UvA-DARE (Digital Academic Repository) Intraarterial ... · van Zwam WH°, Majoie CBLM°, for the...

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Intraarterial treatment for acute ischemic stroke

Berkhemer, O.A.

Link to publication

Citation for published version (APA):Berkhemer, O. A. (2016). Intraarterial treatment for acute ischemic stroke

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 10 Jun 2018

http://dare.uva.nl/personal/pure/en/publications/intraarterial-treatment-for-acute-ischemic-stroke(9bf71181-6113-4151-8bc9-282ae58d8627).html

51

Chapter 3

Effect of Baseline Alberta Stroke Program Early CT

Score on Safety and Efficacy of

Intra‐arterial Treatment: a Subgroup Analysis of a

Randomized Phase 3 Trial (MR CLEAN)

Yoo AJy, Berkhemer OAy, Fransen PSS, van den Berg LA, Beumer D, Lingsma HF,

Schonewille WJ, Sprengers MES, van den Berg R, van Walderveen MAA, Beenen LFM, Wermer MJH, Lycklama à Nijeholt GJ, Boiten J, Jenniskens SFM, Bot JCJ, Boers AMM, Marquering HA, Roos YBWEM°, van Oostenbrugge RJ°, Dippel DWJ°, van der Lugt A°,

van Zwam WH°, Majoie CBLM°, for the MR CLEAN investigators y°Contributed equally

Lancet Neurol 2016;15(7):685‐694.

Chapter 3

52

Abstract

Background Whether infarct size modifies intra‐arterial treatment effect is not certain, particularly in patients with large infarcts. We examined the effect of the baseline Alberta Stroke Program Early CT Score (ASPECTS) on the safety and efficacy of intra‐arterial treatment in a subgroup analysis of the Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands (MR CLEAN). Methods MR CLEAN was a randomized, controlled, open‐label, phase 3 trial of intra‐arterial treatment in patients (aged 18 years from the Netherlands) with proximal arterial occlusion of the anterior circulation, given intra‐arterial treatment within 6 h of stroke onset. The primary outcome was 90‐day modified Rankin Scale (mRS) score. We estimated the intra‐arterial treatment effect for all patients in MR CLEAN who had ASPECTS graded by using multivariable ordinal logistic regression analysis (a proportional odds model) to calculate the adjusted common odds ratio for a shift towards a better functional outcome according to the mRS for intra‐arterial treatment and usual care than for usual care alone. We entered an interaction term into the model to test for interaction with prespecified ASPECTS subgroups: 0–4 (large infarct) versus 5–7 (moderate infarct) versus 8–10 (small infarct). Findings 496 patients—232 (47%) in the intra‐arterial treatment and usual care group and 264 (53%) in the usual care alone group—were included in the analysis. We noted no significant difference in intra‐arterial treatment effect between the ASPECTS subgroups according to 90‐day ordinal mRS (adjusted common odds ratio interaction term relative to ASPECTS 8–10: ASPECTS 0–4: 0.79 [95% CI 0.20–3.19], P=0.740; and ASPECTS 5–7: 1.02 [0.44–2.35], P=0.966). Intra‐arterial treatment did not cause a significant increase in the proportion of patients with at least one serious adverse event in any of the ASPECTS subgroups (ASPECTS 0–4: eight [73%] of 11 patients in treatment and usual care group vs. 11 [58%] of 19 in usual care alone group, P=0.42; ASPECTS 5–7: 32 [59%] of 54 vs. 19 [49%] of 39, P=0.31; ASPECTS 8–10: 70 [42%] of 167 vs. 82 [40%] of 206; P=0.68). For death within 7 days or within 30 days and hemi‐craniectomy, the differences between the intra‐arterial treatment and usual care versus usual care alone groups were not significant by ASPECTS subgroups. A significantly higher proportion of patients had recurrent ischemic stroke in the intra‐arterial treatment plus usual care group than in the usual care alone group in the ASPECTS 8–10 subgroup (eight [5%] vs. one [<1%]; P=0.007). Interpretation Contrary to findings from previous studies suggesting that only patients with non‐contrast CT ASPECTS of more than 7 benefit from intra‐arterial treatment, data from this study suggest that patients with ASPECTS 5–7 should be treated. Further evidence is needed for patients with ASPECTS 0–4, for whom treatment might yield only marginal absolute benefit.

Effect of Baseline ASPECTS on Safety and Efficacy of Intra‐arterial Treatment

53

Introduction

Evidence suggests that infarct size at baseline can be used to predict functional outcome after intra‐arterial treatment.1,2 Patients with smaller infarcts at baseline have better long‐term functional independence, lower incidence of reperfusion hemorrhage, and lower mortality than do those with larger infarcts.1,2 Whether baseline infarct size modifies intra‐arterial treatment effect, particularly whether patients with large infarcts benefit from catheter‐based reperfusion therapy, is not clear.3,4 Despite the availability of advanced imaging systems, non‐contrast CT is widely used for patient assessment before intra‐arterial treatment.5,6 The Alberta Stroke Program Early CT Score (ASPECTS) is a semi‐quantitative method of estimation of infarct size with non‐contrast CT during the acute phase. Findings from a post‐hoc analysis of the PROACT II trial3 showed an interaction of baseline ASPECTS (>7 vs. 7) with intra‐arterial treatment effect. However, in a similar analysis of the IMS III trial,4 no interaction was noted between ASPECTS and intra‐arterial treatment effect, which might have been confounded by the overall neutral trial results. Additionally, data have suggested that a larger infarct threshold (ASPECTS 0–4) might better define patients likely to have a poor response to intra‐arterial treatment than the previous ASPECTS threshold of 7.7

The results from the Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands (MR CLEAN)8,9 provide a unique opportunity to further examine use of non‐contrast CT for intra‐arterial treatment selection because of its randomized controlled design, the positive overall effect of intra‐arterial treatment in the trial, and the absence of explicit imaging exclusion criteria for infarct size. We sought to examine the effect of baseline ASPECTS (0–4 vs. 5–7 vs. 8–10) on the safety and efficacy of intra‐arterial treatment effect in this prespecified subgroup analysis of MR CLEAN.

Methods

Study design and patients MR CLEAN was a randomized, controlled, open‐label, phase 3 trial of intra‐arterial treatment and usual care versus usual care alone in patients from 16 centers in the Netherlands who were 18 years of age or older with a proximal arterial occlusion in the anterior circulation shown with vessel imaging, treated within 6 h of symptom onset. The study methods and patient eligibility criteria have been reported previously.8,9 In this subgroup analysis, we included all patients who had ASPECTS graded. Imaging criteria for inclusion were a CT or MRI scan ruling out hemorrhage and CT, magnetic (MR), or digital subtraction angiography showing occlusion of the intracranial internal carotid artery, middle cerebral artery (M1 or M2 segments), or anterior

Chapter 3

54

cerebral artery (A1 or A2 segments). No imaging exclusion criteria based on ASPECTS or the extent of early ischemic changes on non‐contrast CT were used. All patients or their legal representatives provided written informed consent before randomization. The study protocol was approved by a central medical ethics committee and the research board of each participating center.

Procedures Given that ASPECTS is a score based on a topographical map of the middle cerebral artery territory, we restricted this analysis to patients with occlusions of the intracranial internal carotid artery or the middle cerebral artery territory (i.e., territory at risk involves the middle cerebral artery territory). After closure of the MR CLEAN trial database, we identified additional baseline non‐contrast CT scans for three patients that were closer in time to randomization than were the original scans used in the MR CLEAN primary analysis. We assessed these scans for ASPECTS and used them for the analysis of these three patients because they better reflect the baseline infarct size before treatment than do the original scans. We graded ASPECTS on baseline non‐contrast CT images using previously described methods.7 We scored early ischemic changes, such as hypodensity within the basal ganglia and loss of grey‐white matter distinction in the cortical regions, according to the ten ASPECTS regions of the middle cerebral artery territory. For each involved region, we subtracted one point from a total possible score of ten (i.e., higher ASPECTS signifies a smaller infarct). Each non‐contrast CT scan was read by two independent readers from a panel of experts who were masked to treatment allocation and final outcome. A third reader provided a final adjudication in cases of discrepancies between readers in assignment of the trichotomized ASPECTS: 0–4 versus 5–7 versus 8–10. We graded collateral flow with baseline CT angiography using a 4‐point scale, with 0 representing absent collateral flow (absent filling of the occluded territory), 1 representing poor collateral flow (less than 50% filling of the occluded territory), 2 representing intermediate collateral flow (between 50% and 100% filling of the occluded territory), and 3 representing excellent collateral flow (100% filling of the occluded territory). We assessed arterial recanalization status with follow‐up CT angiography 24 h after symptom onset. We assessed follow‐up infarct volume on non‐contrast CT 5–7 days after symptom onset.10

Outcomes The primary outcome of MR CLEAN was the score on the modified Rankin Scale (mRS) at 90 days. The mRS is a 7‐point scale: 0 indicates no symptoms, 1 indicates no clinically significant disability, 2 indicates slight disability (able to look after own affairs without assistance, but unable to carry out all previous activities), 3 indicates moderate disability (requires some help, but able to walk unassisted), 4 indicates moderately


55

severe disability (unable to attend to bodily needs without assistance and unable to walk unassisted), 5 indicates severe disability (requires constant nursing care and attention), and 6 indicates that the patient is dead. A score of 2 points or fewer indicates functional independence.11 Secondary outcomes were 90‐day functional independence (mRS score of 0‐2), neurological assessment with the National Institutes of Health Stroke Score (NIHSS) at 24 h and at 5‐7 days or at discharge if patients were discharged before 5‐7 days, activities of daily living measured with the Barthel index, and health‐related quality of life measured with the EuroQol Group 5‐Dimension Self‐Report Questionnaire at 90 days. Radiological outcomes were arterial recanalization measured with CT angiography or MR angiography at 24 h, modified Thrombolysis in Cerebral Infarction (mTICI) score with digital subtraction angiography, and final infarct volume with non‐contrast CT at 5‐7 days. The mTICI score is a 5‐point scale, which ranges from 0 (no reperfusion) to 3 (complete antegrade reperfusion of the previously ischemic territory, with absence of visualized occlusion in all distal branches).12 Prespecified safety outcomes were hemorrhagic complications, progression of ischemic stroke, recurrent ischemic stroke, and death. Symptomatic intracranial hemorrhage was defined as neurological deterioration of 4 or more points on the NIHSS and confirmed with neuroimaging. Procedure‐related safety variables were vessel perforations and dissections and new embolus outside of the target downstream territory.12

Statistical analysis For the relation between baseline ASPECTS and functional outcome, we used multivariable ordinal logistic regression analysis (a proportional odds model) to calculate the adjusted common odds ratio (acOR) for a shift in direction towards a better functional outcome on the mRS for intra‐arterial treatment and usual care than for usual care alone. The common odds ratio (OR) is thus a proportional OR. For the relation between baseline ASPECTS and intra‐arterial treatment effect, we entered an interaction term into the model to test for interaction between treatment and the prespecified baseline ASPECTS subgroups: 0‐4 versus 5‐7 versus 8‐10 (as a categorical variable).7 Furthermore, we tested for interaction between treatment allocation and baseline numerical ASPECTS (as a continuous variable) for optimal statistical power. We adjusted the acOR and all secondary effect parameters for potential imbalances in major prespecified prognostic variables adapted from the original MR CLEAN trial protocol statistical analysis plan: age, stroke severity (NIHSS) at baseline, time since onset to randomization, previous stroke, atrial fibrillation, diabetes, and internal carotid artery terminus occlusion. We analyzed the independent effect of baseline ASPECTS on functional outcome and mortality with multivariable ordinal logistic regression analysis adjusted for the same prespecified variables and treatment allocation.

Chapter 3

56

We report acORs and unadjusted common ORs with 95% CIs. We analyzed binary outcomes with logistic regression and reported them as adjusted and unadjusted ORs with 95% CIs. We analyzed continuous outcomes with linear regression and reported adjusted and unadjusted with 95% CIs. We measured inter‐observer agreement for trichotomized ASPECTS scoring using the weighted statistic with quadratic weighting. We assessed agreement for baseline ASPECTS using the concordance correlation coefficient. We calculated means and SDs of differences in ASPECTS scores. Interpretation of these statistics followed the proposed standards of Landis and Koch.13 All p values are two sided, with significance at less than 0.05. We did statistical analyses with Stata (version 13.1).

Role of the funding source The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Results

Figure 3.1 shows the trial profile. Treatments in the intra‐arterial treatment group were mechanical thrombectomy (195 [84%] of 233 patients, of which 190 [97%] had stent retrievers and 24 [12%] received an additional intra‐arterial thrombolytic agent), an intra‐arterial thrombolytic agent alone (one [<1%]), and no intervention (37 [16%]). 496 patients — 232 (47%) in the intra‐arterial treatment and usual care group and 264 (53%) in the usual care alone group — were included in the prespecified subgroup analysis of MR CLEAN. ASPECTS was not graded in four (1%) of 500 patients in the MR CLEAN trial (one [<1%] in the intra‐arterial treatment and usual care group [anterior cerebral artery stroke] and three [1%] in the usual care alone group [two (1%) had an anterior cerebral artery stroke and one (<1%) had missing non‐contrast CT]) and these patients were excluded from the subgroup analysis. The baseline characteristics of the patients are summarized in Table 3.1 and the appendix. 30 (6%) of 496 patients had ASPECTS 0‐4 (11 [37%] in the intra‐arterial treatment and usual care group vs 19 [63%] in the usual care alone group), 93 (19%) had ASPECTS 5‐7 (54 [58%] vs. 39 [42%]), and 373 (75%) had ASPECTS 8–10 (167 [45%] vs. 206 [55%]). Patients with lower ASPECTS had higher median NIHSS, worse CT angiography collateral grades (combined grades 0‐1 16 [55%] of 29 patients with ASPECTS 0‐4, 34 [37%] of 91 with ASPECTS 5‐7, and 110 [30%] of 371 with ASPECTS 8‐10), higher rates of transfer from an outside hospital, longer median times from stroke onset to non‐contrast CT imaging, and a higher incidence of internal carotid artery terminus occlusions (Table 3.1).


57

Figure 3.1 Trial profile. NCCT=non‐contrast CT. We noted no differences in the proportion of patients who were treated with intravenous alteplase or the time to intravenous alteplase treatment between groups (Table 3.1). However, for patients with lower ASPECTS, we noted significantly faster times from qualifying non‐contrast CT imaging to intra‐arterial treatment and to angiographic reperfusion or final angiogram. Nevertheless, overall times from stroke onset to intra‐arterial treatment and to reperfusion or final angiogram were longer with lower ASPECTS (Table 3.1). Successful intra‐arterial revascularization (final mTICI 2B‐3) was achieved in four (40%) of ten patients with ASPECTS 0‐4, 29 (59%) of 49 patients with ASPECTS 5‐7, and 81 (60%) of 136 patients with ASPECTS 8‐10 (P=0.48).

Chapter 3

58

Table 3.1 Baseline characteristics.

ASPECTS 0‐4 (N=30)



P‐value

Age ‐ median (IQR) 62 (52‐69) 61 (52‐71) 67 (57‐77) 0 001 Male sex ‐ n (%) 20 (67) 61 (66) 209 (56) 0 069 NIHSS ‐ median (IQR) 21 (17‐22) 19 (17‐22) 17 (13‐21) 0 0001 Clinical localization: Left hemisphere ‐ n (%) 17 (57) 58 (62) 190 (51) 0 110 Atrial Fibrillation ‐ n (%) 5 (17) 21 (23) 109 (29) 0 063 History of Ischemic Stroke ‐ n (%) 1 (3) 13 (14) 40 (11) 0 680 History of Hypertension ‐ n (%) 15 (50) 36 (39) 174 (47) 0 600 History of Diabetes Mellitus ‐ n (%) 4 (13) 9 (10) 54 (15) 0 405 History of Myocardial Infarction ‐ n (%) 6 (20) 13 (14) 56 (15) 0 686 History of Peripheral Artery Disease ‐ n (%) 1 (3) 6 (7) 17 (5) 0 829 History of Hyperlipidemia ‐ n (%) 11 (37) 22 (24) 95 (26) 0 423 History of Smoking ‐ n (%) 11 (37) 26 (28) 105 (28) 0 471 Current Statin Use ‐ n (%) 10 (33) 23 (25) 109 (29) 0 890 Current Anticoagulant Use ‐ n (%) 0 (0) 9 (10) 30 (8) 0 382 Current Antiplatelet Use ‐ n (%) 9 (30) 24 (26) 111 (30) 0 680 Systolic blood pressure ‐ mean mmHg (SD) 141 5 (19) 145 9 (26) 145 6 (26) 0 546 Pre‐stroke modified Rankin Scale score ‐ n (%) 0 62 0 28 (93) 79 (85) 294 (79) 1 1 (3) 10 (11) 39 (11) 2 1 (3) 4 (4) 40 (11) Level of Occlusion ‐ n (%)** 0.10 ICA 0 (0) 0 (0) 4 (1) ICA‐T 12 (41) 35 (38) 87 (23) M1 17 (57) 55 (59) 246 (66) M2 0 (0) 3 (3) 36 (10) Collateral grade 0 02 Grade 0 3 (10) 8 (9) 15 (4) Grade 1 13 (45) 26 (29) 95 (26) Grade 2 11 (38) 38 (42) 149 (40) Grade 3 2 (7) 19 (21) 112 (30) Transferred from outside hospital – n (%) 26 (87) 65 (70) 127 (34) 0 0001 Treatment with IV alteplase ‐ n (%) 26 (87) 83 (89) 333 (89) 0 739 Onset to IV alteplase in min ‐ median (IQR) 90 (60‐135) 83 (65‐120) 85 (66‐110) 0 988 Onset to randomization in min ‐ median (IQR) 269 (219‐308) 234 (192‐276) 188 (145‐246) <0 0001 Onset to intra‐arterial treatment in min – median (IQR;n)

360 (240‐381;11) 286 (240‐315;51) 243 (205‐300;153) <0 0001

Onset to reperfusion or last angiogram in min ‐ median (IQR;n)

434 (378‐467;10) 354 (303‐404;50) 319 (260‐389;140) 0 0006

Onset to qualifying NCCT in min ‐ median (IQR;n) 241 (177‐291;30) 178 (99‐229;93) 93 (64‐172;373) 0 0001 Qualifying NCCT to randomization in min ‐ median (IQR;n)

26 (21‐35;30) 39 (24‐73;93) 64 (37‐100;371) 0 0001

Qualifying NCCT to intra‐arterial treatment in min ‐ median (IQR;n)

62 (55‐83;12) 96 (64‐141;51) 121 (77‐164;153) 0 0001

Qualifying NCCT to reperfusion or last angiogram in min ‐ median (IQR;n)

149 (121‐167;11) 171 (132‐215;50) 182 (150‐251;140) 0 004

Data are median (IQR), number (%), mean (SD), or median (IQR); number of patients with available data. ASPECTS=Alberta Stroke Program Early CT Score. NIHSS=National Institutes of Health Stroke Score. NCCT=non‐contrast CT. *One patient with ASPECTS 0–4 did not have pretreatment vessel imaging.

We noted no significant treatment‐by‐trichotomized ASPECTS interaction for the primary endpoint of ordinal mRS (acOR interaction term relative to ASPECTS 8‐10: ASPECTS 0‐4: 0.79 [95% CI 0.20‐3.19], P=0.740; ASPECTS 5‐7: 1.02 [0.44‐2.35], P=0.966; Table 3.2 and appendix). We also noted no interaction between ASPECTS and intra‐


59

arterial treatment effect when we analyzed by numerical ASPECTS (adjusted p for interaction=0.51 [data not shown]). acORs for improvement in 90‐day mRS with intra‐arterial treatment were 1.3 (95% CI 0.2‐9.3) for ASPECTS 0–4, 1.9 (0.9‐4.1) for ASPECTS 5–7, and 1.6 (1.1–2.3) for ASPECTS 8–10; Figure 3.2A). With the exception of the 24 h NIHSS, we noted no significant treatment‐by‐trichotomized ASPECTS interaction for any of the other secondary endpoints (Table 3.2 and appendix). In the adjusted analysis, 24 h NIHSS score was greater in the intra‐arterial treatment and usual care group than in the usual care alone group for ASPECTS 0‐4, whereas it was lower in the intra‐arterial treatment group for ASPECTS 8‐10 (Table 3.2). We noted no significant interaction for NIHSS score at 5‐7 days (Table 3.2). The distribution of 90‐day mRS scores is provided by both treatment allocation and ASPECTS subgroup in Figure 3.2B. Intra‐arterial treatment and usual care increased the proportion of patients with mRS 0‐2 scores by 9% (95% CI not applicable) for ASPECTS 0‐4, 19% (2‐35) for ASPECTS 5‐7, and 12% (3‐21) for ASPECTS 8‐10, compared with usual care alone. Similarly, intra‐arterial treatment resulted in an increase in mRS 0‐3 scores of 2% (‐26 to 30) for ASPECTS 0‐4, 18% (‐2 to 38) for ASPECTS 5‐7, and 15% (5‐25) for ASPECTS 8‐10. For intra‐arterial treatment and usual care compared with usual care alone, we noted a reduction at 90 days in mortality of 6% (‐30 to 42) for ASPECTS 0‐4 (four [36%] of 11 vs. eight [42%] of 19), 2% (‐16 to 19) for ASPECTS 5‐7 (13 [24%] of 54 vs. ten [26%] of 39), and 0% (‐8 to 8) for ASPECTS 8‐10 (32 [19%] of 167 vs. 40 [19%] of 206). Figure 3.2 Figure 2: Treatment effect for improvement at 90‐day mRS for the entire population and

trichotomized ASPECTS subgroups (A) and by treatment group per ASPECTS subgroup (B). The numbers and percentages of patients in the ASPECTS subgroups on part B are shown according to the distribution of the mRS. ASPECTS=Alberta Stroke Program Early CT Score. mRS=modified Rankin Scale.

Chapter 3

60

The overall (combined intra‐arterial treatment and usual care and usual care alone) number of patients with a 90‐day mRS score of 0‐2 was one (3%) of 30 in ASPECTS 0‐4, 20 (22%) of 93 in ASPECTS 5‐7, and 102 (27%) of 373 in ASPECTS 8‐10. Similarly, the number of patients with a 90‐day mRS score of 0‐3 was five (17%) of 30 for ASPECTS 0‐4, 41 (44%) of 93 for ASPECTS 5‐7, and 163 (44%) of 373 for ASPECTS 8‐10. For both mRS scores of 0‐2 and 0‐3, only the ASPECTS 0‐4 subgroup was significantly different from the other subgroups in pairwise comparisons (mRS 0‐2: ASPECTS 0‐4 vs. 5‐7 P=0.02, ASPECTS 0‐4 vs. 8‐10 P=0.004; mRS 0‐3: ASPECTS 0‐4 vs. 5‐7 P=0.009, ASPECTS 0‐4 vs. 8‐10 P=0.004). In a multivariable analysis including adjustment for treatment allocation, higher baseline trichotomized ASPECTS was associated with better 90‐day ordinal mRS score (relative to ASPECTS 8‐10: ASPECTS 0‐4 acOR: 0.38 [95% CI 0.16‐0.89]; ASPECTS 5‐7: 0.90 [0.48‐1.67]). Trichotomized ASPECTS was also an independent predictor of 90‐day mortality (relative to ASPECTS 8‐10: ASPECTS 0‐4: 5.84 [1.87‐18.22]; ASPECTS 5‐7: 1.92 [0.77‐4.84]), with 12 (40%) deaths in ASPECTS 0–4, 23 (25%) in ASPECTS 5‐7, and 72 (19%) in ASPECTS 8‐10 subgroups. Median final infarct volumes were. 143 ml (IQR 138–207) in ASPECTS 0‐4, 79 ml (41‐143) in ASPECTS 5‐7, and 56 ml (24‐107) in ASPECTS 8‐10 subgroups (P=0.001). Overall, more patients in the lower ASPECTS subgroups died within 7 days and within 30 days, had hemicraniectomy, and had at least one serious adverse event (Table 3.3). Among the serious adverse events, we noted more cases of ischemic stroke progression in patients with worse ASPECTS. We noted no significant differences in symptomatic intracerebral hemorrhage between ASPECTS subgroups. For death within 7 days or within 30 days and hemicraniectomy, the differences between the intra‐arterial treatment and usual care versus usual care alone groups were not significant for any of the ASPECTS subgroups (appendix). Intra‐arterial treatment did not result in a significant increase in patients with at least one serious adverse event in any of the ASPECTS subgroups (ASPECTS 0‐4: eight [73%] of 11 patients in treatment and usual care group vs. 11 [58%] of 19 for usual care alone group, P=0.42; ASPECTS 5‐7: 32 [59%] of 54 vs. 19 [49%] of 39, P=0.31; ASPECTS 8‐10: 70 [42%] of 167 vs. 82 [40%] of 206; P=0.68; appendix). A significantly higher proportion of patients had recurrent ischemic stroke in the intra‐arterial treatment plus usual care group than in the usual care alone group in the ASPECTS 8‐10 subgroup (eight [5%] patients vs one [<1%] patient; P=0.007; appendix). The observed agreement for trichotomized ASPECTS was 384 (78%) of 494, yielding a weighted of 0.56 (SE 0.04). The concordance correlation coefficient for numerical ASPECTS was 0.62 (95% CI 0.57‐0.67). The mean difference in scores was 0.0 (SD 1.7).


61

Table 3.2

Treatm

ent e

ffect by tricho

tomize

d AS

PECT

S grou

p.

Outcome

Effect

parameter

0‐4

Unadjusted

effect

(95%

CI)

0‐4

Adjusted

effect

(95%

CI)

5‐7

Unadjusted

effect

(95%

CI)

5‐7

Adjusted

effect

(95%

CI)

8‐10

Un

adjusted

effect

(95%

CI)

8‐10

Ad

justed

effect

(95%

CI)

Unadjusted

P‐value for

interaction

0‐4 vs. 8

‐10

Una

djusted

P ‐value for

interaction

5‐7 vs. 8

‐10

Adjusted

P‐value for

interaction

0‐4 vs. 8‐10

Adjusted

P‐value for

interaction

5‐7 vs. 8

‐10

Primary

mRS

at 9

0 da

ys

cOR*

13

(03 to 52)

13

(02 to 93)

17

(08 to 35)

19

(09 to 41)

17

(12 to 24)

16

(11 to 23)

0.69

4 0.91

6 0.740

0.96

6

Second

ary ‐ clinical

mRS

0‐1

OR*

* n/a

n/a

07

(01 to 37)

06

(01 to 41)

25

(12 to 51)

24

(11 to 49)

n/a

0.17

0 n/a

0.15

3

mRS

0‐2

OR*

* n/a

n/a

31

(1 to

94)

3

(09 to 96)

19

(12 to 29)

18

(11 to 3)

n/a

0.39

1 n/a

0.35

8

mRS

0‐3

OR*

* 12

(02 to 85)

12

(0 to

44)

21

(09 to 48)

23

(09 to 58)

19

(12 to 28)

19

(12 to 31)

0.66

0 0.82

3 0.825

0.75

7

NIHSS after 24 ho

urs –

mean (SD)

Beta***

64

(06 to 123)

52

(‐26 to 131)

‐51

(‐82 to ‐2

) ‐4

7 (‐7

5 to ‐1

9)

‐26

(‐43 to ‐0

9)

‐2

(‐34 to ‐0

6)

0.00

5 0.20

7 0.003

0.14

2

NIHSS at 5

‐7 days o

r discharge – mean (SD)

Beta***

08

(‐75 to 9)

‐21

(‐15 to 109)

‐52

(‐9 to

‐13)

‐46

(‐84 to ‐0

8)

‐31

(‐49 to ‐1

4)

‐24

(‐4 to

‐09)

0.31

1 0.34

0 0.529

0.16

6

Second

ary ‐ rad

iological

CTA follow‐up available

in – n. / to

tal n.

21/30

21/30

67/93

67/93

302/373

302/373

No intra‐cran

ial

occlusion on

follow‐up

CTA^

OR*

* 49

(07 to 35)

n/a

61

(21 to 181)

74

(22 to 253)

66

(39 to 109)

7 (4

1 to 12)

0.77

6 0.91

2 0.967

0.71

2

Final infarct volum

e on

NCC

T follow‐up available

in – n. / to

tal n.

11/30

11/30

48/93

48/93

245/373

245/373

Vo

lume in m

illilitres –

mean (SD)

Beta***

‐290

(‐127

6 to 697)

10

(‐166

0 to 167

9)

‐374

(‐837 to 9)

‐414

(‐874 to 45)

‐163

(‐34 to 13)

‐123

(‐296 to 5)

0.82

0 0.36

8 0.562

0.15

3

* Odd

s of improved

outcome (i.e., shift to lower m

RS) w

ith intra‐arteria

l treatment; **

Odd

s of achieving

end

point w

ith intra‐arteria

l treatment; *** Mean diffe

rence

of variable (in

tra‐arteria

l treatment v

alue

minus con

trol value

). n/a: not eno

ugh data available to calculate statistical param

eter.

Chapter 3

62

Table 3.3 Serious adverse events by trichotomized ASPECTS group.




P‐value

Safety parameters Death within 7 days ‐ n (%) 10 (33 3) 18 (19 4) 31 (8 3) <0 0001 Death within 30 days ‐ n (%) 10 (33 3) 22 (23 7) 60 (16 1) 0 007 Hemicraniectomy ‐ n (%) 5 (16 7) 6 (6 5) 16 (4 3) 0 009 Serious Adverse Events Patients with at least one SAE ‐ n (%) 19 (63 3) 51 (54 8) 152 (40 8) 0 001 Symptomatic ICH ‐ n (%) 2 (6 7) 8 (8 6) 25 (6 7) 0 715 Parenchymal hematoma type 1 (PH1) ‐ n (%) 0 (0) 0 (0) 2 (0 54) 0 051 Parenchymal hematoma type 2 (PH2) ‐ n (%) 1 (3 3) 7 (7 5) 20 (5 4) 0 748 Hemorrhagic infarction type 1 (HI1) ‐ n (%) 0 (0) 0 (0) 1 (0 27) 0 593 Haemorrhagic infarction type 2 (HI2) ‐ n (%) 0 (0) 1 (1 1) 1 (0 27) 0 639 Subarachnoid hemorrhage ‐ n (%) 1 (3 3) 0 (0) 1 (0 27) 0 090 Recurrent ischemic stroke ‐ n (%) 2 (6 7) 3 (3 2) 9 (2 4) 0 209 Progression of Ischemic Stroke ‐ n (%) 12 (40) 24 (25 8) 57 (15 3) 0 0001 Pneumonia ‐ n (%) 1 (3 3) 13 (14) 56 (15) 0 141 Other Infection ‐ n (%) 1 (3 3) 3 (3 2) 22 (5 9) 0 293 Cardiac Ischemia ‐ n (%) 0 (0) 0 (0) 5 (1 3) 0 231 Extracranial Hemorrhage ‐ n (%) 0 (0) 1 (1 1) 1 (0 3) 0 639 Allergic Reaction ‐ n (%) 0 (0) 1 (1 1) 0 (0) 0 231 Other Complication ‐ n (%) 4 (13 3) 19 (20 4) 42 (11 3) 0 110

Discussion

This prespecified subgroup analysis of MR CLEAN does not support exclusion of patients from intra‐arterial treatment on the basis of infarct volume assessed with use of ASPECTS on baseline non‐contrast CT. Specifically, we found no significant modification of intra‐arterial treatment effect by trichotomized ASPECTS. Additionally, we did not find an interaction for the primary endpoint when we analyzed by numerical ASPECTS. Moreover, baseline infarct volumes did not affect the safety of intra‐arterial treatment. The PROACT II investigators found evidence of an ASPECTS‐by‐treatment interaction using dichotomized ASPECTS (>7 vs. 7).3 A five‐times higher proportion of patients with small infarcts (ASPECTS >7) had good functional outcome with intra‐arterial treatment than did controls, whereas patients with large infarcts (ASPECTS 7) had no treatment benefit. Data from IMS III4 did not substantiate these findings, although the reason that these results might not have agreed could have been related to the overall neutral results of the trial as well as the early timing of the non‐contrast CT scans, which could have impaired the reliability of infarct assessment.14 Unlike the PROACT II analysis, our findings suggest that patients with baseline ASPECTS 5‐7 (i.e; moderate sized infarcts) have a treatment benefit. In addition to the principal finding that no significant modification of intra‐arterial treatment effect exists between ASPECTS 5‐7 and 8‐10, we also noted a significant benefit of intra‐arterial therapy in the


63

ASPECTS 5‐7 subgroup for a 90‐day mRS score of 0‐2 in the unadjusted analysis. Although these results are encouraging, they should be substantiated in a future study. In terms of overall prognosis, slightly fewer patients with ASPECTS 5‐7 had good functional outcome than in the ASPECTS 8‐10 group and an equivalent number had moderate functional outcome, consistent with an analysis of the pooled PICS and Pivotal dataset7 and further highlighting their similar clinical behavior. These findings have immediate relevance to clinical practice, because up to 40% of centers that use ASPECTS grading have a threshold of 7 (or higher) for treatment exclusion.6 Notably, many of these patients were excluded in randomized intra‐arterial treatment trials. The REVASCAT15 trial excluded patients with non‐contrast CT ASPECTS of lower than 7, whereas ESCAPE16 and SWIFT PRIME17 excluded patients with ASPECTS less than 6. On the basis of previous work, we anticipated that patients with ASPECTS 0‐4 would have a markedly reduced benefit of intra‐arterial treatment.7 Ample evidence in patients who had a stroke and anterior circulation proximal occlusions supports the idea that large final infarct volumes are determinants of long‐term dependence or death.18,19 Because the direct goal of early reperfusion therapy is to prevent infarct growth and thus limit final infarct size, patients who present with extensive infarcts at baseline would seem to be unlikely to benefit from reperfusion and might even be harmed.7,20,21 Nevertheless, the absence of a significant treatment‐by‐ASPECTS interaction in this study prohibits the recommendation to exclude patients with ASPECTS 0‐4 from intra‐arterial treatment. Conversely, one should be wary of acceptance of a potentially false‐negative result (i.e., falsely accepting that a homogeneous treatment benefit exists across the ASPECTS subgroups when in fact a heterogeneous response exists). In light of the strong a‐priori justification for a reduced treatment effect among large infarcts and the small number of patients with ASPECTS 0‐4 in MR CLEAN, our analysis is very likely to be underpowered to adequately test treatment modification in this group. This problem prevails in subgroup analyses, given that trials are powered on the basis of overall treatment effect rather than interaction effects.22,23 Moreover, the false‐negative risk is particularly high when more than two subgroups exist and unequal subgroup sizes exist, as in this case.23 Indeed, in the ASPECTS 0‐4 subgroup, the 9% absolute difference in good outcome in favor of intra‐arterial treatment represented a difference of only one patient. Furthermore, for moderate outcome, the absolute benefit of intra‐arterial treatment was only 2% and represented a difference of only one patient between groups. Further data are needed to precisely estimate the intra‐arterial treatment effect among patients with ASPECTS 0‐4. The difference between relative and absolute benefit is another crucial issue with regard to the ASPECTS 0‐4 subgroup. Absolute benefit is a more clinically relevant measure of treatment effect than relative benefit because it establishes the number needed to treat.22 Even if a relative treatment benefit was shown for patients with

Chapter 3

64

ASPECTS 0‐4, the absolute benefit would be very likely to be marginal given the overall poor prognosis of these patients despite treatment. Trichotomized ASPECTS was an independent predictor of functional outcome and mortality in MR CLEAN, owing primarily to the dismal outcomes in the ASPECTS 0‐4 group. Overall, fewer than 5% of patients with ASPECTS 0‐4 achieved independence, and 40% were dead by 3 months, which was a consequence of the large final infarct volumes in this population. This finding was mirrored in the pooled PICS and Pivotal dataset,7 for which the proportion of independence was 5% and mortality was 55% of 40 patients with ASPECTS 0‐4 who had thrombectomy. Similarly, the IMS III investigators found that baseline ASPECTS was a strong prognostic variable, with patients with ASPECTS 0‐4 faring the worst.4 However, they found a good functional outcome proportion of 21% of 92 patients with ASPECTS 0‐4, which was much higher than they expected. Further data are needed to understand this heterogeneity, but infarct measurement in IMS III could possibly have been impaired by the early imaging time points and substantial variability in imaging quality that was noted by the investigators. If the good functional outcome proportion is confirmed to be very low, the absolute treatment benefit would necessarily have to be small, further raising the question of whether intra‐arterial treatment is cost‐effective in this population. Finally, no benefit of intra‐arterial treatment in PROACT II in the ASPECTS subgroup of less than or equal to 7 might be related to a high proportion of patients with ASPECTS 0‐4 in that study. As expected, baseline ASPECTS was associated with several markers of stroke severity, including NIHSS, collateral strength, and the presence of terminus occlusion in the internal carotid artery. However, the only modifiable risk factor was longer time from stroke onset to non‐contrast CT. Patients with ASPECTS 0‐4 had much longer times from stroke onset to non‐contrast CT imaging than did the other two groups, providing direct evidence that the deleterious effect of delayed treatment is mediated through infarct expansion. This delay is probably due, in part, to interfacility transfer time, as patients with lower ASPECTS had a significantly higher proportion of transfers from an outside hospital. Patient transfer from primary to comprehensive stroke centers for consideration of intra‐arterial treatment might take as long as a median of 90‐100 min, and during this time patients might move outside of the treatment window or sustain greater infarct burden than those who present directly to a comprehensive stroke center.24,25 These findings underscore the importance of identification of the most appropriate triage mechanisms for patients with stroke. In many circumstances, those patients deemed likely to have a proximal artery occlusion would be best served by bypassing the primary stroke center and going directly to the nearest comprehensive stroke center for possible intra‐arterial treatment. An important consideration for adoption of ASPECTS scoring for intra‐arterial treatment selection in clinical practice is the issue of inter‐observer agreement. Similar to a previous study26 of patients with large‐vessel occlusion who had intra‐arterial therapy, we found substantial inter‐observer agreement along the entire ASPECTS scale.


65

Agreement for trichotomized categories was moderate. These results are encouraging, but probably overestimate the reliability of ASPECTS in the real‐world setting, given that the reads in MR CLEAN were done by highly experienced neuroradiologists without the time pressures encountered in the acute setting. Further work is necessary to improve inter‐observer agreement in clinical practice, and adoption of standards for training and certification should be the ultimate goal. The main limitation of this analysis is the small number of patients in the ASPECTS 0‐4 category, which probably resulted in an underpowered test for interaction between ASPECTS and treatment allocation. However, because MR CLEAN was the only intra‐arterial treatment trial that did not have an explicit imaging exclusion criterion based on the extent of parenchymal ischemic changes, our study population was the largest with ASPECTS 0‐4 among the recent randomized trials of intra‐arterial therapy. Despite this limitation, we had sufficient numbers to show that trichotomized ASPECTS was an independent predictor of functional outcome and mortality. This prespecified subgroup analysis of MR CLEAN suggests that patients with baseline ASPECTS 5‐7 should not be excluded from intra‐arterial treatment, as investigators of other studies have proposed. Further data are needed to better estimate intra‐arterial treatment effect in patients with ASPECTS 0‐4. Our findings show a high probability of poor outcome in this group regardless of treatment, which might constrain the potential absolute treatment benefit and limit the cost‐effectiveness of intra‐arterial treatment in these patients.

Chapter 3

66

References

1. Yoo AJ, Verduzco LA, Schaefer PW, Hirsch JA, Rabinov JD, Gonzalez RG. MRI‐based selection for intra‐arterial stroke therapy: value of pretreatment diffusion‐weighted imaging lesion volume in selecting patients with acute stroke who will benefit from early recanalization. Stroke 2009;40:2046‐54.

2. Olivot JM, Mosimann PJ, Labreuche J, et al. Impact of diffusion‐weighted imaging lesion volume on the success of endovascular reperfusion therapy. Stroke 2013;44:2205‐11.

3. Hill MD, Rowley HA, Adler F, et al. Selection of acute ischemic stroke patients for intra‐arterial thrombolysis with pro‐urokinase by using ASPECTS. Stroke 2003;34:1925‐31.

4. Hill MD, Demchuk AM, Goyal M, et al. Alberta Stroke Program early computed tomography score to select patients for endovascular treatment: Interventional Management of Stroke (IMS)‐III Trial. Stroke 2014;45:444‐9.

5. Sheth KN, Terry JB, Nogueira RG, et al. Advanced modality imaging evaluation in acute ischemic stroke may lead to delayed endovascular reperfusion therapy without improvement in clinical outcomes. J Neurointerv Surg 2013;5 Suppl 1:i62‐5.

6. Mehta B, Leslie‐Mazwi TM, Chandra RV, et al. Assessing variability in neurointerventional practice patterns for acute ischemic stroke. J Neurointerv Surg 2013;5 Suppl 1:i52‐7.

7. Yoo AJ, Zaidat OO, Chaudhry ZA, et al. Impact of pretreatment noncontrast CT Alberta Stroke Program Early CT Score on clinical outcome after intra‐arterial stroke therapy. Stroke 2014;45:746‐51.

8. Fransen PS, Beumer D, Berkhemer OA, et al. MR CLEAN, a multicenter randomized clinical trial of endovascular treatment for acute ischemic stroke in the Netherlands: study protocol for a randomized controlled trial. Trials 2014;15:343.

9. Berkhemer OA, Fransen PS, Beumer D, et al. A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med 2015;372:11‐20.

10. Boers AM, Marquering HA, Jochem JJ, et al. Automated cerebral infarct volume measurement in follow‐up noncontrast CT scans of patients with acute ischemic stroke. AJNR Am J Neuroradiol 2013;34:1522‐7.

11. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke 1988;19:604‐7.

12. Zaidat OO, Yoo AJ, Khatri P, et al. Recommendations on angiographic revascularization grading standards for acute ischemic stroke: a consensus statement. Stroke 2013;44:2650‐63.

13. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159‐74.

14. Bal S, Bhatia R, Menon BK, et al. Time dependence of reliability of noncontrast computed tomography in comparison to computed tomography angiography source image in acute ischemic stroke. Int J Stroke 2015;10:55‐60.

15. Jovin TG, Chamorro A, Cobo E, et al. Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med 2015;372:2296‐306.

16. Goyal M, Demchuk AM, Menon BK, et al. Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med 2015;372:1019‐30.

17. Saver JL, Goyal M, Bonafe A, et al. Solitaire with the intention for thrombectomy as primary endovascular treatment for acute ischemic stroke (SWIFT PRIME) trial: protocol for a randomized, controlled, multicenter study comparing the solitaire revascularization device with IV tPA with IV tPA alone in acute ischemic stroke. Int J Stroke 2015;10:439‐48.

18. Yoo AJ, Chaudhry ZA, Nogueira RG, et al. Infarct volume is a pivotal biomarker after intra‐arterial stroke therapy. Stroke 2012;43:1323‐30.

19. Zaidi SF, Aghaebrahim A, Urra X, et al. Final infarct volume is a stronger predictor of outcome than recanalization in patients with proximal middle cerebral artery occlusion treated with endovascular therapy. Stroke 2012;43:3238‐44.

20. Singer OC, Humpich MC, Fiehler J, et al. Risk for symptomatic intracerebral hemorrhage after thrombolysis assessed by diffusion‐weighted magnetic resonance imaging. Ann Neurol 2008;63:52‐60.

21. Lansberg MG, Thijs VN, Bammer R, et al. Risk factors of symptomatic intracerebral hemorrhage after tPA therapy for acute stroke. Stroke 2007;38:2275‐8.


67

22. Kent DM, Rothwell PM, Ioannidis JP, Altman DG, Hayward RA. Assessing and reporting heterogeneity in treatment effects in clinical trials: a proposal. Trials 2010;11:85.

23. Brookes ST, Whitely E, Egger M, Smith GD, Mulheran PA, Peters TJ. Subgroup analyses in randomized trials: risks of subgroup‐specific analyses; power and sample size for the interaction test. J Clin Epidemiol 2004;57(3):229‐36.

24. Prabhakaran S, Ward E, John S, et al. Transfer delay is a major factor limiting the use of intra‐arterial treatment in acute ischemic stroke. Stroke 2011;42:1626‐30.

25. Sun CH, Connelly K, Nogueira RG, et al. ASPECTS decay during inter‐facility transfer predicts patient outcomes in endovascular reperfusion for ischemic stroke: a unique assessment of dynamic physiologic change over time. J Neurointerv Surg 2015;7:22‐6.

26. Gupta AC, Schaefer PW, Chaudhry ZA, et al. Interobserver reliability of baseline noncontrast CT Alberta Stroke Program Early CT Score for intra‐arterial stroke treatment selection. AJNR Am J Neuroradiol 2012; 33:1046‐9.

Chapter 3

68

Supplemental material

Table S3.1

Baseline characteristics b

y tricho

tomize

d AS

PECTS and treatm

ent a

llocatio

n.

0‐4

Interven

tion

(N=1

1)

0‐4

Control

(N=19)

5‐7

Interven

tion

(N=54)

5‐7

Control

(N=39)

8‐10

Interven

tion

(N=167)

8‐10

Co

ntrol

(N=206)

P‐value

0‐4

P‐value

5‐7

P‐value

8‐10

Age ‐ m

edian (IQ

R)

604

(495‐68

6)

636

(545‐70

9)

602

(519‐72

8)

611

(521‐68

7)

675

(566‐76

6)

666

(567‐77)

0897

0832

084

7

Male sex ‐ n

(%)

6 (545)

14 (7

37)

36 (6

67)

25 (6

41)

93 (5

57)

116 (563)

0284

0797

090

4 NIHS

S ‐ m

edian (IQ

R)

19 (1

7‐22)

21 (1

7‐22)

19 (1

7‐22)

20 (1

6‐23)

16 (1

3‐21)

17 (1

4‐22)

0778

0343

026

5 Clinical localization: Le

ft hem

isphe

re ‐ n (%

) 4 (364)

13 (6

84)

30 (5

56)

28 (7

18)

81 (4

85)

109 (529)

0088

0111

039

7 Atria

l Fibrillatio

n ‐ n

(%)

1 (9

1)

4 (211)

12 (2

22)

9 (231)

53 (3

17)

56 (2

72)

0397

0923

033

6 History of Ischem

ic Stroke ‐ n

(%)

1 (9

1)

0 (0)

8 (148)

5 (128)

20 (1

2)

20 (9

7)

0181

0784

048

2 History of Hypertension ‐ n

(%)

3 (273)

12 (6

32)

22 (4

07)

14 (3

59)

72 (4

31)

102 (495)

0058

0636

021

8 History of Diabe

tes M

ellitus ‐ n (%

) 2 (182)

2 (105)

6 (111)

3 (7

7)

26 (1

56)

28 (1

36)

0552

0582

059

0 History of M

yocardial Infarction ‐ n

(%)

2 (182)

4 (211)

7 (13)

6 (154)

24 (1

44)

32 (1

55)

0850

0740

075

5 History of Peripheral A

rtery Disease ‐ n

(%)

0 (0)

1 (5

3)

4 (7

4)

2 (5

1)

4 (2

4)

13 (6

3)

0439

0659

007

1 History of Hyperlipidaemia ‐ n (%)

6 (545)

5 (263)

13 (2

41)

9 (231)

39 (2

34)

56 (2

72)

0122

0911

039

8 History of Smoking ‐ n

(%)

2 (182)

9 (474)

16 (2

96)

10 (2

56)

47 (2

81)

58 (2

82)

0110

0672

099

8 Cu

rren

t Statin

Use ‐ n (%

) 4 (364)

6 (316)

15 (2

78)

8 (205)

46 (2

75)

63 (3

06)

0789

0423

052

1 Cu

rren

t Anticoa

gulant Use ‐ n (%)

0 (0)

0 (0)

5 (9

3)

4 (103)

13 (7

8)

17 (8

3)

‐ 0872

086

9 Cu

rren

t Antiplatelet U

se ‐ n (%

) 3 (273)

6 (316)

13 (2

41)

11 (2

82)

48 (2

87)

63 (3

06)

0804

0653

069

9 Systolic blood

pressure ‐ m

ean mmHg (SD)

1455 (218)

1392 (177)

1491 (276)

1415 (235)

1449 (259)

1461 (251)

0400

0165

063

2 Pre‐stroke mod

ified

Rankin Scale score ‐ n

(%)

0538

0365

090

2 0

11 (1

00)

17 (8

95)

44 (8

15)

35 (8

97)

135 (808)

159 (772)

1

0 (0)

1 (5

3)

7 (13)

3 (7

7)

14 (8

4)

25 (1

21)

2 0 (0)

1 (5

3)

3 (1

6)

1 (2

6)

18 (1

08)

22 (1

07)

Treatm

ent w

ith IV

alte

plase ‐ n

(%)

10 (9

09)

16 (8

42)

46 (8

52)

37 (9

49)

147 (88)

186 (903)

0603

0137

048

2 Onset to

IV alte

plase in m

in ‐ med

ian (IQ

R)

845 (56‐11

0) 91

5 (65‐157)

80 (6

0‐110)

87 (7

3‐120)

85 (7

0‐110)

86 (6

5‐113)

0398

0283

093

2 Level of O

cclusio

n ‐ n

(%)

0668

0455

027

3 ICA

0 (0)

0 (0)

0 (0)

0 (0)

1 (0

6)

3 (1

5)

ICA‐T

4 (364)

8 (421)

23 (4

26)

12 (3

08)

32 (1

92)

55 (2

67)

M1

7 (636)

10 (5

26)

29 (5

37)

26 (6

67)

118 (707)

128 (621)

M2

0 (0)

0 (0)

2 (3

7)

1 (2

6)

16 (9

6)

20 (9

7)


69

Table S3.1

(con

tinue

d)

0‐4

Interven

tion

(N=1

1)

0‐4

Control

(N=19)

5‐7

Interven

tion

(N=54)

5‐7

Control

(N=39)

8‐10

Interven

tion

(N=167)

8‐10

Co

ntrol

(N=206)

P‐value

0‐4

P‐value

5‐7

P‐value

8‐10

Collateral grade

0679

0106

065

6 Grade

0

1 (9

1)

2 (111)

2 (3

8)

6 (158)

6 (3

6)

9 (4

4)

Grade

1

5 (455)

8 (444)

19 (3

89)

7 (184)

47 (2

83)

48 (2

34)

Grade

2

5 (455)

6 (333)

21 (3

96)

17 (4

47)

62 (3

74)

87 (4

24)

Grade

3

0 (0)

2 (111)

11 (2

08)

8 (211)

51 (3

07)

61 (2

98)

Onset to

rand

omisa

tion in m

in ‐ med

ian (IQ

R)

316

(218

‐349)

263

(219

‐279)

232

(191

‐273)

234

(192

‐276)

189

(148

‐238)

1865

(142

‐256)

0037

0901

077

5

Onset to

intra‐arteria

l treatmen

t in min

– med

ian (IQ

R;n)

360

(240

‐381;11)

n/a

286

(240

‐315;51)

n/a

243

(205

‐300

;153)

n/a

‐ ‐

‐

Onset to

repe

rfusion or last angiogram

in m

in

‐ med

ian (IQ

R;n)

434

(378

‐467;10)

n/a

354

(303

‐404;50)

n/a

319

(260

‐389;140)

n/a

‐ ‐

‐

Onset to

qualifying

NCC

T in m

in ‐ med

ian (IQ

R;n)

300

(177

‐325;11)

235

(161

‐252

;19)

180

(87‐229;54)

178

(99‐232;39)

93

(63‐172;167)

93 (6

4‐172;20

6)

0019

0867

075

5

Qualifying

NCC

T to ra

ndom

isatio

n in m

in

‐ med

ian (IQ

R;n)

28

(24‐32

;11)

24

(19‐36;19)

40

(26‐73;54)

38

(22‐72;39)

60

(37‐98;165)

66 (3

7‐100;20

6)

06207

09907

08322

Qualifying

NCC

T to intra‐arteria

l treatmen

t in min

‐ med

ian (IQ

R;n)

62 (5

5‐83;12)

n/a

96 (6

4‐14

1;51)

n/a

121

(77‐164;153)

n/a

‐ ‐

‐

Qualifying

NCC

T to rep

erfusio

n or last ang

iogram

in m

in ‐ med

ian (IQ

R;n)

149

(121

‐167;11)

n/a

171

(132

‐215;50)

n/a

182

(150

‐251

;140)

n/a

‐ ‐

‐

n/a: data no

t available

Chapter 3

70

Table S3.2 The odds ratio estimates and 95% confidence intervals for the interaction terms, unadjusted (top) and adjusted (bottom).

Outcome ASPECTS subgroup. in reference to ASPECTS 8‐10

OR* interaction term (95% CI)

P‐value interaction term

Unadjusted Primary

0‐4 0 76 (0 20 to 2 95) 0 694 mRS at 90 days 5‐7 0 96 (0 43 to 2 15) 0 916

Secondary ‐ clinical 0‐4 n/a n/a mRS 01 5‐7 0 28 (0 05 to 1 72) 0 170 0‐4 n/a n/a mRS 02 5‐7 1 69 (0 51 to 5 55) 0 391 0‐4 0 64 (0 09 to 4 76) 0 660 mRS 03 5‐7 1 11 (0 43 to 2 85) 0 823 0‐4 9 04 (2 76 to 15 3) 0 005 NIHSS after 24 hours* 5‐7 ‐2 47 (‐6 30 to 1 37) 0 207 0‐4 3 92 (‐3 68 to 11 52) 0 311 NIHSS at 5‐7 days or discharge* 5‐7 ‐2 04 (‐6 23 to 2 16) 0 340

Secondary ‐ radiology 0‐4 0 54 (0 14 to 2 02) 0 360 No intra‐cranial occlusion on follow‐up CTA 5‐7 0 88 (0 36 to 2 15) 0 781 0‐4 ‐12 66 (‐122 28 to 96 95) 0 820 Volume in milliliters on NCCT follow‐up* 5‐7 ‐21 05 (‐66 99 to 23 89) 0 368

Adjusted Primary

0‐4 0 79 (0 20 to 3 19) 0 740 mRS at 90 days 5‐7 1 02 (0 44 to 2 35) 0 966

Secondary ‐ clinical 0‐4 n/a n/a mRS 01

5‐7 0 26 (0 04 to 1 65) 0 153 0‐4 n/a n/a mRS 02

5‐7 1 82 (0 51 to 6 49) 0 358 0‐4 0 79 (0 10 to 6 49) 0 825 mRS 03

5‐7 1 18 (0 42 to 3 32) 0 757 0‐4 8 01 (2 65 to 13 36) 0 003 NIHSS after 24 hours* 5‐7 ‐2 45 (‐5 73 to 0 82) 0 142 0‐4 2 17 (‐4 61 to 8 95) 0 529 NIHSS at 5‐7 days or discharge* 5‐7 ‐2 65 (‐6 41 to 1 10) 0 166

Secondary ‐ radiological 0‐4 1 04 (0 13 to 8 43) 0 967 No intra‐cranial occlusion on follow‐up CTA 5‐7 1 26 (0 37 to 4 34) 0 712 0‐4 ‐31 35 (‐137 57 to 74 87) 0 562 Volume in milliliters on NCCT follow‐up* 5‐7 ‐32 65 (‐77 44 to 12 15) 0 153

* beta


71

Table S3.3

Serio

us adverse events b

y tricho

tomize

d AS

PECT

S and treatm

ent a

llocatio

n.

0‐4

Interven

tion

(N=11)

0‐4

Control

(N=19)

5‐7

Interven

tion

(N=54)

5‐7

Control

(N=39)

8‐10

Interven

tion

(N=167

)

8‐10

Co

ntrol

(N=206

)

P‐value

0‐4

P‐value

5‐7

P‐value

8‐10

Safety Param

eters

De

ath with

in 7 days ‐ n (%

) 4 (364)

6 (316)

10 (1

85)

8 (205)

19 (1

14)

12 (5

8)

079

081

005

De

ath with

in 30 days ‐ n (%

) 4 (364)

6 (316)

12 (2

22)

10 (2

56)

28 (1

68)

32 (1

55)

079

070

075

He

micraniectomy ‐ n

(%)

3 (273)

2 (105)

4 (7

4)

2 (5

1)

7 (4

2)

9 (4

4)

024

066

093

Serio

us Adverse Events

Patie

nts w

ith at least one

SAE

‐ n (%

) 8 (727)

11 (5

79)

32 (5

93)

19 (4

87)

70 (4

19)

82 (3

98)

042

031

068

Symptom

atic IC

H ‐ n

(%)

1 (9

1)

1 (5

3)

7 (13)

1 (2

6)

10 (6

) 15

(73)

069

008

062

Paren

chym

al hem

atom

a type

1 (P

H1) ‐ n (%

) 0 (0)

0 (0)

0 (0)

0 (0)

0 (0)

2 (1

0)

‐ ‐

020

Paren

chym

al hem

atom

a type

2 (P

H2) ‐ n (%

) 0 (0)

1 (5

3)

6 (111)

1 (2

6)

8 (4

8)

12 (5

8)

044

012

066

Ha

emorrhagic infarctio

n type

1 (H

I1) ‐ n (%

) 0 (0)

0 (0)

0 (0)

0 (0)

1 (0

6)

0 (0)

‐ ‐

027

Ha

emorrhagic infarctio

n type

2 (H

I2) ‐ n (%

) 0 (0)

0 (0)

1 (1

9)

0 (0)

0 (0)

1 (0

5)

‐ 039

037

Subarachno

id haemorrhage ‐ n

(%)

1 (9

1)

0 (0)

0 (0)

0 (0)

1 (0

6)

0 (0)

018

‐

027

Re

curren

t ische

mic stroke ‐ n (%

) 2 (182)

0 (0)

3 (5

6)

0 (0)

8 (4

8)

1 (0

5)

005

014

000

7 Progression of Ischem

ic Stroke ‐ n

(%)

4 (364)

8 (421)

13 (2

41)

11 (2

82)

29 (1

74)

28 (1

36)

076

065

031

Pn

eumon

ia ‐ n (%

) 0 (0)

1 (5

3)

7 (13)

6 (154)

20 (1

2)

36 (1

75)

044

074

014

Other Infection ‐ n

(%)

1 (9

1)

0 (0)

1 (1

9)

2 (5

1)

14 (8

4)

8 (3

9)

018

038

007

Cardiac Ischem

ia ‐ n (%

) 0 (0)

0 (0)

0 (0)

0 (0)

1 (0

6)

4 (1

9)

‐ ‐

026

Extracranial Haemorrhage ‐ n

(%)

0 (0)

0 (0)

0 (0)

1 (2

6)

0 (0)

1 (0

5)

‐ 024

037

Allergic Reaction ‐ n

(%)

0 (0)

0 (0)

1 (1

9)

0 (0)

0 (0)

0 (0)

‐ 039

‐

Other Com

plication ‐ n

(%)

1 (9

1)

3 (158)

10 (1

85)

9 (231)

17 (1

02)

25 (1

21)

060

059

055

UvA-DARE (Digital Academic Repository) Intraarterial ... · van Zwam WH°, Majoie CBLM°, for the...

Documents

Transcript of UvA-DARE (Digital Academic Repository) Intraarterial ... · van Zwam WH°, Majoie CBLM°, for the...