Short-Term Outcomes After Symptomatic Internal Carotid Artery Occlusion

Introduction

The clinical consequences of acute internal carotid artery (ICA) occlusion can range from being completely asymptomatic to severe stroke.¹ Patients presenting symptomatically with an occluded ipsilateral ICA have a poor prognosis in the long term.^2–5 For those who present with transient or minor ischemic stroke symptoms, the annual risk of subsequent stroke has been determined to be 5% to 7% and the annual mortality rate 6%.³ For patients with more severe strokes, the incidence of recurrent stroke is 10% and mortality 45% after an average follow-up of 1.2 years.⁵ The short-term prognosis of patients with symptomatic ICA occlusion has been less frequently studied.^1,5,6 Available studies have had relatively small numbers of patients with limited evaluations of short-term outcomes. Thus, it remains unclear if ICA occlusion is an independent predictor of outcomes such as stroke recurrence, neurological worsening, and functional outcome.

The purpose of the present study was to evaluate the short-term outcomes of patients presenting to hospital with symptomatic ICA occlusion compared with patients with varying degrees of ICA stenosis in a large, multicenter cohort of patients with acute ischemic stroke/transient ischemic attack (TIA).

Methods

We used data from the Registry of the Canadian Stroke Network (RCSN). The RCSN is a hospital-based registry of consecutive patients with acute stroke presenting to 11 stroke centers in Ontario, Canada.^7,8 All patients in this registry were identified prospectively and data were collected throughout their time in the hospital by trained research nurses using a standardized case record form and custom data entry software.⁸ Approval for the RCSN was obtained from the Research Ethics Board at each of the participating centers and approval for this analysis was obtained through the Publications Committee of the RCSN.

For this study, we included patients admitted to the hospital with an ischemic stroke or TIA in the anterior circulation between July 1, 2003, and March 31, 2008. The following patients were excluded: (1) patients with vertebrobasilar symptoms as defined by the Oxford Community Stroke Project classification of “posterior circulation stroke”⁹; (2) patients with presumed cardioembolic stroke etiology, defined as the presence of atrial fibrillation (either from the medical history or an electrocardiogram during the hospital admission), and/or a final diagnosis of cardioembolic stroke (documented in the hospital discharge summary); (3) patients with a clearly documented nonatherosclerotic etiology (arterial dissection, vasculitis, prothrombotic state, or cortical vein/sinus thrombosis); and (4) patients without vascular imaging data.

The resulting cohort was divided into 4 groups based on angiography (either CT, MR, or conventional angiography) or carotid Doppler findings for the symptomatic ICA: (1) ICA occlusion; (2) severe (70% to 99%) ICA stenosis; (3) moderate (50% to 69%) ICA stenosis; and (4) mild to no (<50%) ICA stenosis. The degree of carotid stenosis on which the study groups were defined was based on the data definitions for ICA stenosis in the RCSN database operations manual. The ICA occlusion group was deemed the comparison group. The flow chart for cohort creation is presented in the Figure.

• Download figure
• Download PowerPoint

The following characteristics were collected at baseline: age, gender, history of hypertension, diabetes mellitus, peripheral vascular disease, smoking status, previous stroke or TIA, previous myocardial infarction, previous carotid endarterectomy or stenting, preadmission acetylsalicylic acid, warfarin, and antihypertensive medications, time from last seen normal to emergency department arrival, admission random glucose, creatinine, international normalized ratio, systolic and diastolic blood pressure, Charlson comorbidity index, preadmission functional status (independent or dependent), Canadian Neurological Scale,¹⁰ level of consciousness, index event (ischemic stroke or TIA), and contralateral ICA stenosis/occlusion. We also collected data on in-hospital treatments received: recombinant tissue plasminogen activator thrombolysis, acetylsalicylic acid, warfarin, heparin, statin agents, antihypertensive medications, carotid endarterectomy, and carotid angioplasty and/or stenting. At participating centers, intravenous heparin is conventionally used for full anticoagulation and subcutaneous heparin for deep vein thrombosis prophylaxis. However, the specific indications, formulations, and dosages of heparin were not captured in the RCSN database.

The following in-hospital outcome data were collected: neurological complications (recurrent stroke, seizure, and neurological worsening), myocardial infarction, death, modified Rankin Scale score¹¹ on hospital discharge, discharge destination (home, rehabilitation, nursing home, other hospital), and length of hospital stay (dichotomized to 1 to 7 days versus ≥8 days). Poor functional outcome was defined as a modified Rankin Scale score 3 to 6.

Statistical Analyses

Statistical analyses were performed using a commercially available software package (SAS Version 9.1.3 statistical software; SAS Institute Inc, Cary, NC). Baseline characteristics were summarized using descriptive statistics comparing the 4 groups of patients with various degrees of ICA stenosis/occlusion. Categorical variables were analyzed using the χ² test. Means and medians were compared using 1-way analysis of variance and Kruskal-Wallis test, respectively. A probability value <0.05 was considered statistically significant. Mean values were presented with SD and median values with interquartile range. No comparisons were made when a cell value was ≤5 to protect the privacy of patients in the database based on privacy policies at the Institute for Clinical Evaluative Sciences where the RCSN database is held. Logistic regression modeling was used to evaluate the association between the degree of stenosis/occlusion of the symptomatic ICA and outcomes described previously and included the following baseline variables: age, gender, stroke symptoms (weakness, aphasia, visual field defects, or brain stem/cerebellar signs), stroke severity (Canadian Neurological Scale score), time from last seen normal to emergency arrival, level of consciousness, Charlson comorbidity index, contralateral ICA patency, random glucose, creatinine, international normalized ratio, systolic and diastolic blood pressure, and treatment with recombinant tissue plasminogen activator. The Charlson comorbidity index was included in the regression models to adjust for patients' baseline level of comorbid illness. Variables were entered into regression models concurrently and were selected based on clinical relevance as factors associated with stroke outcome. Results of univariate and multivariate logistic regression analyses were presented as ORs with 95% CIs. A 95% CI that did not include 1.0 was considered statistically significant. Length of stay analysis included only patients who survived through hospital discharge.

Results

Of the 13 461 consecutive patients who were admitted with a final diagnosis of ischemic stroke or TIA, 4144 patients met study criteria (Figure). Baseline characteristics and in-hospital treatments are presented in Tables 1 and 2, respectively. The mean age was 71 (SD 13.1) years and 45% of patients were female. In total, 82.5% of patients had an index stroke (17.5% had TIA). The symptomatic ICA was occluded in 283 (6.8%) patients, showed a severe stenosis in 414 (10%) patients, a moderate stenosis in 409 (9.9%) patients, and a mild or no stenosis in 3038 (73.3%) patients (Figure). Among the 4 patient groups, significant differences were observed in baseline characteristics (Table 1) and in-hospital treatments (Table 2). Initial stroke severity was highest for the patients with ICA occlusion (P<0.0001).

Outcomes are presented in Tables 3 and 4. Compared with patients with symptomatic ICA occlusion, severe stenosis was associated with a lower risk of in-hospital death (adjusted OR, 0.40; 95% CI, 0.20 to 0.79), neurological complications (adjusted OR, 0.54; 95% CI, 0.36 to 0.80), neurological worsening (adjusted OR, 0.52; 95% CI, 0.34 to 0.78), poor functional outcome (adjusted OR, 0.62; 95% CI, 0.41 to 0.94), and greater odds of being discharged home (adjusted OR, 1.54; 95% CI, 1.03 to 2.31). The association between severe stenosis and recurrent stroke (versus the ICA occlusion group) was not statistically significant (adjusted OR, 0.78; 95% CI, 0.40 to 1.57) and patients with severe stenosis were less likely to have a length of stay ≤7 days (adjusted OR, 0.57; 95% CI, 0.39 to 0.84). In the comparisons between ICA occlusion and moderate stenosis or mild/no stenosis, either degree of stenosis was associated with a lower risk of recurrent stroke, death, neurological complications, neurological worsening, and poor functional outcome and more likely with discharge home (Table 4).

Discussion

Our study showed that patients with symptomatic ICA occlusion were more likely to have in-hospital death, neurological worsening, and poor functional outcome and were less likely to be discharged home compared to patients with severe, moderate, or mild/no stenosis. Recurrent in-hospital stroke was less likely for patients with moderate or mild/no stenosis and not significantly different for those with severe stenosis when compared with ICA occlusion.

In our cohort of 283 patients with symptomatic ICA occlusion, recurrent in-hospital stroke occurred in 6.7% of patients, myocardial infarction in 2.5%, and mortality in 12% with an average length of stay of 18 days. In a previous study of 75 symptomatic patients with ICA occlusion, the incidence of stroke, myocardial infarction, and mortality at 30 days follow-up was 8%, 0%, and 7%, respectively.⁶ Our risk of stroke was similar, although mortality was higher, possibly reflecting methodological differences, including inclusion criteria related to stroke etiology and index event timing as well as the vascular imaging modalities used for ICA patency definitions. Another study of 177 patients with ischemic stroke and ICA occlusion reported a higher 30-day mortality of 30%, but that study excluded patients with TIA.⁵ In our study, we assessed additional short-term outcomes, including neurological worsening, poor functional status, and discharge to destinations other than home, that have not previously been studied in this patient population. In our cohort, a high proportion of patients with ICA occlusion had these adverse outcomes.

To reduce early recurrent stroke risk after acute ICA occlusion, it is possible that earlier, more intensive antiplatelet or anticoagulant therapy might be warranted. In our cohort, antiplatelet therapy use in the hospital was similar in all patient groups, but the timing of initiation was not captured. Warfarin was also used, but the indications for its use were not captured. Early initiation of antiplatelet agents after ischemic stroke has been shown to reduce the risk of recurrent stroke, death, and dependency.¹² However, studies have not specifically evaluated patients with symptomatic ICA occlusion and it remains unknown what is the optimal medical management of these patients. Patients with ICA occlusion may also have hemodynamic compromise and thus are at a higher risk of subsequent stroke.¹³ Therefore, patients with an ICA occlusion and hemodynamic compromise might benefit from urgent revascularization procedures such as acute thrombectomy, angioplasty with stenting, or extracranial/intracranial bypass. However, these interventions would only be considered for acute occlusions in the setting of fresh thrombus, and it is unclear if the potential benefits outweigh the risks.

Our study has some limitations. First, this is a clinical database analysis with data collected by chart review. To improve data reliability, RCSN variables were defined in an operations manual and the data entry software was designed to reduce errors. In addition, study nurses were trained and evaluated on chart abstraction. An important limitation is that RCSN variables were used to identify the study cohort. We attempted to create a homogenous cohort of patients with presumed ICA atherosclerotic stroke because atherosclerosis is by far the most common cause of ICA occlusion.^5,14,15 However, our methodology may have resulted in ascertainment bias and some patients with atherosclerotic disease may have been missed and other patients incorrectly included. Second, the vascular imaging modality used was not uniform, because it depended on participating hospital resources and treating physician preference. Carotid Doppler imaging, used as the only imaging modality in 68% of the cohort, may not have sufficiently distinguished high-grade stenosis from occlusion.¹⁶ Third, we could not determine whether patients presented with an acute or chronic ICA occlusion, which may have different outcomes. Fourth, a few patients with ICA occlusion or mild/no stenosis were listed as receiving in-hospital endarterectomy, angioplasty, or stenting. We have 3 possible explanations for this: (1) the intervention may have been on the side contralateral to the occluded ICA; (2) the intervention may have been completed for a patient-specific indication; or (3) chart data abstraction was not accurate (for example, a patient was coded as having an intervention, but the intervention was aborted once ICA occlusion was confirmed). Fifth, the RCSN database does not effectively capture information on intracranial vascular stenosis that may have implications for outcomes in this patient population. In addition, the database does not capture the location of recurrent stroke events, limiting our ability to discern if a recurrent event occurred within the territory of the symptomatic carotid artery. Finally, the mild/no stenosis group likely included a heterogeneous group of patients with various stroke etiologies that may have different outcomes and may have affected the results. Still, our results showed that symptomatic ICA occlusion had worse outcomes than moderate or severe stenosis, which is a more relevant comparison.

In conclusion, results of this study demonstrate that patients with ischemic stroke or TIA with an ipsilateral ICA occlusion are at high risk of early recurrent stroke, poor functional outcome, and death. Given the potential poor short-term prognosis in this patient population, further studies are warranted to determine if additional investigations and alternative management strategies may improve outcomes.

Sources of Funding

The Registry of the Canadian Stroke Network is funded by an operating grant from the Ontario Ministry of Health and Long-Term Care. The results and conclusions are those of the authors and should not be attributed to any of the sponsoring or funding agencies. The funding agencies had no role in the design or conduct of the study or the collection, management, analysis, or interpretation of the data. The article was reviewed and approved by the publications committee of the Registry of the Canadian Stroke Network. M.D.I.V. is supported by an unrestricted grant from the Niels Stensen Foundation. R.H.S. receives salary support from the Heart and Stroke Foundation/Canadian Institutes of Health Research (CIHR)/Canadian Stroke Network “Focus on Stroke” Research Scholarship. M.K.K. is supported by a CIHR New Investigator Award as well as the Canadian Stroke Network and the University Health Network Women's Health Program. F.L.S. receives salary support from the Canadian Stroke Network.

Disclosures

None.

Footnotes