Author + information
- Received August 23, 2006
- Revision received September 27, 2006
- Accepted November 28, 2006
- Published online February 20, 2007.
- Hsien-Li Kao, MD⁎,1,⁎ (, )
- Mao-Shin Lin, MD†,1,
- Chia-Sung Wang, MD⁎,
- Yen-Hong Lin, MD⁎,
- Lung-Chun Lin, MD⁎,
- Chia-Lun Chao, MD⁎,
- Jiann-Shing Jeng, MD‡,
- Ping-Keung Yip, MD‡ and
- Shih-Chung Chen, MD§
- ↵⁎Reprint requests and correspondence:
Dr. Hsien-Li Kao, National Taiwan University Hospital, 7 Chung-Shan South Road, Taipei, Taiwan.
Objectives This study sought to report technical details and clinical results of the first series of endovascular recanalization for cervical internal carotid artery (ICA) occlusion.
Background Cervical ICA occlusion is associated with impaired cerebral perfusion, which may lead to ischemic cerebral symptoms and hemodynamic infarcts. Neither surgical nor endovascular revascularization has been shown to benefit this population.
Methods Endovascular recanalization was attempted in 30 patients with ICA occlusions (27 men; age 72.1 ± 8.0 years, range 48 to 85 years). Recurrent neurologic deficit or cerebral ischemia by perfusion study, after known ICA occlusion, was noted in all patients. Strategies and devices for coronary occlusion intervention were applied, including microcatheter-supported tapered-tip stiff coronary guidewires. Contralateral ICA stenosis was found in 9 patients (30%). All patients underwent independent neurologic and duplex ultrasound follow-up.
Results The overall technical success rate was 73% (22 of 30). No neck hematoma, intracranial hemorrhage, or hyperperfusion was noted. One (3.3%) fatal brainstem infarction occurred 1 day after a successful ICA procedure, with angiographically proven acute basilar artery occlusion and patent ICA stent. Baseline ophthalmic artery flow was reversed in 15 of the 22 successfully recanalized patients, and was normalized in 12 after the procedure. There was no new cerebral ischemic event or neurologic death for a mean follow-up of 16.1 ± 18.5 months.
Conclusions Endovascular recanalization for cervical ICA occlusion is feasible with acceptable midterm clinical results.
The risk of artery-to-artery embolism in cervical internal carotid artery (ICA) occlusion should be low, because antegrade flow no longer exists. However, inadequate cerebral perfusion via collateral vessels still can cause ischemic symptoms. Actually, cervical ICA occlusion was associated with an annual risk of 6% to 20% of ipsilateral recurrent stroke despite intensive medical management (1,2). Extracranial-to-intracranial (EC-IC) artery bypass may improve cerebral hemodynamics in patients with ICA occlusion (3,4), but it failed to reduce the risk of ischemic stroke (5). Evidence has accumulated showing that there might be a role for revascularization in a highly selected group of patients in whom inadequate hemodynamics can be shown (6,7). However, the significant complexity and technical difficulty of the procedure mandate trials to prove its clinical efficacy.
Carotid endarterectomy (CEA) prevents ischemic stroke in patients with cervical ICA stenosis, but not in occlusion (8,9). Endovascular carotid artery stenting (CAS) is an alternative to CEA in high-surgical-risk patients (10,11), but patients with cervical ICA occlusion were not included in previous CAS trials. The application of endovascular intervention in cervical ICA occlusion has never been explored in the literature. We report our initial experience of endovascular recanalization for cervical ICA occlusion, focusing on its feasibility and the interventional technique.
Since October 2002, endovascular intervention was attempted in 30 patients (27 men; mean age 72.1 ± 8.0 years, range 48 to 85 years) with unilateral cervical ICA occlusion, out of the 255 CAS procedures done in the same time period (11.8%). Occlusion was documented by carotid ultrasound or magnetic resonance angiography. Then the patient was followed up clinically by an independent neurologist for symptom progression under intensive medical therapy. The most recent cerebral infarction, if documented, should be at least 2 weeks before the intervention. Twenty-five patients (83%) suffered from progression or recurrence of ipsilateral neurologic deficit after documenting the occlusion. Another 5 patients with prior ipsilateral stroke and stable neurologic status had ipsilateral hemisphere ischemia shown by perfusion computed tomography (CT) or Diamox stress 133Xe CT. After obtaining informed consent, all 30 patients were admitted for selective cerebral angiography and endovascular intervention. The institutional review board approved the retrospective review of medical and radiologic records.
A selective carotid angiogram was performed using a femoral approach in all patients, including cervical and intracranial views in 2 orthogonal projections in a biplanar setting. All angiograms were taken after an intra-arterial bolus of 200 μg nitroglycerin. The occluded segment was examined carefully to rule out pseudo-occlusion by the characteristics described in a previous publication (12). The distal collateral circulation was also evaluated carefully, with emphasis on the late retrograde opacification of the ICA segment distal to the occlusion. Angiographic criteria for true occlusion were: 1) discontinuation of ICA lumen at least 5 mm in length, 2) Thrombolysis In Myocardial Infarction (TIMI) antegrade flow grade 0 distal to the occlusion, and 3) established collateral filling to the ipsilateral intracranial ICA territory, either via anterior communicating artery, posterior communicating artery, or ipsilateral ophthalmic artery (OA).
Quantitative measurement was done off line by an independent cardiologist with calibration of the system using the known diameter of the catheters. Measurement of common carotid artery and cervical ICA diameters and occlusion length were done after pre-dilatation and establishment of antegrade flow in ICA. Final residual diameter stenosis was determined with the use of the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria, with the distal nontapering portions of the post-intervention ICA after nitroglycerin as the reference segment (8).
Patients received aspirin 100 mg and clopidogrel 75 mg per day for at least 7 days before the procedure. Heparin bolus was given to maintain the activated clotting time between 200 to 250 s during intervention. An 8-F JR4 guiding catheter, which provided not only adequate backup support but also the possibility of passing multiple devices in parallel fashion, was positioned transfemorally into the common carotid artery proximal to the occlusion. A 0.014-inch coronary angioplasty guidewire with intermediate stiffness (Rinato, Asahi Intecc, Aichi, Japan) was advanced through a microcatheter (Transit, Cordis, Miami, Florida) in the guiding catheter to engage the stump of the ICA occlusion. The wire was then exchanged to a tapered-tip stiff 0.014-inch coronary angioplasty guidewire specially designed for chronic coronary occlusion (Conquest, Asahi Intecc, Aichi, Japan) through the microcatheter. The Conquest wire was carefully manipulated with a penetrate-and-advance fashion, avoiding unnecessary rotational or drilling motion. The wiring was aborted if crossing was unsuccessful after 30 min of manipulation, or consumption of 300 ml contrast, or obvious extraluminal course of the wire was observed.
Once the wire crossed the occluded segment, confirmation of its position in the distal true lumen was done with multiple angiographic projections. The microcatheter was exchanged for a small-diameter (1.5 or 2.0 mm) coronary angioplasty balloon, which was inflated to 6 to 8 atm to pre-dilate the occlusion. An embolic protection device (FilterWire EZ, Boston Scientific, Mountainview California, or PercuSurge GuardWire, Medtronic, Minneapolis, Minnesota) was then advanced parallel to the Conquest wire and deployed distally if an adequate distal landing zone with vessel diameter >3 mm could be identified. A self-expanding stent (Carotid Wallstent, Boston Scientific, Galway, Ireland) was then placed across the occlusion, followed by post-dilation using a 4- to 6-mm-diameter balloon catheter to achieve a residual diameter stenosis of <20%. A final ipsilateral intracranial angiogram was obtained to confirm re-established antegrade perfusion. The intervention was considered a technical success if the occlusion was crossed and stented, with a final residual diameter stenosis <20% and TIMI distal antegrade flow grade 3. An exemplary case is shown in Figures 1 and 2.⇓
Daily clopidogrel 75 mg and aspirin 100 mg were continued for at least 3 months after successful intervention. No more heparin was given after the procedure. All patients were sent to the intensive care unit for overnight monitoring of hemodynamic and neurologic status, and systolic blood pressure was carefully maintained within 100 to 140 mm Hg.
Data collection and follow-up
All clinical, angiographic, and procedural data were retrospectively collected from the medical chart and recorded on standard forms by a physician.
Complete neurologic examination, including assessment of National Institutes of Health Stoke Scale, was performed by the neurologist within 24 h after the procedure. Patients were then followed up by the neurologist at 1, 6, and 12 months after the procedure. Carotid ultrasound and transocular duplex evaluation of OA flow direction was also performed by the neurologist at 1, 6, and 12 months. Clinical events including any stroke, transient ischemic attack (TIA), or death were documented on the chart. Follow-up angiography was arranged when the findings on carotid ultrasonography suggested that restenosis (i.e., more than 50% stenosis) had developed. Angiographic restenosis was defined as diameter stenosis >50% in the stented artery segment.
A TIA was defined as a focal neurologic deficit that resolves completely within 24 h. A minor stroke was defined as a nondisabling deficit persisting longer than 24 h but resolving completely with 1 week. A major stroke was classified as persistent neurologic deficit longer than 7 days.
All continuous variables were expressed as mean ± SD, and categorical variables in numbers and percentage.
Baseline clinical characteristics of the patients were listed in Table 1.The average age was 72.1 ± 8.0 years old, with male predominance (90%). All patients were symptomatic according to NASCET definition, mostly presenting with prior ipsilateral strokes within 180 days. Nine (30%) patients also had contralateral ICA stenosis >50%, and 7 received CAS for the contralateral lesion according to the established CAS indications (11) (5 after and 2 before recanalization of the ICA occlusion).
Angiographic characteristics and procedural data
Table 2summarizes the angiographic measurements of the target lesions and the procedural data. The average estimated occlusion length was 27.9 ± 16.2 mm. Wire crossing was successful in 73% (22 of 30) of the cases, all followed by successful balloon pre-dilatation and stent deployment. Distal embolic protection devices were used in 77% (17 of 22) of the crossed lesions, and not in the other 5 because of small distal vessel diameter. The final residual diameter stenosis after stent deployment and post-dilatation was 10 ± 7%. The stents were deployed across the take-off of the external carotid artery, jailing its orifice in all recanalized lesions.
The technical success rate was 73% (22 of 30). All 8 failed procedures were caused by an inability to pass the guidewire across the occlusion. There was no vessel perforation or extravasation observed. One patient (3.3%) suffered a fatal nonipsilateral major stroke 1 day after the procedure. A 55-year-old man with known left ICA occlusion suffered from recurrent cerebral stroke. The angiography also showed >80% stenosis at right vertebral artery ostium, hypoplastic left vertebral artery, and patent basilar artery. The left ICA occlusion was successfully crossed and stented with FilterWire protection. He was sent to the intensive care unit for routine overnight observation, with stable hemodynamic and neurologic status. Consciousness change and bilateral limbs weakness was noticed 8 h later, and CT showed brainstem and cerebellar infarction with marked swelling. Emergent angiography showed basilar artery acute occlusion, and the left ICA stent was patent. Intra-arterial thrombolytic therapy was given, but reperfusion was not achieved. Massive brainstem hemorrhage ensued and the patient died 2 days later.
Bradycardia and hypotension were noted in 3 (10%) patients after stenting; all resolved after fluid supplement or inotropic agent infusion within 24 h. No hyperperfusion was observed in the present cohort.
Follow-up and OA flow direction
All patients, except for the mortality mentioned above, were discharged 2.4 ± 1.3 days after intervention. There was no new TIA or ischemic stroke noted in 28 patients during the 16.1 ± 18.5 months of follow-up (range 1 to 86 months), but 1 (3.4%) nonneurologic death was recorded in an 85-year-old male patient 11 months after successful recanalization. In-stent restenosis was found in 3 of the 22 successful cases (13.6%); all patients were asymptomatic and confirmed with angiography at 6 months. Two of these were re-occlusions, and were left without further treatment because no recurrent symptoms could be documented. The remaining case was a >80% diameter stenosis and was re-stented smoothly after inadequate balloon angioplasty. Table 3summarizes the in-hospital and follow-up results.
Good-quality transocular duplex evaluation of the ipsilateral OA was obtained in 25 of 30 patients (83%) before the intervention, and reversed flow direction was found in 21 (84%, 21 of 25). In the 22 patients successfully recanalized, pre-procedural OA flow direction was documented in 17 (77%), and in 15 was reversed (88%, 15 of 17). Follow-up duplex at 1 month showed normalization of OA flow in 12 of the 15 patients (80%), with another 2 remaining reversed, and 1 unexamined because of the fatal event mentioned previously. The 2 patients with persistent reversed OA flow after successful intervention were later both found to have re-occlusion.
In this study, we have shown that it is possible to recanalize ICA occlusion with an endovascular procedure. Applying the guidewire experience obtained in coronary artery occlusion intervention, the technical success rate of the current series was 73% (22 of 30). No neck hematoma, intracranial hemorrhage, or hyperperfusion occurred. One acute basilar occlusion with fatal posterior infarction was the only complication, but the etiology was most likely an unfortunate independent embolic episode, either artery-to-artery or cardiogenic. Thus, the periprocedural complication rate of the current series was not different from that of previous CAS studies (10,11,13), and certainly compared favorably with the CEA or EC-IC bypass surgery results in ICA occlusion (5,14).
Internal carotid artery occlusion is a relatively uncommon but important cause of TIA and cerebral infarction. Although a sizable proportion of patients could be treated successfully with intensive medical therapy, the annual ipsilateral stroke rate was still around 6% to 20% in this population (1,2). Carotid endarterectomy prevents stoke in patients with ICA stenosis (8,15), but the success rate in recanalizing occlusions was as low as 34%, because of technical difficulties (14). Surgical bypass may be a natural resolution for ICA occlusion, but the large international randomized EC-IC bypass trial failed to show any benefit (5). Many physicians thus abandoned the effort to “reopen a closed door.” Recently, researchers focused on the hemodynamic failure distal to ICA occlusion and the identification of a subpopulation that may benefit from revascularization. Compromised cerebral blood flow plays an important role in causing ipsilateral ischemic events in patients with ICA occlusion (6). Grubb et al. (7) used positron-emission tomography scanning to describe the importance of hemodynamic factors in predicting outcomes among these patients. It implies that medical management alone may not be suitable for certain patients (16), and efforts to select patients with impaired hemodynamics for EC-IC bypass surgery are ongoing (17).
With occlusion and absence of antegrade flow through ICA, in theory, the risk of distal embolization should be minimal. Rothwell et al. (18) reported a lower risk of ischemic stroke in the patients with severe ICA stenosis and reduced distal vessel diameter. They concluded that the diminished flow distal to the lesion is insufficient to carry emboli forward, and post-stenotic ICA narrowing protects the brain from infarction. Similar results were also found in the NASCET study (19), and analysis showed that CEA was of little benefit in patients with near-occlusion (20). However, impaired perfusion, instead of embolism, may be the actual pathophysiology in patients with known ICA occlusion and recurrent symptoms. In the Harvard Stroke Registry, 20% (95 of 471) of stroke patients suffered from deterioration after the initial event, and 33% of them had large artery occlusion (21). It was postulated that the progression in patients is most likely caused by decreased perfusion and/or a decreased potential for developing collateral to the ischemic zones (22). This concept was compatible with the observation by Henderson et al. (23), who reported that the existence of collateral circulation was associated with a lower risk of hemispheric events in patients with severe ICA stenosis. Therapy that improves cerebral blood supply thus may be beneficial in patients with impaired perfusion caused by ICA occlusion. In our series, the majority of the patients (25 of 30, 83%) suffered from progression of neurologic deficit or repeated TIA after known ICA occlusion. The other 5 patients with a stable neurologic condition also had hemodynamic insufficiency documented by the perfusion study. Therefore, our cohort should benefit from revascularization, and our results did show a dramatic absence of new neurologic events after successful recanalization during follow-up.
Reversal of OA flow direction has been a good marker of hemodynamic insufficiency caused by proximal ICA stenosis (24–27). The ipsilateral OA flow direction was reversed in 84% (21 of 25) of our patients before ICA recanalization, and the reversed OA flow was normalized in 80% of the cases just 1 month after successful intervention. The convincing evidence showed that ICA recanalization re-establishes normal cerebral hemodynamics. Failure to enjoy this rapid flow normalization was also found to be a good marker for in-stent re-occlusion. Therefore, we showed clearly the importance of duplex OA flow analysis in the evaluation and follow-up of patients with ICA occlusion.
The technical challenge for endovascular recanalization is the passage of a guidewire across the occlusion. The concepts and equipment used in chronic coronary occlusion were applied. A microcatheter was first placed proximal to the occlusion to improve the back-up support for the subsequent tapered-tip stiff wire manipulation. Excessive rotational or drilling motion of the wire was avoided, but successive small penetrate-and-advance steps were made carefully along the imaginary tract of the occluded vessel segment. A high-quality biplanar imaging system is desirable to limit the procedure time and contrast volume. With careful handling of the wire and meticulous confirmation of its position, we did not experience any vessel perforation or hematoma formation. The potential risk of dislodging a trailing thrombus in the ICA distal to the occlusion was not observed in our series, perhaps because of the very gentle undersized pre-dilatation and the frequent usage (77%) of an embolic protection device before actual stent deployment.
The current study has several limitations. Although the reversed OA flow is a marker of hemodynamic insufficiency, perfusion studies such as Diamox stress 133Xe CT or perfusion CT were not performed systematically to document impaired cerebral perfusion. Comparison of the results of these imaging studies before and after recanalization is mandatory in future research. Second, the small case number and use of conventional end points such as a new TIA or stroke are inadequate to fully show the efficacy of the procedure. Incorporating neurocognitive functional changes before and after recanalization in a larger patient population is necessary in the future.
In conclusion, this small single-center series shows that it is feasible to treat ICA occlusion with endovascular techniques by experienced interventionalists. Future prospective randomized studies with larger patient numbers are mandatory to establish its clinical efficacy and indications.
↵1 Drs. Kao and M.-S. Lin contributed equally to this work.
- Abbreviations and Acronyms
- carotid stenting
- carotid endarterectomy
- computed tomography
- internal carotid artery
- ophthalmic artery
- transient ischemic attack
- Received August 23, 2006.
- Revision received September 27, 2006.
- Accepted November 28, 2006.
- American College of Cardiology Foundation
- Derdeyn C.P.,
- Gage B.F.,
- Grubb R.L. Jr..,
- Powers W.J.
- Neff K.W.,
- Horn P.,
- Dinter D.,
- Vajkoczy P.,
- Schmiedek P.,
- Duber C.
- Klijn C.J.M.,
- Kappelle L.J.,
- Tulleken C.A.F.,
- van Gijn J.
- Rothwell P.M.,
- Warlow C.P.,
- European Carotid Surgery Trialists’ Collaborative Group
- Morgenstern L.B.,
- Fox A.J.,
- Sharpe B.L.,
- Eliasziw M.,
- Barnett H.J.M.,
- Grotta J.C.,
- North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group
- Rothwell P.M.,
- Gutnikov S.A.,
- Warlow C.P.,
- European Carotid Surgery Trialist’s Collaboration
- Caplan L.R.
- Henderson R.,
- Eliasziw M.,
- Fox A.J.,
- et al.
- Rutgers D.R.,
- Klijn C.J.,
- Kappelle L.J.,
- van Huffelen A.C.,
- van der Grond J.
- Anzola G.P.,
- Gasparotti R.,
- Magoni M.,
- Prandini F.