Mechanical endovascular thrombectomy for acute ischemic stroke: a retrospective multicenter study in Belgium

Transcription

1 DOI /s ORIGINAL ARTICLE Mechanical endovascular thrombectomy for acute ischemic stroke: a retrospective multicenter study in Belgium Niels Fockaert 1 Marieke Coninckx 2 Sam Heye 3 Luc Defreyne 4 Denis Brisbois 5 Pierre Goffette 6 Jan Gralla 7 Pasquale Mordasini 7 Andre Peeters 8 Philippe Desfontaines 9 Dimitri Hemelsoet 2 Vincent Thijs 1,10,11 Robin Lemmens 1,10,11 Received: 22 July 2015 / Accepted: 25 September 2015 Belgian Neurological Society 2015 Abstract Clinical trials have shown a beneficial effect of mechanical thrombectomy in acute ischemic stroke patients treated within six up to even 12 h after symptom onset. This treatment was already performed in selected hospitals in Belgium before completion of the randomized controlled trials. Outcome data on these procedures in Belgium have not been published. We performed a retrospective multicenter study of all patients with acute ischemic stroke treated with mechanical endovascular therapy in four hospitals in Belgium. Clinical outcomes, as measured by the modified Rankin Scale (mrs), site of arterial occlusion, reperfusion and the association between these variables were studied. The study included 80 patients: 65 patients with an occlusion in the anterior circulation and 15 with an occlusion in the posterior circulation. Good functional outcome (GFO) rates, defined as mrs 0 2 at 90 days, were 42 % in all patients, 44 % in anterior circulation stroke and 34 % in posterior circulation stroke. Reperfusion was achieved in 78 % of patients; more (100 %) in patients with posterior compared to patients with anterior circulation stroke (72 %; p = 0.02). The rate of GFO was greater in patients with reperfusion versus patients in whom reperfusion was not achieved (adjusted OR 8.2, 95 % CI ). Symptomatic intracerebral hemorrhage was documented in 5 % of all patients. Endovascular treatment with mechanical devices for acute ischemic stroke in Belgium results in GFO and reperfusion rates similar to recently published results in the endovascular-treated arms of randomized clinical trials. Rates of symptomatic intracranial hemorrhage are low and comparable to other cohort studies and clinical trials. Keywords Endovascular Thrombectomy Ischemic stroke Reperfusion Outcome & Robin Lemmens robin.lemmens@uzleuven.be Department of Neurology, University Hospitals Leuven, 3000 Leuven, Belgium Department of Neurology, University Hospitals Gent, 9000 Ghent, Belgium Department of Radiology, University Hospitals Leuven, 3000 Leuven, Belgium Department of Vascular and Interventional Radiology, University Hospitals Gent, 9000 Ghent, Belgium Department of Interventional Neuroradiology, Centre Hospitalier Chrétien, 4000 Liège, Belgium Department of Neuroradiology, Cliniques Universitaires Saint-Luc, Université catholique de Louvain, Brussels, Belgium Department of Diagnostic and Interventional Neuroradiology, University Hospital Bern, Bern, Switzerland Department of Neurology, Cliniques Universitaires Saint- Luc, Université catholique de Louvain, 1200 Brussels, Belgium Department of Neurology, Centre Hospitalier Chrétien, 4000 Liège, Belgium Department of Neurosciences, Experimental Neurology and Leuven Research Institute for Neuroscience and Disease (LIND), KU Leuven, University of Leuven, 3000 Leuven, Belgium VIB, Vesalius Research Center, Laboratory of Neurobiology, 3000 Leuven, Belgium

2 Introduction Intravenous thrombolysis has been the standard treatment for acute ischemic stroke patients in the first 4.5 h after symptom onset, aimed at reperfusion and thereby improving clinical outcomes [1]. Recanalization rates following thrombolysis are low for occlusions of intracranial large vessels and large clots (C8 mm) and, therefore, the role of intra-arterial therapy has been explored to improve these recanalization rates [2]. Initially, randomized clinical trials (RCTs) failed to show benefit on outcome when compared to standard medical treatment [3 5]. These trials had numerous limitations including the absence of a pretreatment confirmed occlusion, the limited use of stentrievers and probably inclusion bias. The MR CLEAN was the first RCT to demonstrate a beneficial effect of endovascular therapy in anterior circulation stroke patients treated up to 6 h after symptom onset [6]. These findings in patients with large vessel occlusions were recently confirmed by four other RCTs: ESCAPE, EXTEND-IA, SWIFT PRIME and REVASCAT [7 10]. During the period when the evidence for endovascular treatment in acute ischemic stroke was lacking, mechanical endovascular thrombectomy was already performed in some centers in Belgium. However, no data have been published on procedural and clinical outcomes in the Belgian population. The aim of this retrospective study was to determine the reperfusion rates and clinical outcomes in multiple centers and compare these data to recently reported results in RCTs. Methods Study design This retrospective multicenter case study was designed to analyze the efficacy and safety of mechanical thrombectomy in patients with acute ischemic stroke in Belgium. Patients receiving mechanical thrombectomy between January and December in the following four centers were included: University Hospital Gent, Centre Hôpitalier Chrétien de Liege, Cliniques universitaires Saint-Luc in Brussels and the University hospitals Leuven. The ethic committees of each participating center approved the study protocol. Only fully authorized personnel with medical qualifications had access to the medical records and scans. Inclusion criteria Patients with a minimum age of 18 years who suffered an acute ischemic stroke and met local criteria for thrombectomy were included. There was no upper age limit, nor a minimum or maximum stroke severity as measured with the National Institutes of Health Stroke Scale (NIHSS). Patients were eligible for inclusion in this analysis if a large vessel occlusion in either the anterior or posterior occlusion was present: internal carotid artery (ICA), medial cerebral artery (MCA) or basilar artery (BA). Time of start of the endovascular procedure needed to be within 12 h after symptom onset. Patients could have been treated with intravenous t-pa, but persistent occlusion had to be present on the digital substraction angiography (DSA). Competent patients gave informed consent before the procedure was started. Otherwise, a witnessed waiver of consent was possible. Data collection Patient characteristics were obtained from medical records: age, sex, time of stroke onset, NIHSS, glucose level at admission, treatment with intravenous thrombolysis, time between symptom onset to groin puncture, device used, general anesthesia, and stroke etiology [11]. Time between onset and reperfusion was not consistently reported in all cases and was, therefore, not included in the analysis. Neuro-imaging analysis Intra-arterial digital subtraction angiographies (IADSA) were reviewed by two independent raters and scored by consensus (JG and PM). Measurements were performed blinded to clinical information and radiology reports. Images were reviewed for the site of arterial occlusive lesion (AOL) and reperfusion grade according to the modified Thrombolysis in Cerebral Infarction score (mtici) [3]. MTICI grade 0 was defined as no perfusion; grade 1 as perfusion past the initial obstruction but limited distal branch filling with little or slow distal perfusion; grade 2a as perfusion of less than half of the vascular distribution of the occluded artery; grade 2b as perfusion of half or greater of the vascular distribution of the occluded artery; grade 3 as full perfusion with filling of all distal branches. Symptomatic intracerebral hemorrhage was reported by each center individually according to SITS- MOST: type 2 parenchymatous intracerebral hemorrhage with neurological deterioration, defined as an increase of 4 or more points in the score on the NIHSS [12]. Outcome measures The primary clinical endpoint was good functional outcome (GFO) at 90 days, defined as a mrs 0 2, obtained by the treating neurologist not blinded for the result of the mechanical thrombectomy. The secondary outcome was excellent functional outcome (EFO), defined as mrs 0 1.

3 The safety outcome measures were the presence of symptomatic intracerebral hemorrhages following the procedure and mortality at 90 days. Outcome measures were analyzed for all patients and were stratified by anterior versus posterior circulation stroke and further by site of AOL. Statistical analysis We intended to compare outcomes in anterior versus posterior circulation stroke and to analyze the association between reperfusion and the various clinical outcomes in the overall patient series, anterior circulation, internal carotid artery (ICA) or middle cerebral artery (MCA), and posterior circulation, basilar artery (BA), strokes. Therefore, the mtici scores were dichotomized and defined as substantial reperfusion (mtici 2B-3) versus no substantial reperfusion (mtici 0-1-2A). Regression analyses were performed to examine the association between reperfusion and GFO. We used a multivariate linear logistic regression model in which GFO and EFO were the dependent variable, and age and baseline NIHSS were forced as prespecified variables for which adjustment was needed. During the preliminary variable selection, collected variables that were significant at an alpha B 0.10 were considered candidates for the final model. For the final model, variables were added sequentially starting with the variable with the lowest p value from the group of candidate predictors. We retained variables with p values B0.05 in the final model. In a secondary analysis, we evaluated the effect of reperfusion on GFO, EFO, mortality and sich using the same model. In a tertiary analysis, we used the full range of the mrs as the dependent variable. We compared the mrs distribution, divided into 6 categories (0, 1, 2, 3, 4, 5, 6), between patients with and without reperfusion using ordinal logistic regression. SPSS22.0 was used to perform the statistical analyses. Results Between January and December , 85 patients underwent endovascular therapy for acute ischemic stroke in four Belgian centers. One patient was excluded due to the lower age limit; a second patient was excluded because of spontaneous recanalization. In three patients, a final angiogram was lacking for central read and, therefore, these patients were not included in the analysis. Here, we report on 80 patients: 65 (81 %) with an occlusion in the anterior circulation, 18 ICA (22 %) and 47 MCA (59 %); and 15 (19 %) with an occlusion in the posterior circulation. Baseline characteristics are summarized in Table 1. Intravenous t-pa was administered in 45 patients (56 %). In 18 patients (22 %), the onset to groin interval was not documented. The median onset to groin interval in the other 62 patients was 267 min. In 45 patients (10 ICA, 32 MCA and 3 BA), endovascular treatment was initiated in the first 6 h after stroke onset and in 17 patients (5 ICA, 8 MCA, 4 BA) between 6 and 12 hours after stroke onset. In 75 patients (94 %), stentrievers were used. Two patients (2.5 %) were lost to follow-up and for those the clinical outcome at discharge was carried forward. GFO was achieved in 34 patients (42 %) without differences in outcome when comparing patients with anterior (29 patients, 44 %) versus posterior circulation stroke (5 patients, 34 %) (Fig. 1). In the subgroup of patients with anterior circulation stroke who were treated within the first 6 h after symptom onset, GFO was obtained in 20 patients (47.5 %). Rates of EFO were 26 % (21 patients) in the overall patient series and were comparable between patients with anterior circulation stroke (19 patients, 29 %) versus posterior stroke circulation (2 patients, 14 %). Substantial reperfusion, defined as mtici 2B-3, was achieved in 62 patients (78 %) in the overall patient series. The percentage of reperfusion differed between patients with posterior circulation stroke, 100 %, versus anterior circulation stroke, 72 % (47 patients; p = 0.02) (Fig. 2). In the multivariate model with GFO as the dependent variable, no other variables were retained in the final model apart from the pre-specified predictors age and baseline NIHSS. Reperfusion was associated with an increased rate of GFO (adjusted OR 8.2, 95 % CI ) and marginally with EFO (adjusted OR 5.3, 95 % CI ; p = 0.05) in the overall population. Since reperfusion was achieved in all patients with occlusions in the posterior circulation, the association between reperfusion and clinical outcomes could only be assessed in patients with anterior circulation occlusions. Baseline characteristics for anterior stroke circulation stroke patients stratified by reperfusion are shown in Table 2. The distribution of MCA and ICA occlusions differed in patients with versus without reperfusion as shown by a larger percentage of MCA occlusions in patients in whom reperfusion was achieved (81 versus 50 %; p = 0.03). In patients with anterior circulation occlusions, there was a strong association between reperfusion and GFO (adjusted OR 7.5; 95 % CI ) and EFO (adjusted OR 6.1; 95 % CI ) (Fig. 3). The site of AOL (MCA versus ICA) did not modify the effect of reperfusion on GFO (p for interaction = 0.9) or EFO (p for interaction = 0.5). The sich rate was 5 % in the overall patient series with no significant difference between patients with anterior versus posterior circulation occlusions: 3 % (2 patients) with anterior versus 13 % (2 patients) with posterior circulation stroke (p = 0.2). Reperfusion was not associated with sich (p = 0.5).

4 Table 1 Baseline characteristics of all included patients Parameter Overall (N = 80) Total anterior (N = 65) ICA (N = 18) MCA (N = 47) BA (N = 15) Age (years) Median Range Sex no. (%) Male 35 (43) 22 (34) 7 (39) 15 (32) 13 (87) Female 45 (57) 43 (66) 11 (61) 32 (68) 2 (13) NIHSS score Median Range Stroke cause no. (%) Large artery atherosclerosis 6 (7.5) 6 (9) 2 (11) 4 (8) 0 (0) Cardio-embolism 44 (55) 38 (58) 9 (50) 29 (62) 6 (40) Atrial fibrillation 33 (41) 29 (45) 6 (33) 23 (49) 4 (27) Other 13 (16) 9 (14) 3 (22) 6 (13) 3 (20) Small vessel occlusion 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) Other determined etiology 6 (7.5) 4 (6) 3 (17) 1 (2) 2 (13) Cryptogenic 24 (30) 17 (26) 4 (22) 13 (28) 7 (47) Intravenous t-pa no. (%) 45 (56) 39 (60) 10 (55) 29 (62) 6 (40) Time onset to groin (min) N Median Device type no. (%) Merci 4 (5) 3 (6) 2 (11) 1 (2) 1 (7) Solitaire 73 (92) 59 (91) 16 (89) 43 (91) 14 (93) Codman 2 (2) 2 (3) 0 (0) 2 (5) 0 (0) Penumbra 1 (1) 1 (0) 0 (0) 1 (2) 0 (0) ICA internal carotid artery, MCA medial cerebral artery, BA basilar artery, NIHSS National Institutes of Health Stroke Scale, TOAST the trial of ORG in acute stroke treatment Mortality was 26 % in the overall patient series with a trend towards a higher mortality rate in patients with posterior circulation stroke: 7 patients (47 %) with posterior circulation stroke, versus 14 patients (22 %) with anterior circulation stroke (p = 0.06). There was no relationship between reperfusion and mortality (p = 0.5). In a tertiary analysis, the effect of reperfusion on the full distribution of mrs scores at day 90 was determined, stratified by AOL. Reperfusion was associated with better functional outcomes when analyzed over the full distribution of mrs scores in all patients (adjusted proportional OR 4.9; 95 % CI ); and in anterior circulation stroke (adjusted proportional OR 4.8; 95 % CI ) without evidence of an interaction between site of AOL (ICA versus MCA) on the effect of reperfusion on functional outcome (p for interaction = 0.5). Discussion Recently, robust evidence has been provided by randomized clinical trials showing a beneficial effect of endovascular therapy for patients with acute ischemic stroke due to large vessel occlusions in the anterior circulation [6 10]. In this Belgian retrospective study, we report on clinical outcomes and reperfusion rates in a period that endovascular treatment was experimentally being introduced. Outcomes were similar to the results of randomized clinical trials and a strong association between reperfusion and functional outcome was confirmed. In this multicenter study, there was no control group and, therefore, the effect of the endovascular therapy versus standard medical treatment could not be assessed. To gain insight into our results, the outcomes in anterior circulation stroke were compared to the endovascular treatment arms of recently performed randomized clinical trials:

5 Fig. 1 Functional outcome in all patients, stratified by AOL. In the overall population, GFO was achieved in 42 % of patients and EFO in 26 % EFO. When stratified by anterior versus posterior circulation stroke, no differences were observed: The rate of GFO was 44 % in patients with anterior circulation stroke versus 34 % in patients with basilar artery stroke. ICA internal carotid artery, MCA medial cerebral artery, BA basilar artery, mrs modified Rankin Scale, GFO good functional outcome (mrs 0 2), EFO excellent functional outcome (mrs 0 1) Fig. 2 Distribution of postprocedure mtici scores, stratified by site of AOL. Substantial reperfusion, defined as an mtici 2B-3, was achieved in 78 % of patients in the overall patient series; in 73 % with an occlusion in the anterior circulation versus 100 % with an posterior circulation occlusion (p = 0.02). ICA internal carotid artery, MCA medial cerebral artery, BA basilar artery, AOL arterial occlusive lesion, mtici modified thrombolysis in cerebral infarction MR CLEAN, ESCAPE, SWIFT PRIME, EXTEND IA and REVASCAT [6 10]. When comparing various trials, differences in inclusion criteria and methodology need to be taken into account. Variation in predictors of outcome between studies can result in differences in outcome not based on the therapeutic effect, but due to dissimilarities in trial design. Age, baseline NIHSS and time to treatment have been shown to be strong predictors of outcome. An upper age limit (of 80) was only used in REVASCAT and SWIFT PRIME, but mean ages did not vary greatly between the various studies. Median NIHSS in the various studies was comparable although an upper NIHSS limit of 29 was used in SWIFT PRIME. The onset to treatment time window showed a wide range per protocol: 6 h in MR CLEAN, EXTEND-IA and SWIFT PRIME, 8 h in REVASCAT and 12 h in ESCAPE and our study. The median onset to treatment times, however, did not differ importantly, with a range between 3 and 4.5 h. An additional predictive factor which could influence the response to endovascular treatment is the use of imaging-based selection. In EXTEND-IA, ESCAPE, REVASCAT and SWIFT PRIME, patients could only be included if the presence of a large infarct core was excluded before randomization of the patient. This selection was established to optimize the potential of endovascular treatment, since this therapy will not likely be beneficial in patients in whom brain tissue has already been lost. A final important predictor of outcome is reperfusion. Reperfusion has been associated with good clinical outcomes and, therefore, the reperfusion rate within each trial would

6 Table 2 Baseline characteristics of patients with AOL in the anterior circulation, stratified by reperfusion grade Parameter Reperfusion (N = 47) No reperfusion (N = 18) Age (years) Median Range Sex no. (%) Male 19 (40) 3 (16) Female 28 (60) 15 (84) NIHSS score Median Range Site of AOL no. (%) MCA 38 (81) 9 (50) ICA 9 (19) 9 (50) Intravenous t-pa no. (%) 28 (60) 11 (61) Time onset to groin (min) N (%) 38 (81) 15 (83) Median Range ICA internal carotid artery, MCA medial cerebral artery, AOL arterial occlusive lesion, NIHSS National Institutes of Health Stroke Scale subsequently correlate with the percentage of patients with good clinical outcomes. Both the use of imaging-based selection and the rate of substantial reperfusion following the endovascular procedure are nicely reflected in the clinical outcomes of the various studies. In MR CLEAN and our patient series, no imaging selection was used and the rates of GFO (mrs 0 2) were among the lowest of all studies: 32 % in MR CLEAN and 44 % in our study. In REVASCAT, the percentage of GFO was the same as in our patient series: 44 %. In ESCAPE and SWIFT PRIME, these rates were higher, 53 and 60 %; and in EXTEND-IA in which CTA/CTP imaging was combined to determine salvageable tissue and small cores the rate of GFO was even 71 %. The same trend is noticeable when analyzing the relationship between reperfusion and GFO rates. In MR CLEAN, the lowest reperfusion rate of 58 % was documented, and the studies with the highest reperfusion rates, SWIFT PRIME and EXTEND-IA, have shown the best GFO rates in the intervention arms. Taken all the various predictive variables into account in this indirect analysis, outcomes in endovascular-treated patients in our retrospective analysis are in line with recently published results in the endovascular-treated arms of randomized clinical trials. Additionally, this indirect comparison underscores that using Fig. 3 Distribution of 90-day functional outcome according to the modified Rankin Scale (mrs) stratified by reperfusion status. The graphs show the distribution of 90-day functional outcomes according to the modified Rankin Scale (mrs) stratified by reperfusion status in all patients (a) and in patients with anterior circulation stroke (b). Reperfusion is associated with GFO in all patients (OR 8.2; 95 % CI ) and in patients with occlusions in the anterior circulation (OR 8.2; 95 % CI )

7 imaging-based selection in concordance with achieving high substantial reperfusion rates will result in increased good clinical outcomes in endovascular-treated ischemic stroke patients with large vessel occlusions in the anterior circulation. Additional studies are required to determine the most optimal imaging selection criteria, since exclusive selection of patients who experience large benefit might result in excluding of patients who benefit from endovascular treatment with smaller effect sizes. Although only a limited number of patients with basilar artery occlusions were included in this patient series, we compared our results to a recently published large cohort of posterior circulation stroke patients treated with endovascular therapy since data on RCT are currently lacking. The Endovascular stroke therapy (ENDOSTROKE) trial evaluated clinical outcomes and revascularization rates in 148 patients with a confirmed basilar artery occlusion [13]. As in our study, GFO was obtained in 34 % of patients in ENDOSTROKE. Reperfusion rates were high in both studies although slightly higher in our patient series, 100 versus 79 %. Mortality was lower in ENDOSTROKE (33 versus 47 %), probably due to lower median NIHSS and shorter median time to treatment. Due to the 100 % reperfusion rate in our study, we were unable to examine the association between reperfusion and clinical outcome in posterior circulation strokes. Surprisingly the ENDO- STROKE trial revealed that recanalization did not significantly predict clinical outcome; NIHSS and collateral status, however, were independent predictors of clinical outcome in a multivariate analysis. We believe our study to have several strengths. Firstly, this was a multicenter assessment of clinical outcomes of all acute ischemic stroke patients in the participating centers who were endovascular treated during the early years of this treatment. Therefore, the results represent the real life experience of endovascular treatment with stentrievers for ischemic stroke. Secondly, the site of AOL and reperfusion status pre- and post-intervention were assessed by two independent raters blinded for the clinical and other neuroimaging data. Thirdly, in all included patients a large vessel occlusion was determined before the endovascular treatment was initiated. As with all multicenter retrospective studies, there are various limitations of our study. Firstly, data were retrospectively collected and, therefore, data on some potential predictive variables were lacking. For instance, adverse events during the procedure were not registered and could, therefore, not be included in the analysis. Secondly, all participating centers had individual protocols to decide on endovascular treatments of patients. Therefore, we cannot exclude that some selection may have occurred. Thirdly, the assessment of clinical outcomes and sich was not blinded. Consequently, there is a potential bias in centers reporting better outcomes compared to blinded assessment. Fourth, the numbers in the study are relatively small, thereby decreasing the power to identify associations (e.g., hampering additional analysis of an association between reperfusion stratified by site of AOL in the anterior circulation: ICA versus MCA). In conclusion, clinical outcomes in Belgian patients treated for anterior circulation occlusions are comparable to recently published results in the endovascular-treated arms of randomized clinical trials. Reperfusion is a strong predictor of GFO in patients with a large vessel occlusion in the anterior circulation. Rates of symptomatic intracranial hemorrhage are low and similar as reported in recent literature. This evidence underscores the potential of Belgian centers to perform endovascular therapy as standard therapy in patients with a large vessel occlusion in the anterior circulation. Acknowledgments We would like to thank our colleagues of the neurology and neuroradiology department at the University Hospital Gent, Centre Hôpitalier Chrétien de Liege and Cliniques universitaires Saint-Luc in Brussels to provide clinical and radiological data. VT and RL are Senior Clinical Investigators of the FWO Flanders. Compliance with ethical standards Conflict of interest Jan Gralla reports grants from Covidien. All other authors have no disclosures. Ethical approval This study was performed in accordance with the ethical standards of the institutional research committees and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. References 1. Emberson J, Lees KR, Lyden P et al (2014) Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke: a metaanalysis of individual patient data from randomised trials. Lancet 384(9958): Riedel CH, Zimmermann P, Jensen-Kondering U, Stingele R, Deuschl G, Jansen O (2011) The importance of size: successful recanalization by intravenous thrombolysis in acute anterior stroke depends on thrombus length. Stroke J Cereb Circ 42: Broderick JP, Palesch YY, Demchuk AM et al (2013) Endovascular therapy after intravenous t-pa versus t-pa alone for stroke. N Engl J Med 368: Ciccone A, Valvassori L (2013) Endovascular treatment for acute ischemic stroke. N Engl J Med 368: Kidwell CS, Jahan R, Gornbein J et al (2013) A trial of imaging selection and endovascular treatment for ischemic stroke. N Engl J Med 368: Berkhemer OA, Fransen PS, Beumer D et al (2015) A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med 372: Campbell BC, Mitchell PJ, Kleinig TJ et al (2015) Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med 372:

8 8. Goyal M, Demchuk AM, Menon BK et al (2015) Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med 372: Jovin TG, Chamorro A, Cobo E et al (2015) Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med 372(24): Saver JL, Goyal M, Bonafe A et al (2015) Stent-retriever thrombectomy after intravenous t-pa vs. t-pa alone in stroke. N Engl J Med 372(24): Adams HP Jr, Bendixen BH, Kappelle LJ et al (1993) Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org in acute stroke treatment. Stroke J Cereb Circ 24: Wahlgren N, Ahmed N, Davalos A et al (2007) Thrombolysis with alteplase for acute ischaemic stroke in the Safe Implementation of Thrombolysis in Stroke-Monitoring Study (SITS- MOST): an observational study. Lancet 369: Singer OC, Berkefeld J, Nolte CH et al (2015) Mechanical recanalization in basilar artery occlusion: the ENDOSTROKE study. Ann Neurol 77: