|Home | About | Journals | Submit | Contact Us | Français|
Optimal treatment for patients with symptomatic intraluminal carotid thrombus (ICT) remains poorly defined.
We performed a retrospective chart review of patients presenting to our institution between 2001 and 2011 with symptomatic ICT.
Twenty-four patients (16 males, 8 females) with ICT presented with ischemic stroke (n=18) or TIA (n=6). All were treated initially with anticoagulation ± antiplatelet drugs. Eight of these patients had no or mild carotid stenosis on initial angiography and were treated with medical management alone. The remaining 16 patients had moderate or severe carotid stenosis on initial angiography. Of these, 10 underwent delayed revascularization (endarterectomy, n=8; angioplasty and stenting, n=2), 2 refused revascularization, and 4 were treated with medical therapy alone. One patient had multiple TIAs despite medical therapy and eventually underwent CEA; the remaining 23 patients had no TIAs. No patient suffered ischemic or hemorrhagic stroke while on anticoagulation, during the perioperative period or in long-term follow up; one patient died of an unrelated condition (mean follow-up = 16.4 months).
Our results suggest that initial anticoagulation of symptomatic ICT results in a low rate of recurrent ischemic events and that carotid revascularization, if indicated, can be safely performed in a delayed manner.
Intraluminal carotid artery thrombus (ICT) is an infrequent finding in patients presenting with transient ischemic attack (TIA) or ischemic stroke. It can be a focal process limited to the cervical internal carotid artery (ICA), or a more extensive condition that involves both the extracranial and intracranial ICA.6 ICT can be found in the presence or absence of carotid stenosis. When identified in patients without underlying carotid stenosis, it is often associated with a hypercoagulable state.3,6 When identified in patients with underlying carotid stenosis, its incidence is correlated to the severity of atherosclerotic disease. In the North American Symptomatic Carotid Endarterectomy Trial (NASCET), the frequency of ICT was 1.1% in patients with <70% stenosis vs. 4.3% in patients with >70% stenosis and 5.5% in patients with >85% stenosis.4,18
Once ICT is identified, its presence merits special consideration due a high risk of recurrent ischemic events. In NASCET, patients randomized to antiplatelet medical therapy who had ICT on initial angiography had a significantly higher risk of stroke and death at 1 month as compared to medically treated patients without ICT (11% vs. 2%, respectively).4,18 Due to this aggressive clinical course, some consider ICT an urgent or emergent condition that requires early carotid endarterectomy (CEA);7,19 however, the risks associated with this strategy have been considerable.6 In NASCET, patients undergoing CEA in the presence of ICT had twice the risk of perioperative stroke and death as compared to patients undergoing CEA without ICT (12% vs. 6%, respectively).4,18 Similarly, in a multicenter retrospective review of symptomatic patients undergoing CEA, patients with ICT had a nearly threefold higher risk of perioperative stroke as compared to patients without ICT (14% vs. 5%, respectively).10
Optimal management of patients with symptomatic ICT, therefore, remains poorly defined. Reported treatment options include- 1) medical management with anticoagulation ± antiplatelet drugs,6 2) early carotid revascularization with CEA5,6 or carotid artery stenting (CAS),13,14 and 3) initial medical management with anticoagulation ± antiplatelet drugs followed by delayed revascularization via CEA6,8 or CAS.11 However, results with each of these therapeutic strategies are based on a few small case series, and available evidence does not provide clarity as to best practice for this patient population.
At our institution, patients with symptomatic ICT are treated with initial anticoagulation followed by delayed revascularization via CEA or CAS, if indicated (Figure 1). In this study, we retrospectively evaluated our experience in the management of patients presenting with TIA or acute ischemic stroke due to ICT, and compared our results to the published literature.
This study was approved by the Human Research Protection Office.
In our practice, digital subtraction angiography (DSA) is obtained as part of routine care in patients presenting with ischemic symptoms and Doppler ultrasound evidence of ≥50% stenosis of the ipsilateral carotid artery, in order to determine whether they are candidates for surgical or endovascular revascularization. Patients that are not good candidates for either procedure owing to their health or the severity of their stroke do not undergo DSA. A computerized database of adult patients who underwent DSA at our institution between January 2001 and February 2011 was queried and patients with a diagnosis of ICT, defined as a filling defect in the arterial lumen that was visible on at least two planes, were identified. All studies were performed by one of three senior interventional neuroradiologists.
Patients who were symptomatic (TIAs or stroke) with intraluminal thrombus in the common carotid artery (CCA) or extracranial internal carotid arteries (ICA) were included in this study. Exclusion criteria were- 1) Thrombus confined to intracranial ICA, 2) Thrombus associated with trauma and/or dissection, 3) History of CAS ipsilateral to the thrombus (i.e. instent thrombosis), 4) Ischemic stroke with hemorrhagic transformation, and 5) Extensive medical co-morbidities at the time of presentation that limited life expectancy.
Medical records of identified patients were retrospectively reviewed and the following data collected- age, sex, comorbidities (hypertension, diabetes mellitus, hyperlipidemia, coronary artery disease, chronic heart failure), history of previous stroke or TIA, current medications (especially antiplatelet agents or anticoagulants), smoking history, clinical presentation (TIA or stroke, number of events prior to diagnosis, time from first event to diagnosis, NIHSS at presentation), admission laboratory values (including hemoglobin, hematocrit, and platelet count), angiographic findings at presentation (degree of stenosis bilaterally, location of thrombus relative to stenosis if present, and hemodynamic effect of the thrombus), treatment (use of anticoagulant and antiplatelet agents, duration of anticoagulation), and follow up (clinical and radiographic). Comorbidities and current medications were identified from self reported history. Clinical presentation was considered to be TIA if the patient presented with an acute focal neurological deficit(s) lasting less than 24 hours, and stroke if there was acute onset focal neurological deficit(s) lasting greater than 24 hours. The duration of symptoms was recorded as 1 day if symptoms were present for 24 hours or less. NIHSS score was assessed retrospectively from examination findings at admission if not directly available in the medical records.15 Outcome was assessed using radiographic and clinical evaluation by a treating neurologist or neurosurgeon at follow up. The degree of stenosis was classified per NASCET criteria.2 For patients who were receiving anticoagulant drugs at last follow up, the interval from initiation of treatment to last follow up was considered to be the duration of anticoagulation.
In our practice, patients with ICT who have minimal atherosclerotic disease on initial angiography undergo a diagnostic workup for underlying hypercoagulable conditions including factor V Leiden mutation, prothrombin G20210A mutation, protein C deficiency, protein S deficiency, antithrombin deficiency, antiphospholipid antibody syndrome and hyperhomocysteinemia. In addition, we have previously identified coexisting anemia and thrombocytosis as a risk factor for ICT in patients without significant underlying carotid stenosis.3 These data were also extracted from the medical records.
Twenty-four patients (16 males, 8 females) with ICT satisfying the inclusion and exclusion criteria were identified. The median age at presentation was 55 years (IQR: 46 to 68 yrs). The majority of patients had known risk factors for stroke: hypertension (54%), hyperlipidemia (50%), smoking (33%), history of stroke/TIA (25%), coronary artery disease (25%), diabetes mellitus (21%), and atrial fibrillation (4.2%). There was a history of use of oral contraceptives in one female patient (12.5%). At presentation, 7 patients (29%) were on antiplatelet drugs, and 1 patient (4.2%) was on warfarin for atrial fibrillation with a therapeutic INR (2.2).
Among the 24 patients, 7 patients had a hypercoagulable state – 1 with antiphospholipid antibody syndrome, 1 with systemic lupus erythematosus and thrombocytosis, 1 with essential thrombocytosis, and 4 with iron deficiency anemia with thrombocytosis. A diagnostic work up for other hypercoagulable conditions including factor V Leiden mutation, prothrombin G20210A mutation, protein C deficiency, protein S deficiency, antithrombin deficiency, antiphospholipid antibody syndrome and hyperhomocysteinemia was negative in patients with iron deficiency anemia and thrombocytosis. The median age of patients presenting with hypercoagulability was significantly lower than those without a hypercoagulable state (43 vs. 59 years, p=0.006 by Mann Whitney test). Patient characteristics are further outlined in Tables 1 and and22.
Most patients (75%) presented with ischemic stroke; the remainder presented with TIA (25%). Duration of symptoms ranged from <3 hours to 4.5 months (median: 1 day; interquartile rage (IQR): 1 to 2 days). However, all patients were acutely symptomatic, with the last symptomatic episode being < 1 week from presentation. Median NIHSS score at presentation was 4 (IQR: 1 to 10) in patients with ischemic stroke. There was insufficient data for NIHSS assessment in two patients.
Among the 24 patients, 12 had high-grade stenosis, 4 had moderate stenosis and 8 had no or mild stenosis on initial angiography per NASCET criteria.1 One patient who presented with TIA and had no carotid stenosis on initial angiography had a history of ipsilateral CEA for severe stenosis one year prior. Decreased distal flow and/or luminal collapse of the distal ICA due to the ICT was noted in 14 of 24 patients.
Among the 7 patients with a hypercoagulable state, 6 had no or minimal stenosis and one had severe stenosis on initial angiography. Decreased distal flow and/or luminal collapse of the distal ICA was seen in 3 of these patients.
All patients were treated at presentation with anticoagulation. The anticoagulation regimen varied– unfractionated heparin (UFH) only (n=3), low molecular weight heparin (LMWH) only (n=2), warfarin only (n=1), UFH followed by warfarin (n=10), UFH with LMWH bridge followed by warfarin (n=4), LMWH followed by warfarin (n=3), and UFH followed by LMWH (n=1). None of the patients developed anticoagulation associated complications. Nineteen patients also received concurrent antiplatelet therapy of which 17 were treated with aspirin alone while 2 received aspirin and clopidogrel.
Among patients with a hypercoagulable state, 4 were anticoagulated with UFH and warfarin, 2 with UFH, LMWH bridge and warfarin, and 1 with LMWH followed by warfarin. Four patients also received antiplatelet agents (aspirin). Treatment details are summarized in Tables 3 and and44.
Of the 8 patients with no or mild carotid stenosis on initial DSA, two had follow up vacular imaging. One patient underwent follow up magnetic resonance angiography 12 days after initiation of anticoagulation to evaluate new onset headache. It showed complete resolution of the ICT and absence of residual carotid stenosis. The other patient underwent follow up computed tomography angiography (CTA) 7 months after initiation of anticoagulation due to uncertainty regarding the degree of stenosis on initial DSA (49%). Complete resolution of the ICT and mild residual stenosis was noted. Of the 16 patients with moderate or high grade stenosis on initial DSA, 11 underwent repeat vascular imaging (all DSA). Complete resolution of the ICT was observed in 10 of these 11 patients, while one had substantial but incomplete resolution of the ICT. Residual moderate to severe stenosis was observed 10 of these 11 patients, while one patient had only minimal residual stenosis. Follow up data is outlined further in Tables 3 and and44.
Among the 8 patients with no or mild carotid stenosis on initial DSA, none were offered delayed revascularization. Among the 16 patients with moderate to high grade carotid stenosis on initial DSA, 8 underwent delayed CEA, 2 underwent delayed CAS, 2 refused delayed revascularization, 3 were not offered delayed revascularization (2 due to severe comorbidities; 1 due to downgrading of carotid stenosis to minimal on repeat DSA), and 1 was lost to follow up after initiation of anticoagulation. Mean interval from clinical presentation to delayed revascularization was 39.3 days (range 15 – 91 days).
Mean duration of anticoagulation was 30 days (range 7 – 53 days) in the 10 patients who underwent delayed revascularization and 103 days (range 7 – 318 days) in the 13 patients who underwent medical therapy alone (duration of anticoagulation was not available in one patient, who was lost to follow up). Mean duration of clinical follow up for the entire patient cohort was 70 weeks (range 5–368 weeks). No patient in this cohort suffered a recurrent ischemic stroke. One patient (4.2%), with high-grade stenosis on initial DSA, experienced TIAs while on anticoagulation. He underwent follow up DSA 28 days after initiation of anticoagulation, which revealed substantial but incomplete resolution of the ICT and 90% residual carotid stenosis. He underwent an uncomplicated CEA the next day where no evidence of ICT was noted, and he experienced no recurrent ischemic events in long-term follow up (74.5 weeks). Among patients who underwent delayed revascularization via CEA or CAS, none suffered peri-operative ischemic stroke or death. One patient had a post-CEA neck hematoma that required surgical evacuation. No procedural complications were noted in those undergoing CAS.
In long-term follow up, one patient died due to complications of metastatic breast cancer. This patient had no ischemic complications during the 90.2 weeks of available follow up. There was no other patient mortality. One patient was lost to follow up.
Symptomatic ICT is associated with high risk of recurrent ischemic events and poor patient outcome.4 Optimal management of this patient population remains controversial, mainly due to the rarity of reported cases. Here, we report excellent results with a strategy of initial anticoagulation followed by delayed revascularization (as indicated). None of our 24 patients (14 of whom were treated with anticoagulation alone; 10 of whom were treated with anticoagulation followed by delayed revascularization) suffered new ischemic or hemorrhagic stroke and none suffered ICT-related death in late clinical follow up (mean = 70 weeks) (Tables 1–4). When comparing our results to published reports with strategies of antiplatelet therapy alone (11% risk of stroke and death at 1 month)4,18 or early CEA (12–14% risk of stroke and death at 1 month),4,10,18 the strategy of initial anticoagulation ± antiplatelet therapy to permit ICT resolution followed by delayed revascularization (if needed) appears superior (Table 5).
Results from smaller case series in which ICT patients were treated with anticoagulation alone or in combination with delayed revascularization generally echo our experience (Table 5). Buchan et al.6 treated 6 ICT patients with anticoagulation alone and reported no recurrent ischemic events and no deaths in late follow up. They treated 3 additional ICT patients with initial anticoagulation followed by delayed CEA and reported no recurrent ischemic events in the interim between initiation of anticoagulation and surgery. One of these 3 patients, however, did suffer a fatal post-CEA stroke. Combe et al.8 treated 5 ICT patients and Gonzalez et al.11 treated 3 ICT patients with initial anticoagulation followed by delayed CEA and reported no recurrent ischemic events in the interim between initiation of anticoagulation and surgery and no incidence of post-CEA stroke or death.
Two alternative approaches to the management of patients with ICT and suspected underlying stenosis have been reported – early CEA and early CAS (Table 5). Regarding early CEA, results from NASCET (12% perioperative stroke/death rate for ICT patients vs. 6% perioperative stroke/death rate for non-ICT patients) and a large multicenter retrospective review (14% perioperative stroke rate for ICT patients vs. 5% perioperative stroke rate for non-ICT patients) strongly suggest that this treatment strategy for ICT patients likely carries unacceptably high risk. Smaller single institution case series generally corroborate this conclusion. Biller et al.5 treated 5 ICT patients with early CEA and noted a 20% risk of perioperative stroke/death. Walters et al.19 treated 11 ICT patients with early CEA and noted a 27% risk of perioperative stroke/death. Buchan et al.6 treated 14 ICT patients with early CEA and noted a 29% risk of perioperative stroke/death. In contradistinction, Caplan et al.7 treated 6 patients with early CEA and noted a 0% risk of perioperative stroke/death.
Regarding early CAS for ICT patients with underlying stenosis, initial results have been more favorable than that reported for early CEA. In the first reported case series, Tsumoto et al.17 treated 6 ICT patients with early CAS and noted a 17% risk of perioperative stroke/death. However, two subsequent series of 9 ICT patients13 and 6 acute stroke patients with ICT14 noted a 0% risk of stroke/death following early CAS. Taken together, these three small case series13,17 and other individual case reports12,17 suggest that early CAS for treatment of patients with symptomatic ICT may be a reasonable approach, though the number of reported cases remains small. However, it should be noted that the degree of stenosis in many of these patients maybe minimal once the thrombus has lysed. Our data suggests that anticoagulation for a week to 10 days followed by re-imaging may avoid a revascularization procedure in many patients, with minimal risk of stroke.
In most cases, ICT occurs in the setting of underlying atherosclerotic disease that predisposes to thromboembolism;6 however, it has also been reported in the absence of significant atherosclerotic disease. The latter usually occurs in the setting of a hypercoagulable state.3,7,9,16,20 In our series, 7 patients had a hypercoagulable condition, the most common of which was thrombocytosis. The formation of ICT in the setting of essential thrombocytosis or anemia with secondary thrombocytosis is rare, with only a few reported cases.3,7,9,16,20 Our results with a strategy of anticoagulation for this subset of ICT patients suggest that medical management may be preferred as compared to early CEA or early CAS.
Our study has several limitations. First, this is a retrospective case series with limitations inherent to the study design. Second, follow up angiography to confirm thrombus resolution was not obtained in all patients. Third, the methods and length of anticoagulation employed in our series were varied. Fourth, the concurrent use of antiplatelet therapy during the period of anticoagulation was not uniform. Fifth, the small sample size of our series could have underestimated the true incidence of stroke/death in patients who are treated with our strategy of anticoagulation alone, or anticoagulation followed by delayed revascularization. Given the rarity of this condition, however, such limitations would be difficult to avoid.
In conclusion, this is the largest case series to date demonstrating that patients with symptomatic ICT can be managed safely and effectively via a strategy of initial anticoagulation followed by delayed carotid revascularization (if indicated).
Source of Funding: This study was supported by NIH grants - NINDS P50 NS055977 (C.P.D., G.J.Z.), NINDS U01 NS58728 (C.P.D.), and NINDS R01 NS051631 (C.P.D).
Presentation Information: Portions of this work were presented in abstract form at the American Association of Neurological Surgeons – 2011 Annual Scientific Meeting, Denver, USA, April 11, 2011.
DISCLOSURE C.P.D. serves as a consultant for W.L.Gore Inc. and Pulse Therapeutics, and is a shareholder in nFocus.