of extracranial and intracranial vascular disease. In: Rutherford RB, ed. Vascular Surgery. Philadelphia, PA: WB Saunders; 1995: 1481–1507.
15. Mathews D, Walker BS, Purdy PD, et al. Brain blood flow SPECT in temporary balloon occlusion carotid and intracerebral arteries. J Nucl Med. 1993; 34: 1239–1243.
16. Ryu YH, Chung TS, Lee JD, et al. HMPAO SPECT to assess neurologic deficits during balloon test occlusion. J Nucl Med. 1996; 37: 551–554.
17. Thizon-de-Gaulle I. Antiplatelet drugs in secondary prevention after acute myocardial infarction. Rev Port Cardiol. 1998; 17: 993–997.
18. Topol E, Byzova T, Plow E. Platelet GPIIb-IIIa blockers. Lancet. 1999; 353: 227–231.
19. Colella JJ, Diamond DL. Blunt carotid injury: Reassessing the role of anticoagulation. Am Surg. 1996; 62: 212–217.
20. Eachempati SR, Vaslef SN, Sebastian MW, Reed RL II. Blunt vascular injuries of the head and neck: is heparinization necessary? J Trauma. 1998; 45: 997–1004.
21. Mokri B. Traumatic and spontaneous extracranial internal carotid artery dissections. J Neurol. 1990; 237: 356–361.
22. Ryu R. Endovascular stents and clinical applications: the state of the art. Appl Radiol. 1997; 26(suppl): 63–70.
23. Higashida RT. Carotid interventions: PTA and stenting. J Vasc Interv Radiol. 1998; 12: 84–88.
24. Bernstein SM, Coldwell DM, Prall JA, Brega KE. Treatment of traumatic carotid pseudoaneurysm with endovascular stent placement. J Vasc Interv Radiol. 1997; 8: 1065–1068.
25. Duke BJ, Ryu RK, Coldwell DM, Brega KE. Treatment of blunt injury to the carotid artery by using endovascular stents: an early experience. J Neurosurg. 1997; 87: 825–829.
26. Naylor A, Bolia A, Abbott R, et al. Randomized study of carotid angioplasty and stenting versus carotid endarterectomy: a stopped trial. J Vasc Surg. 1998; 28: 326–334.
27. Kaplan JA. Delayed fatal hemothorax due to traumatic carotid dissection: a case report of a previously unreported cause of death. J Forensic Sci. 1994; 39: 552–556.
28. Pretre R, Kursteiner K, Reverdin A, Faidutti B. Blunt carotid artery injury: devastating consequences of undetected pseudoaneurysm. J Trauma. 1995; 39: 1012–1014.
29. Haynes R, Mukherjee J, Sackett D, et al. Functional status changes following medical or surgical treatment for cerebral ischemia. Results of the extracranial-intracranial bypass study. JAMA. 1987; 257: 2043–2046.
© 1999 Lippincott Williams & Wilkins, Inc.
30. Rostomily R, Newell D, Grady M, et al. Gunshot wounds of the internal carotid artery at the skull base: management with vein bypass grafts and a review of the literature. J Trauma. 1997; 42: 123–132.