Author + information
- Received February 22, 2016
- Revision received June 3, 2016
- Accepted June 5, 2016
- Published online September 12, 2016.
- Ying-Hsien Chen, MDa,
- Weng-San Leong, MDb,
- Mao-Shin Lin, MDa,c,
- Ching-Chang Huang, MDa,
- Chi-Sheng Hung, MDa,
- Hung-Yuan Li, MDa,
- Kok-Kheng Chan, MDd,
- Chih-Fan Yeh, MDa,
- Ming-Jang Chiu, MD, PhDe and
- Hsien-Li Kao, MDa,∗ ()
- aDepartment of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
- bPantai Hospital Ipoh, Malaysia
- cGraduate Institute of Clinical Medicine, Medical College, National Taiwan University, Taipei, Taiwan
- dKPJ Penang Specialist Hospital, Penang, Malaysia
- eDepartment of Neurology, National Taiwan University Hospital, Taipei, Taiwan
- ↵∗Reprint requests and correspondence:
Dr. Hsien-Li Kao, Department of Internal Medicine, National Taiwan University Hospital, 7 Chung-Shan South Road, 100 Taipei, Taiwan.
Objectives This study sought to determine predictors for successful endovascular treatment in patients with chronic carotid artery occlusion (CAO).
Background Endovascular recanalization in patients with chronic CAO has been reported to be feasible, but technically challenging.
Methods Endovascular attempts in 138 consecutive chronic CAO patients with impaired ipsilateral hemisphere perfusion were reviewed. We analyzed potential variables including epidemiology, symptomatology, angiographic morphology, and interventional techniques in relation to the technical success.
Results The technical success rate was 61.6%. Multivariate analysis showed absence of prior neurologic event (odds ratio [OR]: 0.27; 95% confidence interval [CI]: 0.10 to 0.76), nontapered stump (OR: 0.18; 95% CI: 0.05 to 0.67), distal internal carotid artery (ICA) reconstitution via contralateral injection (OR: 0.19; 95% CI: 0.05 to 0.75), and distal ICA reconstitution at communicating or ophthalmic segments (OR:0.12; 95% CI: 0.04 to 0.36) to be independent factors associated with lower technical success. Point scores were assigned proportional to model coefficients, and technical success rates were >80% and <40% in patients with scores of ≤1 and ≥4, respectively. The c-indexes for this score system in predicting technical success was 0.820 (95% CI: 0.748 to 0.892; p < 0.001) with a sensitivity of 84.7% and a specificity of 67.9%.
Conclusions Absence of prior neurologic event, nontapered stump, distal ICA reconstitution via contralateral injection, and distal ICA reconstitution at communicating or ophthalmic segments were identified as independent negative predictors for technical success in endovascular recanalization for CAO.
Carotid artery occlusion (CAO) is associated with a 6% to 20% annual risk of recurrent ipsilateral ischemic stroke despite intensive medical treatment (1,2), and surgical bypass offers no benefit in preventing stroke (3–5). Feasibility and midterm results of endovascular treatment of chronic CAO have been reported (6,7), successful recanalization restores cerebral perfusion and may improve neurocognitive function (8–10). However, CAO recanalization is technically challenging and its potential complication may be life threatening. Therefore, the acceptance and dissemination of this procedure has been slow. We present a retrospective analysis of CAO recanalization attempts to identify predictors for technical success; a scoring system was constructed to facilitate better case selection for beginning operators.
We conducted a retrospective analysis of consecutive CAO recanalization attempts from October 2004 to April 2015 in National Taiwan University Hospital and affiliated hospitals. CAO was detected by Doppler ultrasound, computed tomography, magnetic resonance imaging, or conventional angiography. Brain computed tomography perfusion with acetazolamide (Diamox) stress was performed before carotid intervention to document ipsilateral hemisphere perfusion abnormality in all patients, with method as described previously (10). Prior neurologic symptoms were defined as ipsilateral transient ischemic attack or ischemic stroke or amaurosis fugax. The duration from the last neurologic event to intervention was recorded and categorized as ≤6 months, >6 months, or no history of neurologic event before the diagnosis of CAO. Endovascular recanalization attempts were made after obtaining informed consent. Clinical and neurologic data, angiography findings, and interventional results were collected and reviewed by independent neurologist and interventionist. The retrospective review of the clinical information and radiologic records of the patients were approved by the Institutional Review Board at National Taiwan University Hospital.
Selective cerebral angiography was done via femoral approach before intervention. Pseudo-occlusion was ruled out using criteria described in the literature (11,12). Angiographic criteria of true occlusion were: 1) discontinuation of lumen >5 mm in length; 2) grade 0 Thrombolysis In Cerebral Infarction (TICI) antegrade flow distal to the occlusion; and 3) established collateral filling to the ipsilateral intracranial internal carotid artery (ICA) territory, either via anterior communicating artery, posterior communicating artery, ipsilateral ophthalmic artery (OA), or other brachiocephalic artery branches. Diameter stenosis was calculated using the NASCET (North American Symptomatic Carotid Endarterectomy Trial) method (13).
All interventions were performed via an 8-F femoral sheath. Aspirin 100 mg and clopidogrel 75 mg daily for at least 7 days were given before the procedure. Heparin was given to maintain activated clotting time within 200 to 250 s. The target common carotid artery (CCA) was engaged with 8-F JR 4 guiding catheter. Intraluminal wiring using coronary guidewires and microcatheters has been described (6) as well as the alternative subintimal tracking with antegrade re-entry technique (14). Wiring was abandoned after 30 minutes of futile effort, consumption of >300 ml of contrast, or when the wire tip is confirmed to be extravascular.
Once wire enters the distal true lumen, the microcatheter was exchanged to a 1.5-mm diameter coronary balloon for pre-dilation. Distal embolic protection device would be deployed if an adequate landing zone can be identified. Properly sized balloon expandable stents (for segments in and above carotid canal) and self-expanding stents (for cervical ICA) were then deployed to scaffold the occlusion. Balloon post-dilation may be done if stent expansion was not adequate. The intervention was considered a technical success if the occlusion segment was stented with final residual diameter stenosis of ≤20%, and establishing grade 3 antegrade TICI flow.
Reading of angiograms
An independent interventionist reviewed the procedural angiograms off-line, and morphologic characteristics regarding the stump, occlusion segment, and distal ICA reconstitution were recorded. A stump was present if there was contrast filling within the segment of cervical ICA after it bifurcates from the CCA, proximal to the occluded segment. Stump angulation was measured using CCA as the reference axis at lateral view, and categorized into >45° or ≤45°. It would be designated as >45° for analysis if stump was absent. The occlusion site was categorized as CCA, cervical ICA, or intracranial ICA. Distal carotid visibility was defined as the presence of contrast reconstitution in the ICA distal to the occlusion, either during selective ipsilateral, contralateral, or vertebral injection. The most proximal level of distal carotid artery reconstitution was categorized as at petrous segment or below, cavernous segment, clinoid segment, ophthalmic segment, and communicating segment. Reversed OA flow was defined as the presence of contrast flow in OA with reversed flow direction during ipsilateral injection. The occlusion length was measured from the occlusion site to the distal reconstituted ICA in lateral projection, in straight line ignoring potential curvature of the occluded segment. The occlusion length was categorized as ≤50 or >50 mm. If the distal ICA was not visible, the occlusion length would be designated as >50 mm.
All continuous variables were expressed as mean ± SD, and categorical variables in numbers and percentage. The chi-square test or Fisher's exact test (if the group's number is 5 or less) was used to compare groups of categorical data. A 2-sided p value of 0.05 was considered statistically significant. Stata/SE 11.0 for Windows (StataCorp LP, Texas) was used for statistical analyses. Sensitivity and specificity were measured by the Youden index.
A total of 138 consecutive CAO recanalization attempts were performed in 138 patients (115 men; age 66.7 ± 9.7 years; range: 41 to 86 years) with overall technical success rate 61.6%. Tables 1 and 2 summarized the demographics and lesion characteristics. Among 138 ICA occlusion patients, 43.5% of the patients did not have history of neurologic events and 47.8% received recanalization attempts within 6 months from the last neurologic events. During cerebral angiography, calcification was present in 37.0% of the occlusions, and tapered stump (76.1%) with a stump angle of >45° (18.8%) was seen in the majority. An occlusion length of >50 mm was seen in 86.2%. Distal carotid artery can be visualized by contralateral CCA injection in 23.2%, and 63.8% with reversed OA flow.
Hyperpefusion syndrome with delayed nonfatal intracranial hemorrhage developed in 2 patients after successful recanalization. Another 2 patients suffered from subarachnoid hemorrhage and intracranial hemorrhage related to wire extravasation; 1 died after craniectomy. Another mortality occurred in a patient with history of neck radiation for nasopharyngeal cancer, due to delayed massive paranasal sinus bleeding possibly related to wire trauma. There is 1 subacute stent thrombosis with stroke. Therefore, 30-day procedural complication rate including stroke, intracranial hemorrhage, and death was 4.3% (6 of 138), and overall 30-day mortality was 1.4% (2 of 138).
Univariate analysis was performed with logistic regression after adjusting age and sex to determine predictors for technical success (Table 3). Success rate was lower in occlusions without prior neurologic event (OR: 0.45; 95% CI: 0.22 to 0.96), nontapered stump (OR: 0.33; 95% CI: 0.13 to 0.82), occlusion site at intracranial ICA (OR: 0.02; 95% CI: 0.001 to 0.50), distal ICA reconstitution via contralateral injection (OR: 0.19; 95% CI: 0.08 to 0.44), with the level of reconstitution at ophthalmic segment (OR: 0.04; 95% CI: 0.01 to 0.20) and at communicating segment (OR: 0.03; 95% CI: 0.01 to 0.17). The technical success rate was greater in patients with hyperlipidemia (OR: 2.10; 95% CI: 1.03 to 4.28), and duration from last neurologic event ≤6 months (OR: 2.2; 95% CI: 1.09 to 4.48).
Multivariate analysis using stepwise backward and forward method was then performed (Table 4). No history of neurologic event (OR: 0.27; 95% CI: 0.10 to 0.76), nontapered stump (OR: 0.18; 95% CI: 0.05 to 0.67), distal ICA reconstitution with contralateral injection (OR: 0.19; 95% CI: 0.05 to 0.75), or level of distal carotid artery reconstitution at communicating or ophthalmic segments (OR: 0.12; 95% CI: 0.04 to 0.36) were identified as independent negative predictors for technical success in CAO recanalization.
CAO scoring system
Using the independent predictors obtained from multivariate analysis, a scoring system can be created to predict success rate of CAO recanalization. Points were assigned to each variable proportional to its regression coefficients rounded to the nearest integer. Table 5 summarized the actual score points assigned, and the success rates of different total score points. For lesions with score of ≤1, the technical success rate was >80%. If the score was ≥3, then the success rate decreased significantly to <40%. The c-index on the basis of area under the curve for this scoring system in predicting technical success was 0.820 (95% CI: 0.748 to 0.892; p < 0.001), with a sensitivity of 84.7% and a specificity of 67.9% (Figure 1).
Example case 1
A 67-year-old man presented with transient left upper limb weakness 4 months before CAO recanalization (score point 0). Angiography showed right ICA occlusion with blunt stump morphology (2 points) (Figure 2A). Distal carotid artery was visible via contralateral contrast injection (2 points), with level of distal ICA reconstitution at communicating segment (2 points) (Figure 2B). The total CAO score for this lesion was 6, and thus the success rate of recanalization would be estimated low (20%).
Example case 2
A 51-year-old man presented with left limbs weakness 2 years before CAO recanalization (0 point). Angiography showed right ICA occlusion with tapered stump morphology (0 point) (Figure 2C). Distal carotid artery was visible with ipsilateral contrast injection (0 point) with level of distal ICA reconstitution at cavernous segment (0 point) (Figure 2D). The total CAO score for this lesion was 0, and thus the success rate of recanalization would be high (>90%).
Chronic CAO is associated with annual stroke rates of 6% to 20%, despite intensive medical treatment (1,2). Although several nonrandomized extracranial-intracranial bypass studies have shown positive results (15–17), both the milestone EC-IC Bypass trial and Carotid Occlusion Surgery Study (3–5) failed to demonstrate any difference in fatal and nonfatal strokes at 2 to 5 years between surgical and medical groups in symptomatic CAO patients. The high periprocedural complications seen in the surgical group is mainly due to the intolerance of procedural hemodynamic fluctuation (18).
Endovascular recanalization of CAO, in contrast, may offer the same reperfusion advantage with less procedural hemodynamic compromise. Successful recanalization of chronic CAO patients may improve global cognitive function, as well as attention and psychomotor processing speed, compared with medical treatment (8,10,19). Unfortunately, endovascular recanalization of CAO is technically challenging, because the occlusion length is usually long with wide individual variation of the occluded vessel segment course. Visual clues for wiring across the occlusion, such as bridging collateral or distal artery reconstitution, are often lacking. Furthermore, potential complications after wiring injury, including hemorrhage, pseudoaneurysm, and carotid-cavernous fistula (20,21), may be catastrophic and deterred inexperienced operators from embarking on the procedure. A systematic pre-procedural evaluation is therefore important to identify patient and lesion characteristics that carry higher success rates, and thus are helpful for the dissemination of this procedure.
It is intuitively reasonable that the duration from last event may affect technical success rate. As an acute occlusion “ages,” its consistency may become hard and calcified, and its length will increase progressively with a thrombotic process. Both of these mechanisms have been observed in coronary artery occlusions (22), increasing the difficulty of endovascular recanalization. When CAO develops insidiously without a clear neurologic event, such “clinically silent” occlusions may have actually progressed over years and decades before the recanalization procedure was attempted. The resulting firm, calcified, and long occlusion segment is therefore more resistant to endovascular intervention. This is clearly shown in our analysis.
A tapered stump at the occlusion facilitates wire entry into the occluded segment, whereas a blunt stump or absence of stump increases the difficulty. This is well-demonstrated in our study. Because the majority of CAO start at the proximal cervical ICA just distal to the CCA bifurcation, less angulation between the CCA and ICA axes may also enhance forward force delivery and wire entry. However, stump angulation was shown to be insignificant in predicting technical success in our study.
Occlusion length was known to affect success rate in endovascular recanalization for chronic total coronary artery occlusion (23). In theory, longer occlusion in CAO should also predict lower technical success. Wiring across a long CAO is difficult due to the variable vessel course, easily results in false lumen creation and a higher chance of vessel injury. Arbitrarily, we categorize occlusion length by the cutoff of 50 mm, and the technical success rates in short and long occlusions are 73.7% and 59.7%, respectively. The occlusion length was found insignificant in predicting technical success (OR: 0.53; 95% CI: 0.17 to 1.59), but only 19 CAO (13.8%) in the current series were shorter than 50 mm. A larger case number would be necessary to clarify whether an inverse relationship do exist between the occlusion length and success rate of CAO recanalization.
Because severe calcification in the occluded segment may hinder guide wire crossing, it has been established as a negative predictor for success in coronary intervention for chronic occlusions (23). Interestingly, the success rate in occlusions with calcification in our analysis was in fact higher (calcified 68.6% vs. noncalcified 57.5%), although not statistically significant (OR: 1.60; 95% CI: 0.75 to 3.44). The potential reason is that calcification may actually provide a roadmap, and thus facilitate wiring procedures in the highly variable cervical ICA anatomy. In addition, most of the dense calcium burden in the current series was located at the proximal occlusion site, but not throughout the entire length of CAO. Once the proximal cap was penetrated with the devices described, the rest of the occlusion segment with loose calcification may be then easily traversed using the subintimal tracking with antegrade re-entry technique. Unlike in coronary interventions where major side branches should be preserved to maintain long-term patency, subintimal tracking with antegrade re-entry technique may be used more liberally in CAO intervention because there is no major side branch proximal to OA.
Visualization of the distal ICA by ipsilateral contrast injection not only provides clear reference for wiring procedure, but because the most frequent ipsilateral collateral route is the reversed OA flow, it also implies lumen patency from the OA takeoff to the terminal ICA bifurcation. When the distal ICA is reconstituted at levels distal to the ophthalmic segment, the difficulty of guide wire manipulation increases significantly. The operator has to negotiate the wire through tortuous petrous/cavernous/clinoid segments of the occluded ICA, and also achieve wire re-entry into true lumen in an intradural position. The potential consequences of vessel injury or perforation are prohibitive for aggressive wire choice or manipulation. In fact, the success rates when distal ICA reconstitution at segments of petrous or below, cavernous, clinoid, ophthalmic, and communicating or above, were 93%, 80%, 73%, 33%, and 29%, respectively, rendering the level of distal ICA reconstitution at ophthalmic or communicating segments predictive for a significantly lower rate of success (OR: 0.12; 95% CI: 0.04 to 0.36).
The construction of a scoring system based on our data can be used as a predictive tool to estimate the technical success rate of endovascular recanalization for individual chronic CAO. It will be especially important and useful for beginners of the procedure, so that cases with less likelihood of success may be avoided. In addition, it also provides a basis for the comparison of complexity and difficulty of CAO cases in studies or discussions. This CAO scoring system should and will be used prospectively in future studies to validate its predictive power.
The case number of the present study is relatively small, and this may obscure relevant factors and undermine the predictive power of the scoring system. More prospective patients are needed to validate current findings in the future. The present analysis only looked at factors associated with technical success, without mention of complication and long-term clinical outcome. Although these CAO patients with objective perfusion abnormality may enjoy improved neurocognitive function after successful recanalization (8,10,19), this potential benefit should be proven to outweigh the risk of procedural complication in further investigations.
No history of neurologic event, nontapered stump, distal ICA reconstitution by contralateral injection, and distal ICA reconstitution at communicating or ophthalmic segments were identified as independent negative predictors for technical success in endovascular treatment for CAO.
WHAT IS KNOWN? Endovascular recanalization of chronic carotid artery occlusion is technically challenging.
WHAT IS NEW? We identified predictors for procedural success and constructed a scoring system based on patient/lesion characteristics. This would help interventionists in their initial learning curves to gauge and select appropriate cases.
WHAT IS NEXT? Further validation of the scoring system in a larger prospective cohort is mandatory in the future.
The authors acknowledge statistical assistance provided by the Taiwan Clinical Trial Statistical Center, Training Center, and Pharmacogenomics Laboratory (which is founded by National Research Program for Biopharmaceuticals [NRPB] at the Ministry of Science and Technology of Taiwan; MOST 104-2325-B-002-032) and the Department of Medical Research in National Taiwan University Hospital.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Chen and Leong contributed equally to this work.
- Abbreviations and Acronyms
- carotid artery occlusion
- common carotid artery
- confidence interval
- internal carotid artery
- ophthalmic artery
- odds ratio
- Received February 22, 2016.
- Revision received June 3, 2016.
- Accepted June 5, 2016.
- 2016 American College of Cardiology Foundation
- Klijn C.J.,
- Kappelle L.J.,
- Tulleken C.A.,
- van Gijn J.
- Flaherty M.L.,
- Flemming K.D.,
- McClelland R.,
- Jorgensen N.W.,
- Brown R.D. Jr..
- Kao H.L.,
- Lin M.S.,
- Wang C.S.,
- et al.
- Thomas A.J.,
- Gupta R.,
- Tayal A.H.,
- Kassam A.B.,
- Horowitz M.B.,
- Jovin T.G.
- Lin M.S.,
- Chiu M.J.,
- Wu Y.W.,
- et al.
- Siddiqui A.H.,
- Hopkins L.N.
- Huang C.C.,
- Chen Y.H.,
- Lin M.S.,
- et al.
- O'Leary D.H.,
- Mattle H.,
- Potter J.E.
- Michael T.T.,
- Papayannis A.C.,
- Banerjee S.,
- Brilakis E.S.
- 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.
- Lin M.S.,
- Lin L.C.,
- Li H.Y.,
- et al.
- Wu C.K.,
- Luo J.L.,
- Lin M.S.,
- Kao H.L.
- Stone G.W.,
- Kandzari D.E.,
- Mehran R.,
- et al.
- Morino Y.,
- Abe M.,
- Morimoto T.,
- et al.