Journal of Thrombosis and Haemostasis, 12: 814–821

DOI: 10.1111/jth.12561

ORIGINAL ARTICLE

Improving prediction of recanalization in acute large-vessel occlusive stroke P . V A N A C K E R , * † D . L A M B R O U , * A . E S K A N D A R I , * P . M A E D E R , ‡ R . M E U L I , ‡ G . N T A I O S § and P. MICHEL* *Neurology Service, Centre Hospitalier Universitaire Vaudois and University of Lausanne, Lausanne, Switzerland; †Department of Neurology, University Hospital Antwerp, Edegem, Belgium; ‡Department of Radiology, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland; and §Department of Medicine, University of Thessaly, Larissa, Greece

To cite this article: Vanacker P, Lambrou D, Eskandari A, Maeder P, Meuli R, Ntaios G, Michel P. Improving prediction of recanalization in acute large-vessel occlusive stroke. J Thromb Haemost 2014; 12: 814–21.

Summary. Background: Recanalization in acute ischemic stroke with large-vessel occlusion is a potent indicator of good clinical outcome. Objective: To identify easily available clinical and radiologic variables predicting recanalization at various occlusion sites. Methods: All consecutive, acute stroke patients from the Acute STroke Registry and Analysis of Lausanne (2003–2011) who had a large-vessel occlusion on computed tomographic angiography (CTA) (< 12 h) were included. Recanalization status was assessed at 24 h (range: 12–48 h) with CTA, magnetic resonance angiography, or ultrasonography. Complete and partial recanalization (corresponding to the modified Treatment in Cerebral Ischemia scale 2–3) were grouped together. Patients were categorized according to occlusion site and treatment modality. Results: Among 439 patients, 51% (224) showed complete or partial recanalization. In multivariate analysis, recanalization of any occlusion site was most strongly associated with endovascular treatment, including bridging therapy (odds ratio [OR] 7.1, 95% confidence interval [CI] 2.2–23.2), and less so with intravenous thrombolysis (OR 1.6, 95% CI 1.0–2.6) and recanalization treatments performed beyond guidelines (OR 2.6, 95% CI 1.2–5.7). Clot location (large vs. intermediate) and tandem pathology (the combination of intracranial occlusion and symptomatic extracranial stenosis) were other variables discriminating between recanalizers and non-recanalizers. For patients with intracranial occlusions, the variables significantly associated with recanalization after 24 h were: baseline National Institutes of Health Stroke Scale (NIHSS) Correspondence: Peter Vanacker, Centre Hospitalier Universitaire Vaudois, 46, Rue de Bugnon, CH-1011 Lausanne, Switzerland. Tel.: +41 21 314 11 11; fax: +41 21 314 13 03. E-mail: [email protected] Received 19 November 2013 Manuscript handled by: F. R. Rosendaal Final decision: F. R. Rosendaal, 8 March 2014

(OR 1.04, 95% CI 1.02–1.1), Alberta Stroke Program Early CT Score (ASPECTS) on initial computed tomography (OR 1.2, 95% CI 1.1–1.3), and an altered level of consciousness (OR 0.2, 95% CI 0.1–0.5). Conclusions: Acute endovascular treatment is the single most important factor promoting recanalization in acute ischemic stroke. The presence of extracranial vessel stenosis or occlusion decreases recanalization rates. In patients with intracranial occlusions, higher NIHSS score and ASPECTS and normal vigilance facilitate recanalization. Clinical use of these predictors could influence recanalization strategies in individual patients. Keywords: mechanical thrombolysis; neuroimaging; stroke; tissue-type plasminogen activator; treatment outcome.

Introduction In patients with acute ischemic stroke and large-vessel occlusion, recanalization has a major impact on clinical outcome if performed early [1,2], particularly in patients with a favorable penumbra/core ratio [3–5]. Recanalization of occluded large arteries, restoration of blood flow and prevention of infarct expansion may occur spontaneously in a subset of patients, but is more effectively achieved by intravenous thrombolysis (IVT), endovascular mechanical and/or thrombolytic treatment, or combinations thereof [1,6–10]. However, recanalization with an acute intervention is not always associated with reperfusion and better clinical outcome [11,12], given that multiple other factors (time, collaterals, core and penumbra volumes, intermediate occlusions, site of ischemia, metabolic states, genetics, etc.) also have effects. Currently, milder baseline stroke deficit, elevated systolic blood pressure, normal glucose values at admission, smoking, an absence of atrial fibrillation, distal vessel occlusion and thrombus length have been shown to predict recanalization [13–22]. However, some of the studies assessing recanalization were performed in small cohorts, performed in specific subgroups of patients © 2014 International Society on Thrombosis and Haemostasis

Predictors of recanalization in stroke 815

(IVT alone, endovascular treatment alone, or combined), or performed without appropriate logistic regression analysis or adjustment for occlusion site. Our objective was to identify independent predictors for recanalization among multiple demographic, clinical, metabolic and radiologic variables and in a large consecutive series of acute ischemic stroke patients, independently of the applied treatment strategy. The study was performed in a mixed cohort of untreated, IVT-treated and endovascular-treated stroke patients. By using such a study design, we hoped to detect clinicoradiologic covariates influencing the 24-h recanalization rates independently of the acute stroke management. We specifically paid attention to occlusion site, multilevel occlusions, and treatment strategies applied.

room for suspected acute ischemic stroke. On acute noncontrast CT, early ischemic changes were considered to be present if the acute Alberta Stroke Program Early CT Score (ASPECTS) (or pc-ASPECTS, for posterior circulation strokes) was < 10 [26,27]. Whenever indicated according to guidelines from the European Stroke Organization, IVT was performed. Endovascular treatment (intra-arterial thrombolysis or mechanical thrombectomy alone or in combination with IVT) was performed if the baseline NIHSS was ≥ 6 and treatment could be initiated within 6 h of symptom onset, according to the Swiss guidelines [28]. Patients receiving a recanalization treatment outside the recommended time windows were marked and analyzed separately. Analysis of recanalization

Patients and methods Study design and patient selection

The Acute STroke Registry and Analysis of Lausanne (ASTRAL) is the collection of all acute ischemic strokes in patients admitted to the stroke unit and/or intensive care unit of the Centre Hospitalier Universitaire Vaudois within 24 h after the last time when they were well [23]. For the current analysis, we selected all acute ischemic strokes in patients from January 2003 to June 2012 with: (i) acute computed tomographic angiography (CTA) performed within 12 h after the last time when patients were known to be well; (ii) arterial occlusion in cervical and/or cerebral arteries on CTA and in relation to the ischemic territory; and (iii) availability of a second arterial imaging finding (CTA, magnetic resonance angiography [MRA], or ultrasonography) at 24 h (range: 12–48 h) after stroke onset and allowing for assessment of recanalization. The 12-h time window was selected because significant salvageable brain tissue may still be present, and the 24-h recanalization assessment was selected because findings made at this time point are stronger predictors for a favorable outcome at 3 months than those made at an earlier assessment (< 2 h) [8,24,25]. The data collected included demographics, pre-existing cardiovascular risk factors and cerebrovascular events, comorbidities, clinical symptoms and signs, and metabolic and physiologic parameters at admission and 24 h later. National Institutes of Health Stroke Scale (NIHSS) scores were determined at admission, 6 and 24 h later, and after 7 days. An altered level of consciousness was considered to be present if the patient was obtunded or responded only with reflex motor or autonomic effects, or was totally unresponsive, flaccid, or areflexic. Clinical outcome at 3 months was assessed with the modified Rankin scale. Scores between 0 and 2 at 3 months were defined as a favorable outcome. Multimodal computed tomography (CT)-based imaging is the preferred initial imaging method in our emergency © 2014 International Society on Thrombosis and Haemostasis

Analyses of acute and subacute vascular imaging were performed both by a neurologist (P.M.) and by a neuroradiologist (R.M. or Ph.M.), using continuous axial and reformatted MIP pictures of extracranial (EC) and intracranial (IC) arteries. In case of discordance, a consensus was reached after an interdisciplinary discussion. Data on the interrater reliability between the different investigators were not collected. Occlusions in ischemic territory and non-ischemic territory were categorized according to their location (IC; EC; large IC; intermediate IC; anterior and posterior circulation). Large IC occlusion was defined as an occlusion in the trunk of the middle cerebral artery (M1), terminal occlusion of the IC internal carotid artery, or basilar artery occlusion. The group of intermediate IC occlusions contained occlusions in the anterior cerebral artery (A1 or A2 segments), the intermediate middle cerebral artery (M2), the posterior cerebral artery (P1 or P2 segments), the IC part of the vertebral artery (V4), and the siphon of the internal carotid artery without distal T-occlusion; the latter two were considered to be ‘intermediate’, because thrombus load and clinical symptoms are usually minor in the absence of extension into the basilar artery and the terminal segment of the carotid, respectively. Combined pathology was defined as the combination of an IC occlusion with EC occlusion or stenosis (> 50% according to the NASCET criteria) of an artery leading to the ischemic territory. A moderate to severe stenosis (NASCET > 50%) or occlusion in the EC part of the vessel leading to the ischemic territory was labeled as significant EC pathology. Occlusion extension was categorized as occlusion in a single vessel segment or in multiple vessel segments. In the ASTRAL, recanalization can be assessed with heterogeneous vascular imaging techniques (for detailed information, see supporting information Tables S1 and S2). In this study, only patients with an initial CTA and follow-up CTA, MRA or ultrasonography at 24 h were included. Recanalization of initially occluded arteries was classified as absent, partial, or complete.

816 P. Vanacker et al

Univariate logistic regression was performed to identify potential predictors of arterial occlusion, with all relevant clinical, radiologic and biological data included in the ASTRAL. Imputation of the missing values of the covariates studied was carried out with the method of chained equations. Multiple logistic regression analysis for the identification of predictors that significantly influence the probability of the absence of recanalization was performed as the combination of five imputed multiple analyses. Stepwise methods were implemented on each imputed dataset to identify significant main effects and interactions. In the multivariate logistic regression analysis, the level of significance was set at 95%. Statistical analysis was performed with R software (Stata, College Station, TX, USA). Collection, analysis and publication of data in the ASTRAL was approved by the ethics commission for research on humans of the Canton of Vaud, subcommission III.

tocol (IVT followed by mechanical thrombectomy). Thirty-six patients (8%) were treated outside current guidelines, as they were non-eligible for the standard stroke care within proven time windows. One-third were treated outside the proven time windows, according to an individualized, interdisciplinary decision. In this group, the majority were treated with endovascular mechanical thrombectomy outside a 6-h time window. The other two-thirds participated in research protocols, mostly of late imaging-based IVT by rt-PA or new fibrinolytic therapies, mechanical flow augmentation trial, and neuroprotective agents. The remaining patients (39%) received general supportive care in a comprehensive stroke center setting. Patients with initial large-vessel occlusion who were excluded because they did not undergo 24-h arterial reassessment had a similar profile regarding the demographics and admission NIHSS score, but they arrived significantly later at the hospital (median 267 min, IQR 177 min vs. median 201 min, IQR 156 min) and were less often thrombolyzed (21% vs. 48%).

Results

Predictors of any vessel recanalization

Statistical analysis

Baseline characteristics

During the observation period, 760 of 2765 consecutive stroke patients (27%) had a large-vessel occlusion in the ischemic territory on initial CTA (< 12 h after stroke onset). The main reasons for exclusion were: absence of a large-vessel occlusion (885 patients), vascular imaging beyond the 12-h window (672), and renal failure (185). To assess recanalization in the study population, goodquality vascular imaging (12–24 h) was necessary. In total, 320 of 760 patients were excluded by the absence of subacute imaging (189) or bad-quality imaging (131) (see supporting information Figure S1). There were several reasons why patients did not have a recanalization assessment at 24 h. First, in the initial years of the ASTRAL, less attention was paid to re-evaluating arterial status at 24 h in the study population. Second, in a subset of the study population, repeating vascular imaging was impossible, for medical and organizational reasons (contrast allergy, non-availability of appropriate imaging modality, etc.). A small group of 5.3% (40/2760) of our study population with initial arterial occlusion died within 24 h, and an unknown proportion of patients were in too poor a condition for vascular reassessment. The median age of patients was 67 years (interquartile range [IQR] 21), 46% were females (202/439), the median admission NIHSS score was 15.0 (IQR 12.0), and the median onset-to-treatment time was 2.6 h (IQR 1.7 h) for IVT performed after < 4.5 h. IVT with recombinant tissue-type plasminogen activator (rt-PA) was given within 4.5 h to 210 of these patients (48%), and endovascular treatment to 22 (5%). Two-thirds of the patients given endovascular treatment were treated with a bridging pro-

Partial or complete recanalization at 24 h was seen in 226 (51.5%) of all patients. Almost half of the recanalized large-vessel occlusions were found to be partially recanalized at 24 h (112/226 patients). The most relevant baseline data, clinical findings and radiologic findings for the recanalized and non-recanalized patient groups are shown in Table 1. In multivariate analysis, recanalization at any site of occlusion was associated with the acute stroke treatment and the vessel characteristics on acute CTA (Table 2). Patients with large-vessel occlusion who received IVT alone were more likely to have recanalization (odds ratio [OR] 1.64, 95% confidence interval [CI] 1.04–2.57). The highest likelihood of recanalization was found for endovascular treatment alone or in combination with IVT (OR 7.10, 95% CI 2.18–23.15). Patients treated outside current guidelines or in research protocols were also more likely to have recanalization (OR 2.58, 95% CI 1.16– 5.70). The occlusion site partially influenced the recanalization rate: recanalization was more likely to occur at both large (OR 2.27, 95% CI 1.14–4.48) and intermediate (OR 2.16, 95% CI 1.11–4.22) IC occlusion locations than at other (EC) locations, and the combination of EC and IC pathology decreased the chances of recanalization by > 50%. The area under the receiver operating characteristic (ROC) curve of this model for prediction of recanalization at any site was 0.72. Predictors of IC vessel recanalization

In multivariate analysis (Table 3), endovascular treatment clearly increased recanalization rates, and IVT showed a possible association (OR 1.30, P = 0.32) in the same © 2014 International Society on Thrombosis and Haemostasis

Predictors of recanalization in stroke 817 Table 1 Baseline data and radiologic findings, dichotomized according to complete or partial recanalization at any site vs. no recanalization

Demographics Age > 65 years* Female* Medical history Atrial fibrillation Smoking Presentation on arrival Hypertension Acute glucose values Baseline NIHSS score* Altered level of consciousness Anterior circulation stroke Stroke etiology (modified TOAST) Large-artery atherosclerosis Cardioembolism Dissection Undetermined cause Multiple causes Acute stroke intervention IVT Endovascular treatment** Recanalization treatment outside guidelines Time from symptom onset to IVT (min) Vessel occlusion characterization Extracranial carotid stenosis Extracranial carotid occlusion Significant extracranial pathology† Intracranial occlusion Large intracranial occlusion Intracranial carotid occlusion Significant intracranial pathology Combined pathology Occlusion extension (> 1 vessel)

Complete or partial recanalization (n = 226)

No recanalization (n = 213)

OR (95% CI)

124 (55) 104 (46)

120 (56) 98 (46)

0.94 (0.65–1.37) 1.00 (0.69–1.46)

74 (33) 54 (24)

58 (28) 48 (23)

1.29 (086–1.95) 1.10 (0.71–1.73)

128 (57) 6.8 mg/dl 16 (10) 36 (17) 190 (86)

125 (59) 6.6 mg/dl 14 (14) 51 (24) 168 (79)

0.92 0.92 1.02 0.63 0.60

(0.63–1.34) (0.84–1.01) (0.99–1.04) (0.39–1.01) (0.36–1.00)

37 99 26 52 8

(17) (45) (12) (22) (4)

66 67 28 36 13

(31) (32) (13) (18) (6)

1.78 2.64 1.66 2.65 1.10

(0.48–6.59) (1.58–4.39) (0.85–3.24) (1.44–4.87) (0.42–2.90)

121 18 23 163

(54) (8) (10) (105)

89 4 13 150

(42) (2) (6) (96)

2.27 7.52 2.96 1.00

(1.50–3.44) (2.43–23.29) (1.40–6.26) (0.99–1.00)

12 46 61 218 150 43 219 47 48

(5) (21) (27) (97) (67) (19) (97) (21) (21)

22 68 114 192 121 65 194 63 70

(10) (32) (54) (90) (57) (31) (91) (30) (33)

0.40 0.43 0.33 3.41 1.52 0.52 3.01 0.63 0.55

(0.19–0.84) (0.29–0.65) (0.22–0.49) (1.41–8.21) (1.03–2.24) (0.33–0.81) (1.15–7.87) (0.41–0.97) (0.36–0.85)

CI, confidence interval; IVT, intravenous thrombolysis; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; TOAST, trial of ORG 10172 in acute stroke treatment. All presented variables were significantly different in univariate comparisons, except for those marked with asterisks (*). Endovascular treatment includes both endovascular treatment alone and bridging of endovascular thrombectomy and IVT (**). Significant extracranial pathology was defined as a symptomatic, high-grade stenosis (NASCET>50%) of extracranial vessels (†).

Table 2 Results of the multivariable model for variables associated with complete or partial recanalization at any site at 24 h (12–48 h) OR (95% CI) Clinical findings Altered level of consciousness Treatment IVT (< 4.5 h) Endovascular treatment* Recanalization treatment outside current guidelines Vessel characteristics on CTA Significant extracranial stenosis in ischemic territory (50–100%) Large intracranial occlusion Intermediate intracranial occlusion

P-value

0.60 (0.35–1.03)

0.06

1.64 (1.04–2.57) 7.10 (2.18–23.15) 2.58 (1.16–5.70)

0.03 < 0.01 0.02

0.38 (0.25–0.58)

< 0.01

2.27 (1.14–4.48) 2.16 (1.11–4.22)

0.02 0.02

CI, confidence interval; CTA, computed tomographic angiography; IVT, intravenous thrombolysis; OR, odds ratio. *Including endovascular thrombectomy combined with IVT (‘bridging’). © 2014 International Society on Thrombosis and Haemostasis

direction. Additional EC vessel pathology decreased the chances of recanalization of IC occlusions, whereas localization of the occlusion within the IC circulation did not play a role. Patients with higher NIHSS scores at admission were more likely to have recanalization (OR 1.04, 95% CI 1.00–1.07 per 1 point of NIHSS), whereas an altered level of consciousness hindered it (OR 0.22, 95% CI 0.10–0.47). In contrast, patients with abnormal ASPECTS were less likely have recanalization (OR 1.13, 95% CI 1.02–1.26). The ROC curve of this model for prediction of recanalization of IC large-vessel occlusions was 0.73 as compared with absence of recanalization. Clinical outcome in the short term and long term

Complete or partial recanalization was associated with a median NIHSS score decrease from 16 (at admission) to 9 (after 24 h) in the recanalization group, and with a

818 P. Vanacker et al Table 3 Results of the multivariable model for variables associated with complete or partial recanalization of intracranial occlusion at 24 h (12–48 h)

Neurologic examination Altered level of consciousness Baseline NIHSS score Treatment IVT (< 4.5 h) Endovascular treatment* Recanalization treatment outside current guidelines Non-contrast CT Normal ASPECTS (10) Vessel characteristics on CTA Significant extracranial stenosis in ischemic territory (50–100%)

OR (95% CI)

P-value

0.22 (0.10–0.47) 1.04 (1.00–1.07)

< 0.01 0.03

1.30 (0.77–2.20) 4.14 (1.49–11.5) 2.35 (1.02–5.44)

0.32 < 0.01 < 0.05

1.13 (1.02–1.26)

0.03

0.37 (0.23–0.61)

< 0.01

ASPECTS, Alberta Stroke Program Early CT Score; CI, confidence interval; CT, computed tomography; CTA, computed tomographic angiography; IVT, intravenous thrombolysis; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio. *Including endovascular thrombectomy combined with IVT (‘bridging’).

decrease from 14 to 13 in the non-recanalization group. In unadjusted analysis, favorable clinical outcome at 3 months (OR 0.49, 95% CI 0.33–0.73) and 12 months (OR 0.45, 95% CI 0.30–0.68) was clearly worse in patients without recanalization. Recanalization did not influence the rate of recurrence of ischemic stroke and transient ischemic attack over the next 12 months (OR 0.73, 95% CI 0.36–1.46). Discussion In this study of 439 patients presenting with acute ischemic stroke and a large-vessel occlusion, we found 24-h recanalization rates to be associated with clinical findings, the intervention type, and radiologic features. The main clinical, independent predictor for the absence of IC recanalization is a decreased level of consciousness. This has not been described before. This observation may partially explain why level of consciousness is repeatedly found to be an independent predictor of poor outcome in patients with acute ischemic stroke [29]. This might reflect the thrombus size, with larger clots being more likely to produce a depressed level of consciousness. The predictive value of this clinical covariate is not only of significance in posterior circulation strokes, but was found to be independently associated with revascularization in a subgroup analysis of anterior circulation strokes (325 observations: OR 0.37, 95% CI 0.17–0.78). Higher baseline NIHSS scores (median 10; IQR 7–14) were correlated with higher rates of arterial recanalization. Previous publications, although limited to IVT-treated patients, have always reported opposite findings [16,17,24]. This paradox can be explained by an indication bias. When there was a large-vessel location of the

occlusion and a higher NIHSS score, patients were more likely to be treated with mechanical revascularization. These endovascular treatments (alone or in combination with IVT) had an impressively high OR of 7 for arterial recanalization of any vessel occlusion, and an OR of 4 for recanalization of IC occlusions. IVT was clearly less potent in achieving recanalization, but our study was not highly powered enough to show a statistically significant superiority of endovascular treatment. Nevertheless, the IVT-treated patients had somewhat higher recanalization rates for most occlusion sites than in recent studies [19,30,31]. Given the fact that effective recanalization alone is not sufficient to guarantee a better outcome [32], randomized studies are needed to show that the radiologic superiority of endovascular treatment translates into clinical benefit. Our data suggest that this may be the case in patients with additional EC stenosis or occlusion, although the additional time delay to endovascular treatment may reduce its benefits. Among other radiologic features, a normal pretreatment ASPECTS on plain CT (median 10; IQR 9–10) is associated with a greater chance of complete and partial recanalization of IC occlusions. This has already been demonstrated in IVT-treated acute middle cerebral artery occlusions, and may reflect the absence of a large core volume, or better collaterals. We have confirmed that additional EC pathology decreases the chances of recanalization, independently of treatment [19]. This may be attributable to technical difficulties in recanalizing EC arteries and in achieving complete recanalization of associated IC occlusions, and hemodynamic factors. Thrombus length extending over more than one vessel segment also decreased recanalization by ~ 50% (Table 1), but this effect did not reach significance in multivariate analysis. However, in contrast to our expectations and the findings of other authors [16], the probabilities for recanalization of large and intermediate IC occlusion sites were similar. This observation needs to be confirmed, or may alternatively be attributable to chance, or non-measured factors such as collateral status or the volumes of core and penumbra. Some previously described predictors, such as high serum glucose and elevated systolic blood pressure at admission, a history of diabetes, smoking or atrial fibrillation, female gender and age were not identified in our logistic regression analysis [13,19,21,22]. This may be related to patient selection in the previous studies (only IVT or only endovascular treatment) or the absence of logistic regression analysis. Recanalization after large-vessel stroke is the most important modifiable predictor of a favorable outcome. A meta-analysis of case series evaluating the effect of recanalization on outcome found a substantial decrease in mortality (OR 0.24, 95% CI 0.16–0.35) and an increase in the likelihood of a good functional outcome among revascularized patients (OR 4.4, 95% CI 3.3–5.9) [1]. Our predictive models are able to identify large-vessel © 2014 International Society on Thrombosis and Haemostasis

Predictors of recanalization in stroke 819

occlusive strokes that are likely to recanalize and to predict the impact of some modifiable covariates on the recanalization. Implementation in routine clinical practice may be hampered by the moderate predictive ability of the models, with areas under the ROC curve of between 0.72 and 0.73. The strengths of our study are the large number of patients and included variables, the inclusion of patients with different recanalization strategies, and the application of logistic regression analysis. Several limitations apply to the generalizability of our results: First, a significant number of patients with arterial occlusion on initial imaging did not undergo good-quality repeat imaging at 24 h, mainly in the initial period of the register and because of the medical condition of the patient. The revascularization was recorded at 24 h (range 12–48 h), but the optimal timing for reassessment of the vessel occlusion remains uncertain [18,20,33]. The decision on the 24-h interval was made on the basis of the fact that persistent recanalization would be a stronger predictor than early recanalization (first 2 h on transcranial ultrasonography) for a favorable outcome at 3 months [9]. The 12-h absolute time window for salvageable brain tissue is still debatable. Therefore, an additional multivariate analysis with an 8-h time window was performed, and gave similar results as only six of 439 patients (1.3%) had their first vascular imaging between 8 h and 12 h. Second, collateral vessel status [33], core/ penumbra volumes and thrombus length [16] were not assessed as potential predictors, but need to be investigated in future studies. These data were lacking in the ASTRAL. To compensate for thrombus lengths, we used the variable ‘occlusion extension’ as an indirect measure of thrombus extent. Third, large-vessel occlusions were classified into large and intermediate arterial occlusions, mainly on the basis of anatomic location. The likelihood of recanalization between the locations may be more heterogeneous than expected. This arbitrary classification is debatable, but clinically relevant in the absence of a standardized classification of the IC arteries. Fourth, the degree of revascularization was not taken into account, and future studies will need to identify predictors of complete recanalization separately. Finally, external validation of this predictive model in a second large dataset has not been performed, as a good validation cohort needs to be derived rigorously, to contain all the necessary clinicoradiologic data, and to use similar classifications for the critical variables. Therefore, our singlecenter-based results contain some inherent uncertainty. Chance variation may have played a role, as a large number of variables were included in the project. Summary We identified several clinical and radiologic factors associated with recanalization in acute ischemic stroke, of which endovascular treatment was the most important. In © 2014 International Society on Thrombosis and Haemostasis

comparison, IVT had a significant but moderate effect. Significant EC vessel pathology strongly decreases the probability of arterial recanalization in large-vessel occlusive stroke. For IC recanalization, a normal level of consciousness and a normal ASPECTS are supplementary predictors for recanalization. The positive correlation between higher NIHSS scores and recanalization may have resulted from an indication bias. Better and earlier identification in acute stroke patients of factors influencing arterial recanalization could influence and individualize treatment strategies. This could result in better clinical outcome in the short term and long term. Addendum P. Vanacker: corresponding author, study design and concept, data collection, data analysis and interpretation, and preparation of the manuscript. D. Lambrou data analysis and data interpretation. A. Eskandari: data collection. P. Maeder data collection and data interpretation. R. Meuli data collection and data interpretation. G. Ntaios: study design and concept, data collection, data interpretation, and critical revision of the manuscript for important intellectual content. P. Michel study design and concept, data collection, data interpretation, critical revision of the manuscript for important intellectual content, and study supervision. Acknowledgements P. Mosimann (Department of Radiology, University Hospital Lausanne, Switzerland), L. Hirt (Department of Neurology, University Hospital Lausanne, Switzerland) and R. Sztajzel (Department of Neurology, University Hospital Geneva, Switzerland) contributed to the development of the recanalization criteria. Funding sources This research was supported by grants from the Swiss Cardiology foundation (P. Michel and P. Vanacker), CardioMet-CHUV (P. Michel), and the European Neurological Society (P. Vanacker). Disclosure of Conflict of Interests P. Vanacker has received research support from the European Neurological Society. P. Michel has received funding for travel or speaker honoraria from Shire, Bayer Schering Pharma, Boehringer-Ingelheim, and Sanofi-Aventis, consulting fees from Pierre-Fabre, Servier, and Lundbeck, honoraria from scientific advisory boards for Bayer and Boehringer-Ingelheim, and research support from the Swiss National Science Foundation, the Swiss Cardiology Foundation, CardioMet CHUV, and Lundbeck. The other authors state that they have no conflict of interest.

820 P. Vanacker et al

Supporting Information Additional Supporting Information may be found in the online version of this article: Data S1. Methods. Table S1. Definition of occlusion and stenosis in the initial arterial study in acute ischemic stroke. Table S2. Definition of partial and complete recanalization. Figure S1. Flowchart of patient selection and distribution in different subgroups

11

12 13

References 14 1 Rha JH, Saver JL. The impact of recanalization on ischemic stroke outcome: a meta-analysis. Stroke 2007; 38: 967–73. 2 Khatri P, Abruzzo T, Yeatts SD, Nichols C, Broderick JP, Tomsick TA. IMS I and II Investigators. Good clinical outcome after ischemic stroke with successful revascularization is time-dependent. Neurology 2009; 73: 1066–72. 3 Olivot JM, Mlynash M, Thijs VN, Kemp S, Lansberg MG, Wechsler L, Schlaug G, Bammer R, Marks MP, Albers GW. Relationships between infarct growth, clinical outcome, and early recanalization in diffusion and perfusion imaging for understanding stroke evolution (DEFUSE). Stroke 2008; 39: 2257–63. 4 Ogata T, Christensen S, Nagakane Y, Ma H, Campbell BC, Churilov L, Lansberg MG, Straka M, De Silva DA, Mlynash M, Bammer R, Olivot JM, Desmond PM, Albers GW, Davis SM, Donnan GA; for the EPITHET DEFUSE Investigators. The effects of alteplase 3 to 6 hours after stroke in the EPITHET-DEFUSE combined dataset: post hoc case-control study. Stroke 2013; 44: 87–93. 5 Davis SM, Donnan GA, Parsons MW, Levi C, Butcher KS, Peeters A, Barber PA, Bladin C, De Silva DA, Byrnes G, Chalk JB, Fink JN, Kimber TE, Schultz D, Hand PJ, Frayne J, Hankey G, Muir K, Gerraty R, Tress BM, et al., for the EPITHET investigators. Effects of alteplase beyond 3 h after stroke in the Echoplanar Imaging Thrombolytic Evaluation Trial (EPITHET): a placebocontrolled randomised trial. Lancet Neurol 2008; 7: 299–309. 6 Fiebach JB, Al-Rawi Y, Wintermark M, Furlan AJ, Rowley HA, Lindst
en A, Smyej J, Eng P, Warach S, Pedraza S. Vascular occlusion enables selecting acute ischemic stroke patients for treatment with desmoteplase. Stroke 2012; 43: 1546–66. 7 Galimanis A, Jung S, Mono ML, Fischer U, Findling O, Weck A, Meier N, De Marchis GM, Brekenfeld C, El-Koussy M, Mattle HP, Arnold M, Schroth G, Gralla J. Endovascular therapy of 623 patients with anterior circulation stroke. Stroke 2012; 43: 1052–7. 8 Mazighi M, Serfaty JM, Labreuche J, Laissy JP, Meseguer E, Lavall
ee PC, Cabrejo L, Slaoui T, Guidoux C, Lapergue B, Klein IF, Olivot JM, Abboud H, Simon O, Niclot P, Nifle C, Touboul PJ, Raphaeli G, Gohin C, Claeys ES, et al. RECANALISE investigators. Comparison of intravenous alteplase with a combined intravenous–endovascular approach in patients with stroke and confirmed arterial occlusion (RECANALISE study): a prospective cohort study. Lancet Neurol 2009; 8: 802–9. 9 Yeo LLL, Paliwal P, Teoh HL, Seet RC, Chan BP, Liang S, Venketasubramanian N, Rathakrishnan R, Ahmad A, Ng KW, Loh PK, Ong JJ, Wakerley BR, Chong VF, Bathla G, Sharma VK. Timing of recanalization after intravenous thrombolysis and functional outcomes after acute ischemic stroke. JAMA Neurol 2013; 70: 353–8. 10 Zaidat OO, Suarez JI, Sunshine JL, Tarr RW, Alexander MJ, Smith TP, Enterline DS, Selman WR, Landis DM. Thrombolytic

15

16

17

18

19

20

21

22

23

24

therapy of acute ischemic stroke: correlation of angiographic recanalization with clinical outcome. Am J Neuroradiol 2005; 26: 880–4. Soares BP, Tong E, Hom J, Cheng SC, Bredno J, Boussel L, Smith WS, Wintermark M. Reperfusion is a more accurate predictor of follow-up infarct volume than recanalization: a proof of concept using CT in acute ischemic stroke patients. Stroke 2010; 41: 34–40. Cloft HJ. Death and destruction in the intra-arterial battle with acute ischemic stroke. Am J Neuroradiol 2011; 32: 1769–70. Arnold M, Mattle S, Galimanis A, Kappeler L, Fischer U, Jung S, Demarchis G, Gralla J, Mono ML, Brekenfeld C, Meier N, Nedeltchev K, Schroth G, Mattle HP. Impact of admission glucose and diabetes on recanalization and outcome after intra-arterial thrombolysis for ischaemic stroke. Int J Stroke 2012; doi:10. 1111/j.1747-4949.2012.00879.x. Zangerle A, Kiechl S, Spiegel M, Furtner M, Knoflach M, Werner P, Mair A, Wille G, Schmidauer C, Gautsch K, Gotwald T, Felber S, Poewe W, Willeit J. Recanalization after thrombolysis in stroke patients: predictors and prognostic implications. Neurology 2007; 68: 39–44. Kimura K, Iguchi Y, Yamashita S, Shibazaki K, Kobayashi K, Inoue T. Atrial fibrillation as an independent predictor for no early recanalization after IV-rtPA in acute ischemic stroke. J Neurol Sci 2008; 246: 57–61. Riedel CH, Zimmerann P, Jensen-Kondering U, Stingele R, Deuschl G, Jansen O. The importance of size: succesful recanalization by intravenous thrombolysis in acute anterior stroke depends on thrombus length. Stroke 2011; 42: 1775–7. Moftakhar P, English JD, Cooke DL, Kim WT, Stout C, Smith WS, Dowd CF, Higashida RT, Halbach VV, Hetts SW. Density of thrombus on admission CT predicts revascularization in large vessel occlusion acute ischemic stroke. Stroke 2013; 44: 243–5. Saqqur M, Uchino K, Demchuk AM, Molina CA, Garami Z, Calleja S, Akhtar N, Orouk FO, Salam A, Shuaib A, Alexandrov AV; CLOTBUST Investigators. Site of arterial occlusion identified by transcranial Doppler predicts the response to intravenous thrombolysis for stroke. Stroke 2007; 38: 948–54. Tsivgoulis G, Saqqur M, Sharma VK, Lao AY, Hoover SL, Alexandrov AV; CLOTBUST Investigators. Association of pretreatment ASPECTS scores with tPA-induced arterial recanalization in acute middle cerebral artery occlusion. J Neuroimag 2008; 18: 56–61. Arnold M, Kappeler L, Nedeltchev K, Brekenfeld C, Fischer U, Keserue B, Remonda L, Schroth G, Mattle HP. Recanalization and outcome after intra-arterial thrombolysis in middle cerebral artery and internal carotid artery occlusion. Does sex matter? Stroke 2007; 38: 1281–5. Kufner A, Nolte CH, Galinovic I, Brunecker P, Kufner GM, Endres M, Fiebach JB, Ebinger M. Smoking–thrombolysis paradox: recanalization and reperfusion rates after intravenous tissue plasminogen activator in smokers with ischemic stroke. Stroke 2013; 44: 407–13. Mendoncßa N, Rodriguez-Luna D, Rubiera M, Boned-Riera S, Ribo M, Pagola J, Pi~ neiro S, Meler P, Alvarez-Sabin J, Montaner J, Molina CA. Predictors of tissue-type plasminogen activator nonresponders according to location of vessel occlusion. Stroke 2012; 43: 417–21. Michel P, Odier C, Rutgers M, Reichhart M, Maeder P, Meuli R, Wintermark M, Maghraoui A, Faouzi M, Croquelois A, Ntaios G. The Acute Stroke Registry and Analysis of Lausanne (ASTRAL). Design and baseline analysis of an ischemic stroke registry including acute multimodal imaging. Stroke 2010; 41: 2491–8. Wunderlich MT, Goertler M, Postert T, Schmitt E, Seidel G, Gahn G, Samii C, Stolz E; Duplex Sonography in Acute Stroke (DIAS) Study Group; Competence Network Stroke. Recanaliza© 2014 International Society on Thrombosis and Haemostasis

Predictors of recanalization in stroke 821

25

26

27

28

29

tion after intravenous thrombolysis: does a recanalization time window exist? Neurology 2007; 68: 1364–8. Ribo M, Alvarez-Sab
ın J, Montaner J, Romero F, Delgado P, Rubiera M, Delgado-Mederos R, Molina CA. Temporal profile of recanalization after intravenous tissue plasminogen activator: selecting patients for rescue reperfusion techniques. Stroke 2006; 37: 1000–4. Puetz V, Dzialowski I, Hill MD, Demchuk AM. The Alberta Stroke Program Early CT score in clinical practice: what have we learned? Int J Stroke 2009; 4: 354–64. Puetz V, Khomenko A, Hill MD, Dzialowski I, Michel P, Weimar C, Wijman CA, Mattle HP, Engelter ST, Muir KW, Pfefferkorn T, Tanne D, Szabo K, Kappelle LJ, Algra A, von Kummer R, Demchuk AM, Schonewille WJ. Extent of hypoattenuation on CT angiography source images in basilar artery occlusion: prognostic value in the Basilar Artery International Cooperation Study. Stroke 2011; 42: 3454–9. Michel P, Arnold M, Engelter S. f€ ur die Themengruppe ‘Akuttherapie des Hirnschlags’ der Zerebrovaskul€aren Arbeitsgruppe der Schweiz. Thrombolyse beim isch€amischen Hirnschlag: aktualisierte Leitlininen. Swiss Med Wkly 2009; 9: 892–4. Bhatia R, Hill MD, Shobha N, Menon B, Bal S, Kochar P, Watson T, Goyal M, Demchuk AM. Low rates of acute recanaliza-

© 2014 International Society on Thrombosis and Haemostasis

30

31

32

33

tion with intravenous recombinant tissue plasminogen activator in ischemic stroke – real-world experience and a call for action. Stroke 2010; 41: 2254–8. Lee KY, Han SW, Kim SH, Nam HS, Ahn SW, Kim DJ, Seo SH, Kim DI, Heo JH. Early recanalization after intravenous administration of recombinant tissue plasminogen activator as assessed by pre- and post-thrombolytic angiography in acute ischemic stroke patients. Stroke 2007; 38: 192–3. Broderick JP, Palesch YY, Demchuk AM, Yeatts SD, Khatri P, Hill MD, Jauch EC, Jovin TG, Yan B, Silver FL, von Kummer R, Molina CA, Demaerschalk BM, Budzik R, Clark WM, Zaidat OO, Malisch TW, Goyal M, Schonewille WJ, Mazighi M, et al. Endovascular therapy after intravenous t-PA versus t-PA alone for stroke. N Engl J Med 2013; 368: 893–903. Ntaios G, Faouzi M, Ferrari J, Lang W, Vemmos K, Michel P. An integer-based score to predict functional outcome in acute ischemic stroke: the ASTRAL score. Neurology 2012; 78: 1916–22. Miteff F, Levi CR, Bateman GA, Spratt N, McElduff P, Parsons MW. The independent predictive utility of computed tomography angiographic collateral status in acute ischaemic stroke. Brain 2009; 132: 2231–8.