[Skip to Navigation]
Sign In
Figure.  Functional Outcomes at 90 Days According to the Score on the Modified Rankin Scale (mRS) (Baseline Ischemic Core >50 mL on Computed Tomographic Perfusion)
Functional Outcomes at 90 Days According to the Score on the Modified Rankin Scale (mRS) (Baseline Ischemic Core >50 mL on Computed Tomographic Perfusion)

Shift analysis by Wilcoxon signed rank test, P = .04.

Table 1.  Baseline and Procedural Characteristics of Patients With Baseline Ischemic Cores Greater Than 50 mLa
Baseline and Procedural Characteristics of Patients With Baseline Ischemic Cores Greater Than 50 mLa
Table 2.  Outcomes for Patients With Baseline Ischemic Cores Greater Than 50 mLa
Outcomes for Patients With Baseline Ischemic Cores Greater Than 50 mLa
Table 3.  Baseline and Procedural Characteristics of Patients With Baseline Ischemic Cores Greater Than 70 mLa
Baseline and Procedural Characteristics of Patients With Baseline Ischemic Cores Greater Than 70 mLa
Table 4.  Outcomes for Patients With Baseline Ischemic Cores Greater Than 70 mLa
Outcomes for Patients With Baseline Ischemic Cores Greater Than 70 mLa
1.
Campbell  BC, Christensen  S, Levi  CR,  et al.  Comparison of computed tomography perfusion and magnetic resonance imaging perfusion-diffusion mismatch in ischemic stroke.  Stroke. 2012;43(10):2648-2653.PubMedGoogle ScholarCrossref
2.
Campbell  BC, Yassi  N, Ma  H,  et al.  Imaging selection in ischemic stroke: feasibility of automated CT-perfusion analysis.  Int J Stroke. 2015;10(1):51-54.PubMedGoogle ScholarCrossref
3.
Albers  GW, Thijs  VN, Wechsler  L,  et al; DEFUSE Investigators.  Magnetic resonance imaging profiles predict clinical response to early reperfusion: the Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution (DEFUSE) study.  Ann Neurol. 2006;60(5):508-517.PubMedGoogle ScholarCrossref
4.
Lansberg  MG, Straka  M, Kemp  S,  et al; DEFUSE 2 Study Investigators.  MRI profile and response to endovascular reperfusion after stroke (DEFUSE 2): a prospective cohort study.  Lancet Neurol. 2012;11(10):860-867.PubMedGoogle ScholarCrossref
5.
Lansberg  MG, Lee  J, Christensen  S,  et al.  RAPID automated patient selection for reperfusion therapy: a pooled analysis of the Echoplanar Imaging Thrombolytic Evaluation Trial (EPITHET) and the Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution (DEFUSE) study.  Stroke. 2011;42(6):1608-1614.PubMedGoogle ScholarCrossref
6.
Olivot  JM, Mlynash  M, Thijs  VN,  et al.  Relationships between infarct growth, clinical outcome, and early recanalization in Diffusion and Perfusion Imaging for Understanding Stroke Evolution (DEFUSE).  Stroke. 2008;39(8):2257-2263.PubMedGoogle ScholarCrossref
7.
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(8):2205-2211.PubMedGoogle ScholarCrossref
8.
Parsons  MW, Christensen  S, McElduff  P,  et al; Echoplanar Imaging Thrombolytic Evaluation Trial (EPITHET) Investigators.  Pretreatment diffusion- and perfusion-MR lesion volumes have a crucial influence on clinical response to stroke thrombolysis.  J Cereb Blood Flow Metab. 2010;30(6):1214-1225.PubMedGoogle ScholarCrossref
9.
Mlynash  M, Lansberg  MG, De Silva  DA,  et al; DEFUSE-EPITHET Investigators.  Refining the definition of the malignant profile: insights from the DEFUSE-EPITHET pooled data set.  Stroke. 2011;42(5):1270-1275.PubMedGoogle ScholarCrossref
10.
Sanák  D, Nosál’  V, Horák  D,  et al.  Impact of diffusion-weighted MRI-measured initial cerebral infarction volume on clinical outcome in acute stroke patients with middle cerebral artery occlusion treated by thrombolysis.  Neuroradiology. 2006;48(9):632-639.PubMedGoogle ScholarCrossref
11.
Yoo  AJ, Verduzco  LA, Schaefer  PW, Hirsch  JA, Rabinov  JD, González  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(6):2046-2054.PubMedGoogle ScholarCrossref
12.
Kidwell  CS, Jahan  R, Gornbein  J,  et al; MR RESCUE Investigators.  A trial of imaging selection and endovascular treatment for ischemic stroke.  N Engl J Med. 2013;368(10):914-923.PubMedGoogle ScholarCrossref
13.
Campbell  BC, Mitchell  PJ, Kleinig  TJ,  et al; EXTEND-IA Investigators.  Endovascular therapy for ischemic stroke with perfusion-imaging selection.  N Engl J Med. 2015;372(11):1009-1018.PubMedGoogle ScholarCrossref
14.
Saver  JL, Goyal  M, Bonafe  A,  et al; SWIFT PRIME Investigators.  Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke.  N Engl J Med. 2015;372(24):2285-2295.PubMedGoogle ScholarCrossref
15.
Jovin  TG, Chamorro  A, Cobo  E,  et al; REVASCAT Trial Investigators.  Thrombectomy within 8 hours after symptom onset in ischemic stroke.  N Engl J Med. 2015;372(24):2296-2306.PubMedGoogle ScholarCrossref
16.
Berger  C, Fiorelli  M, Steiner  T,  et al.  Hemorrhagic transformation of ischemic brain tissue: asymptomatic or symptomatic?  Stroke. 2001;32(6):1330-1335.PubMedGoogle ScholarCrossref
17.
Dehkharghani  S, Bammer  R, Straka  M,  et al.  Performance and predictive value of a user-independent platform for CT perfusion analysis: threshold-derived automated systems outperform examiner-driven approaches in outcome prediction of acute ischemic stroke.  AJNR Am J Neuroradiol. 2015;36(8):1419-1425.PubMedGoogle ScholarCrossref
18.
Mandava  P, Kalkonde  YV, Rochat  RH, Kent  TA.  A matching algorithm to address imbalances in study populations: application to the National Institute of Neurological Diseases and Stroke Recombinant Tissue Plasminogen Activator acute stroke trial.  Stroke. 2010;41(4):765-770.PubMedGoogle ScholarCrossref
19.
Goyal  M, Menon  BK, van Zwam  WH,  et al; HERMES Collaborators.  Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials.  Lancet. 2016;387(10029):1723-1731.PubMedGoogle ScholarCrossref
20.
Schaefer  PW, Souza  L, Kamalian  S,  et al.  Limited reliability of CT perfusion acute infarct volume measurements compared to DWI in anterior circulation stroke.  Stroke. 2015;46(2):419-424.PubMedGoogle ScholarCrossref
21.
Albers  GW, Goyal  M, Jahan  R,  et al.  Ischemic core and hypoperfusion volumes predict infarct size in SWIFT PRIME.  Ann Neurol. 2016;79(1):76-89.PubMedGoogle ScholarCrossref
22.
Tisserand  M, Turc  G, Charron  S,  et al.  Does diffusion lesion volume above 70 mL preclude favorable outcome despite post-thrombolysis recanalization?  Stroke. 2016;47(4):1005-1011.PubMedGoogle ScholarCrossref
23.
Kruetzelmann  A, Köhrmann  M, Sobesky  J,  et al.  Pretreatment diffusion-weighted imaging lesion volume predicts favorable outcome after intravenous thrombolysis with tissue-type plasminogen activator in acute ischemic stroke.  Stroke. 2011;42(5):1251-1254.PubMedGoogle ScholarCrossref
24.
Kimura  K, Iguchi  Y, Shibazaki  K,  et al.  Large ischemic lesions on diffusion-weighted imaging done before intravenous tissue plasminogen activator thrombolysis predicts a poor outcome in patients with acute stroke.  Stroke. 2008;39(8):2388-2391.PubMedGoogle ScholarCrossref
25.
Ribo  M, Tomasello  A, Lemus  M,  et al.  Maximal admission core lesion compatible with favorable outcome in acute stroke patients undergoing endovascular procedures.  Stroke. 2015;46(10):2849-2852.PubMedGoogle ScholarCrossref
26.
Gilgen  MD, Klimek  D, Liesirova  KT,  et al.  Younger stroke patients with large pretreatment diffusion-weighted imaging lesions may benefit from endovascular treatment.  Stroke. 2015;46(9):2510-2516.PubMedGoogle ScholarCrossref
27.
Noorian  AR, Rangaraju  S, Sun  CH,  et al.  Endovascular therapy in strokes with ASPECTS 5-7 may result in smaller infarcts and better outcomes as compared to medical treatment alone.  Interv Neurol. 2015;4(1-2):30-37.PubMedGoogle ScholarCrossref
Original Investigation
January 2017

Endovascular Treatment for Patients With Acute Stroke Who Have a Large Ischemic Core and Large Mismatch Imaging Profile

Author Affiliations
  • 1Department of Neurology, Grady Memorial Hospital, Emory University School of Medicine, Atlanta, Georgia
  • 2Department of Radiology, Grady Memorial Hospital, Emory University School of Medicine, Atlanta, Georgia
  • 3Department of Neurosurgery, Grady Memorial Hospital, Emory University School of Medicine, Atlanta, Georgia
JAMA Neurol. 2017;74(1):34-40. doi:10.1001/jamaneurol.2016.3954
Key Points

Question  Could patients with large baseline ischemic cores on computed tomographic perfusion and a mismatch imaging profile benefit from endovascular therapy for acute ischemic stroke?

Findings  In this matched case-control study, endovascular therapy was significantly associated with a favorable shift in the overall distribution of 90-day modified Rankin Scale scores, higher rates of independent outcomes (90-day modified Rankin Scale scores of 0-2, 25% vs 0%), and smaller final infarct volumes (87 vs 242 mL). Endovascular therapy was also associated with numerically lower rates of parenchymal hematoma type 2, hemicraniectomy, and 90-day mortality.

Meaning  In properly selected patients, endovascular therapy appears to benefit patients with large core and large mismatch profiles.

Abstract

Importance  Endovascular therapy (ET) is typically not considered for patients with large baseline ischemic cores (irreversibly injured tissue). Computed tomographic perfusion (CTP) imaging may identify a subset of patients with large ischemic cores who remain at risk for significant infarct expansion and thus could still benefit from reperfusion to reduce their degree of disability.

Objective  To compare the outcomes of patients with large baseline ischemic cores on CTP undergoing ET with the outcomes of matched controls who had medical care alone.

Design, Setting, and Participants  A matched case-control study of patients with proximal occlusion after stroke (intracranial internal carotid artery and/or middle cerebral artery M1 and/or M2) on computed tomographic angiography and baseline ischemic core greater than 50 mL on CTP at a tertiary care center from May 1, 2011, through October 31, 2015. Patients receiving ET and controls receiving medical treatment alone were matched for age, baseline ischemic core volume on CTP, and glucose levels. Baseline characteristics and outcomes were compared.

Main Outcomes and Measures  The primary outcome measure was the shift in the degree of disability among the treatment and control groups as measured by the modified Rankin Scale (mRS) (with scores ranging from 0 [fully independent] to 6 [dead]) at 90 days.

Results  Fifty-six patients were matched across 2 equally distributed groups (mean [SD] age, 62.25 [13.92] years for cases and 58.32 [14.79] years for controls; male, 13 cases [46%] and 14 controls [50%]). Endovascular therapy was significantly associated with a favorable shift in the overall distribution of 90-day mRS scores (odds ratio, 2.56; 95% CI, 2.50-8.47; P = .04), higher rates of independent outcomes (90-day mRS scores of 0-2, 25% vs 0%; P = .04), and smaller final infarct volumes (mean [SD], 87 [77] vs 242 [120] mL; P < .001). One control (4%) and 2 treatment patients (7%) developed a parenchymal hematoma type 2 (P > .99). The rates of hemicraniectomy (2 [7%] vs 6 [21%]; P = .10) and 90-day mortality (7 [29%] vs 11 [48%]; P = .75) were numerically lower in the intervention arm. Sensitivity analysis for patients with a baseline ischemic core greater than 70 mL (12 pairs) revealed a significant reduction in final infarct volumes (mean [SD], 110 [65] vs 319 [147] mL; P < .001) but only a nonsignificant improvement in the overall distribution of mRS scores favoring the treatment group (P = .18). All 11 patients older than 75 years had poor outcomes (mRS score >3) at 90 days.

Conclusions and Relevance  In properly selected patients, ET appears to benefit patients with large core and large mismatch profiles. Future prospective studies are warranted.

Introduction

The aim of acute ischemic stroke treatment is to reestablish flow to hypoperfused but still viable brain tissue before an irreversible injury develops using intravenous tissue plasminogen activator (IV-tPA), endovascular therapy (ET), or both. Proper patient selection is critical, and advanced imaging techniques have been studied to identify those who can benefit from treatment. Nonviable tissue or infarct core can be recognized using diffusion-weighted magnetic resonance imaging (DW-MRI) or computed tomographic perfusion (CTP) techniques.1,2

The target mismatch profile was studied by the Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution (DEFUSE) group as a concept to identify ideal candidates for reperfusion treatment.3 As per DEFUSE-2, target mismatch on MRI was defined as (1) a ratio of 1.8 or greater between the volumes of critically hypoperfused tissue (time to maximum tissue residue function [Tmax], >6 seconds) and ischemic core on DW-MRI maps, (2) an absolute difference of 15 mL or more, (3) an ischemic core on DW-MRI maps of 70 mL or less, and (4) volume of tissue with severe hypoperfusion (Tmax, >10 seconds) of 100 mL or less.4 Favorable clinical outcomes after early recanalization in patients with target mismatch profiles have been previously established.3-6 However, little is known regarding aggressive reperfusion strategies in the setting of large baseline ischemic cores because these patients are in general thought to have poor outcomes despite treatment with IV-tPA or ET.7-12 Consequently, those patients are usually not offered reperfusion treatment in clinical practice and clinical trials. In fact, a large baseline ischemic core was an exclusion criterion in most of the more recently published randomized clinical trials of ET for acute ischemic stroke. The Extending the Time for Thrombolysis in Emergency Neurological Deficits–Intra-arterial (EXTEND-IA) and Solitaire With the Intention For Thrombectomy as Primary Endovascular Treatment (SWIFT PRIME) trials excluded patients with a baseline ischemic core greater than 70 mL and greater than 50 mL, respectively13,14; the Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE) trial excluded patients with Alberta Stroke Program Early CT (ASPECT) scores less than 6,14 and the Endovascular Revascularization With Solitaire Device vs Best Medical Therapy in Anterior Circulation Stroke Within 8 Hours trial excluded patients with ASPECT scores less than 7 on CT or less than 6 on DW-MRI.15

We hypothesized that CTP imaging may identify a subset of patients with high ischemic core volumes on presentation who remain at high risk for significant infarct expansion and thus could still benefit from endovascular reperfusion as a strategy to reduce their degree of disability. We therefore designed the present study to compare the outcomes of patients with large baseline ischemic cores on CTP undergoing ET with matched controls who had medical care alone.

Methods
Patient Selection and Measures of Outcomes

We reviewed our prospectively collected large vessel occlusion stroke database to identify patients presenting with large ischemic cores at a tertiary care center from May 1, 2011, through October 31, 2015. Patients undergoing ET were compared with a matched control group who did not undergo ET. All patients were evaluated for ET. Given the lack of any specific data, there was no prespecified protocol for the treatment of large core strokes at Grady Memorial Hospital, Emory University School of Medicine, Atlanta, Georgia, during the study period. Therefore, the decision to treat vs not treat any given patient with a large infarct on presentation was left to the treating neuroendovascular specialist with common agreement with the stroke team caring for the patients and the patients’ families. The specific selection criteria for the present study were as follows. Inclusion criteria were intracranial internal carotid artery and/or proximal middle cerebral artery (M1 and/or M2) occlusion on CT angiography; time from last-known normal to treatment of less than 12 hours; imaging screening using CTP with the automated RAPID software, version 4.5.0 (iSchemaView); baseline ischemic cores on CTP (regional cerebral blood flow [rCBF], <30%) greater than 50 mL; and absolute mismatch volume (Tmax, >6 seconds per lesion; rCBF, <30%) greater than 40 mL with involvement of eloquent areas (sensorimotor, language, spatial attention, or visual functions). Exclusion criteria were baseline ischemic cores on CTP greater than 150 mL, acute bihemispheric and/or posterior circulation strokes, and inconclusive or unreliable CTP imaging. The study was approved by the Emory University Institutional Review Board. Consent was waived because this was a retrospective medical record review and no additional intervention was performed.

Baseline characteristics and demographics, as well as imaging and procedural data, were collected. The primary outcome measure was the shift in the degree of disability among the treatment and control groups as measured by the modified Rankin Scale (mRS) (with scores ranging from 0 [fully independent] to 6 [dead]) at 90 days. Secondary end points included the rates of independent and ambulatory outcomes (defined as 90-day mRS scores of 0-2 and 0-3, respectively), rates of hemicraniectomy, and final infarct volumes (FIVs). Safety end points included the rates of parenchymal hematoma type 2 (as defined in the European Cooperative Acute Stroke Study)16 and 90-day mortality. Subgroup analyses were performed for exploratory purposes, including baseline ischemic core greater than 70 mL and age stratification at 65 years or younger, older than 65 years, and older than 75 years.

Image Protocol

The CTP encompassing 8 cm of brain coverage was evaluated with a fully automated software environment (RAPID, version 4.5.0). Ischemic core (irreversibly injured tissue) was defined by a rCBF reduction of less than 30% of the corresponding contralateral normal tissues. The total hypoperfused volume (ischemic penumbra) was defined by a greater than 6-second delay in the Tmax.14

Follow-up imaging included noncontrast CT or MRI documenting the FIVs within 5 days of the treatment. The DW-MRI sequence was preferentially used if MRI was obtained within the first 72 hours of the stroke and the fluid-attenuated inversion recovery sequence if the MRI was performed within 3 to 5 days.4 For noncontrast CT, window-level settings were adjusted to maximize contrast between the normal and infarcted brain. Edema that produced sulcal effacement was not excluded. Hemorrhagic transformation was incorporated in the FIVs whenever present. The FIVs were measured after export of raw Digital Imaging and Communications in Medicine data to the Fiji release of the ImageJ software platform (https://imagej.nih.gov/ij/).17

Matching Method

We adopted a matching method based on weighted Euclidean distances to obtain a pair of individuals considered to be the nearest neighbors in a 3-dimensional space of baseline ischemic core on CTP, age, and pretreatment glucose levels.18 The distance between each intervention-control pair was computed using the %FIND_NEIGHBORS macro in the University Edition of SAS statistical software (SAS Institute Inc). The distribution of Euclidian distances was then studied, and a threshold was determined as follows: Q75 + 1.5 × (Q75 − Q25), where Q25 and Q75 are the 25th and 75th percentiles, respectively.18 Pairs with distances greater than the threshold were considered outliers and eliminated from further consideration.

Statistical Analysis

Continuous variables were reported as mean (SD) or median (interquartile range [IQR]) as appropriate. Categorical variables were reported as proportions. Between groups, comparisons for continuous and ordinal variables were made with the paired t test or Wilcoxon rank sum test as appropriate. Categorical variables were compared by the McNemar test for discordant pairs. The overall distribution of 90-day mRS scores was compared between groups (shift in disability levels) using the Wilcoxon signed rank test to account for the matching.14 Ordinal regression was computed for odds ratios (ORs) to assess the association between ET and mRS score. Significance was set at P < .05, and all P values were 2-sided. Statistical analysis was performed using Statistics 23 (IBM Corp) except for the McNemar test, which was computed using the FREQ procedure in the University Edition of SAS statistical software (SAS Institute Inc).

Results

A total of 28 qualifying patients in the intervention group and 41 in the control group were identified during the study period. After matching the 2 groups for age, baseline ischemic core on CTP, and pretreatment glucose levels, 28 pairs were generated and included in the primary analysis (mean [SD] age, 62.25 [13.92] years for cases and 58.32 [14.79] years for controls; male, 13 cases [46%] and 14 controls [50%]). The same neuroendovascular physician (R.G.N.) treated 21 of the 28 patients undergoing ET (75%), with the remaining patients equally distributed across 3 other physicians. There was no clear pattern to the neuroendovascular call system, with occasional periods of alternating calls and other periods of in-block calls resulting in a reasonable degree of randomness to the treatment paradigm.

Primary Analysis: Ischemic Core Greater Than 50 mL on CTP

Demographic and baseline characteristics were comparable between the 2 groups (Table 1), including the Tmax greater than 6 seconds lesion volumes (205.4 [87.7] vs 210.8 [62.6] mL; P = .77). The median time from last-known normal to puncture in the intervention group was 262 minutes (IQR, 196.2-515.5 minutes). A stent retriever was used in 22 cases (79%). Successful reperfusion (modified Thrombolysis in Cerebral Infarction [mTICI] score of 2b-3) was achieved in 22 patients (79%) undergoing ET. The full hospitalization data set was available in all 56 patients, but detailed 90-day mRS scores were missing in 4 treatment and 5 control patients.

Thrombectomy treatment was associated with a favorable shift in the overall distribution of 90-day mRS scores (OR, 2.56; 95% CI, 2.50-8.47; P = .04) (Table 2 and Figure). Controlling for age, glucose level, and baseline ischemic core on CTP (adjusted OR, 6.93; 95% CI, 1.88-25.53; P = .004) or ASPECT scores (adjusted OR, 8.47; 95% CI, 2.09-34.27; P = .003) yielded the same results. Six of the 24 patients (25%) in the intervention group had good outcomes (mRS scores of 0-2) at 90 days and 0 of 23 patients in the control group had good outcomes (P = .04). The FIVs were significantly smaller in the treatment group (87 [77] vs 242 [120] mL; P < .001). Parenchymal hematoma type 2 occurred in only 2 patients (7%) in the treatment group and 1 (4%) in the control group (P > .99). There were numerically lower rates of hemicraniectomy (2 [7%] vs 6 [21%]; P = .10) and 90-day mortality (7 [29%] vs 11 [48%], P = .75) in the intervention arm.

Secondary Analysis
Baseline Ischemic Core Greater Than 70 mL on CTP

The same matching algorithm was used to create another matched study population with a baseline ischemic core greater than 70 mL on CTP. Twelve pairs were generated and underwent analysis. No differences were found in baseline characteristics (Table 3). The Tmax greater than 6 seconds lesion volumes were again similar in the 2 groups (259.4 [105.0 vs 231.2 [53.6 mL]; P = .32). The median time from last-known normal to puncture in the intervention group was 258 minutes (IQR, 185-498 minutes). A stent retriever was used in 9 cases (75%). The rate of successful reperfusion (mTICI score of 2b-3) was 91.7%. The full hospitalization data set was available in all 24 patients, but detailed 90-day mRS scores were missing in 2 of the treatment and 1 of the control patients.

There was a significant reduction in FIVs (110 [65] vs 319 [147 mL]; P < .001) but only a nonsignificant improvement in the overall distribution of mRS scores that favored the treatment group (P = .18) (Table 4). Four of 10 patients (40%) in the treatment group achieved functional independence at 90 days compared with 0 of 11 patients in the control group (P = .24). None of the 12 intervention patients and 1 of the 12 control patients (8%) developed a parenchymal hematoma type 2 (P > .99). The rates of hemicraniectomy (2 [17%] vs 4 [33%]; P = .31) and 90-day mortality (4 [40%] vs 6 [54.5%]; P = .65) were numerically lower in the treatment arm, but these differences did not reach statistical significance.

Age Stratification

Endovascular treatment resulted in a significant favorable shift in the degree of 90-day disability in patients 75 years or younger (21 pairs; P = .03 for ordinal shift analysis; mRS scores of 0-2, 6 of 17 [35%] vs 0; mRS scores of 5-6, 3 of 17 [18%] vs 8 of 18 [44%]). All patients 75 years or older (6 in the intervention group and 5 in the control group) had poor outcomes (mRS score >3) at 90 days (eTable in the Supplement).

Discussion

Our results suggest that ET may improve functional outcomes when added to medical management in patients with anterior circulation, large vessel occlusion stroke presenting with large ischemic cores (rCBF <30% and ischemic core >50 mL), and large mismatch profiles. Despite the increased risk of poor outcome and hemorrhagic transformation in the large core population, we were able to demonstrate a favorable shift in the overall distribution of 90-day mRS scores with ET. No difference in rates of severe intracranial hemorrhage was found, whereas the hemicraniectomy and 90-day mortality rates were numerically lower in the treatment group. Even though our sample size for the subgroup with rCBF less than 30% and ischemic cores greater than 70 mL was presumably too small to demonstrate a clinically meaningful effect, we were able to reveal a marked reduction in FIVs, which should theoretically lead to better functional outcomes. To the best of our knowledge, this is the first study to assess the effects of ET in patients with a large baseline ischemic core and large mismatch profile using a matched case-control method.

Our study has potentially important implications to the current practice standards considering that the recently published pooled analysis of the 5 major thrombectomy trials failed to demonstrate a treatment benefit in patients with ASPECT scores of 5 or less (121 patients; OR, 1.24; 95% CI, 0.62-2.49).19 At first glance, our results seem contradictory to this study, but it is critical to recognize that many of the patients included in the pooled analysis had unknown mismatch status. As such, it is likely that the lack of significant residual, salvageable tissue along with large completed cores in a significant proportion of their patients attenuated the benefit of reperfusion. In contrast, all patients in our study had significant volumes of viable but critically hypoperfused tissue (mean Tmax >6 seconds and ischemic core >200 mL) despite large baseline ischemic cores.

It has been established that CTP has more limited accuracy to detect ischemic cores compared with DW-MRI. In this context, the rCBF threshold of less than 30% is optimized to avoid overcalls, often leading to an underestimation of true ischemic core volumes.20 As an example, the median absolute difference between the observed and predicted core volume was 13 mL in the SWIFT PRIME trial.21 Notably, SWIFT PRIME had a median baseline ischemic core of only 6 mL (IQR, 0-16 mL), and higher differences between the observed and predicted core volumes likely occur in patients with large baseline ischemic cores. In fact, the median difference was 38 mL in patients enrolled in SWIFT PRIME who had malignant profiles. Therefore, it is probable that a baseline rCBF less than 30% and a baseline ischemic core greater than 50 mL often approximate to a baseline ischemic core greater than 70 mL on DW-MRI. Another important consideration is that recent data support that large lesions seen on DW-MRI might be reversible with early reperfusion.22

Several studies7-11,23,24 have found an association between large baseline ischemic core in acute ischemic stroke and high risk of poor outcome after revascularization treatment. The DEFUSE investigators described poor outcomes and high rates of hemorrhagic transformation after IV-tPA in patients with a malignant imaging profile (defined as a baseline ischemic core >100 mL on DW-MRI and/or a Tmax >8 seconds and a baseline ischemic core >100 mL).3 The data on endovascular reperfusion of patients with large ischemic cores remain sparse but for the most part have been equally discouraging. Yoo et al11 reported poor clinical outcomes in all patients with baseline ischemic cores greater than 70 mL on DW-MRI regardless of their final reperfusion status. In the Mechanical Retrieval and Recanalization of Stroke Clots Using Embolectomy (MR RESCUE) trial, 30 patients with large cores (122.8 mL; IQR, 96.9-171.4 mL) and a nonpenumbral pattern on baseline CTP or MRI underwent ET.12 Only 5 (17%) of those patients achieved a good outcome, whereas 6 patients (20%) were dead at 90 days. Ribo et al25 reported that only 12% of patients who underwent successful reperfusion with baseline ischemic cores greater than 39 mL achieved good outcomes at 90 days.

In contrast to the aforementioned results and consistent with our findings, other series have found good outcomes in more than one-fifth of patients with large baseline ischemic cores. Tisserand et al22 recently reported on 267 patients who were evaluated with acute MRI and treated with IV-tPA within 4.5 hours from stroke onset. Good outcomes were seen in 12 of the 54 patients (22%) with baseline ischemic cores of 70 mL or greater on DW-MRI. Olivot et al7 reported good outcomes in 4 of 19 patients (21%) undergoing ET with baseline ischemic cores greater than 70 mL on DW-MRI. Interestingly, complete endovascular reperfusion (Thrombosis in Myocardial Infarction [TIMI] score of 3, 7 patients) was associated with numerically higher rates of favorable outcomes (43% vs 8%; P = .12) and significantly lower mortality (28% vs 83%; P = .04) compared with failed or incomplete reperfusion (TIMI scores of 0-2, 12 patients). Hemorrhagic transformation occurred in 6 of 19 patients but did not significantly differ in those who underwent full reperfusion (TIMI score of 3, 3 of 7 [43%]) vs not (TIMI scores of 0-2, 3 of 12 [25%]; P = .62). Similarly, Gilgen et al26 described favorable outcomes after ET in 14 of 66 patients (21%) with baseline ischemic cores greater than 70 mL on DW-MRI. Among those who underwent successful reperfusion (mTICI scores of 2b-3), a favorable outcome was achieved in 35.5%. Symptomatic intracranial hemorrhage occurred in 13 of 66 patients (20%). Remarkably, 7 of 39 patients (18%) with baseline ischemic cores greater than 100 mL on DW-MRI achieved good outcomes. Our rate of favorable outcome is in the same range as these 3 series; however, we have had lower rates of severe intracranial hemorrhage compared with most of the aforementioned studies. The potential reasons for that include the higher use of the newer stent-retriever technology (in contrast with the Merci device and/or intra-arterial lytics), which is associated with safer, faster, and more complete reperfusion, as well as the existence of a site protocol for immediate, aggressive intraprocedural blood pressure control as soon as reperfusion is achieved. In their studies, Ribo et al25 and Gilgen et al26 highlighted the significant interactions among age, baseline ischemic core volumes, and outcomes. We found similar associations, with all patients with large ischemic cores who were 75 years or older having poor outcomes regardless of treatment.

Our study raises important philosophical and ethical considerations. The stroke field has long lived under the fallacy of improper outcome definitions and their related end points. This is even more evident when it comes to ET. It is not reasonable to expect an independent outcome in patients who present with a defined large infarct on baseline imaging. In this setting, our goal should be a better outcome compared with what we would have had without ET. There is no logical sense in withholding endovascular reperfusion in a patient with a baseline ischemic core of 80 mL in the setting of a significantly larger mismatch only to perform a hemicraniectomy moments later when one could have potentially performed reperfusion, saved brain tissue, and avoided a more invasive and expensive surgical procedure.27

Limitations

There are significant limitations to our study, mostly inherent to the retrospective design and relatively small sample size. Using DW-MRI as the screening tool would have led to a more reliable measure of the baseline ischemic cores. However, the use of CTP makes our findings more generalizable because immediate MRI evaluation is not available in most stroke centers. Moreover, the RAPID CTP core method is known to underestimate as opposed to overestimate volumes, which should theoretically prevent an artificial inflation of our results. Long-term outcomes were missing in 9 of the 56 patients. However, the inclusion of all patients helped better characterize the effects of treatment on hemorrhagic complications and infarct volume reduction.

Conclusions

Endovascular therapy appears to benefit patients with large ischemic cores and large mismatch profiles. This benefit likely becomes less pronounced with increasing age. Future prospective trials are warranted to confirm these findings.

Back to top
Article Information

Corresponding Author: Raul G. Nogueira, MD, Department of Neurology, Grady Memorial Hospital, Emory University School of Medicine, 49 Jesse Hill Jr Dr SE, Room 333, Atlanta, GA 30303 (raul.g.nogueira@emory.edu).

Accepted for Publication: August 18, 2016.

Published Online: November 7, 2016. doi:10.1001/jamaneurol.2016.3954

Author Contributions: Drs Rebello and Bouslama contributed equally to this work. Drs Bouslama and Nogueira had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Bouslama, Frankel, Nogueira.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Rebello, Bouslama, Nogueira.

Critical revision of the manuscript for important intellectual content: Bouslama, Haussen, Dehkharghani, Grossberg, Belagaje, Frankel.

Statistical analysis: Rebello, Bouslama, Grossberg, Nogueira.

Administrative, technical, or material support: Rebello, Dehkharghani, Frankel, Nogueira.

Study supervision: Rebello, Grossberg, Frankel, Nogueira.

Conflict of Interest Disclosures: Dr Nogueira reported receiving support from Stryker Neurovascular (Trevo Eersus Merci Retrievers for Thrombectomy Revascularisation of Large Vessel Occlusions in Acute Ischaemic Stroke and Trevo and Medical Management Versus Medical Management Alone in Wake Up and Late Presenting Strokes, principal investigator), Covidien (SWIFT and SWIFT PRIME Steering Committee, Solitaire FR Thrombectomy for Acute Revascularization), and Penumbra (3-D Separator Trial Executive Committee). No other disclosures were reported.

References
1.
Campbell  BC, Christensen  S, Levi  CR,  et al.  Comparison of computed tomography perfusion and magnetic resonance imaging perfusion-diffusion mismatch in ischemic stroke.  Stroke. 2012;43(10):2648-2653.PubMedGoogle ScholarCrossref
2.
Campbell  BC, Yassi  N, Ma  H,  et al.  Imaging selection in ischemic stroke: feasibility of automated CT-perfusion analysis.  Int J Stroke. 2015;10(1):51-54.PubMedGoogle ScholarCrossref
3.
Albers  GW, Thijs  VN, Wechsler  L,  et al; DEFUSE Investigators.  Magnetic resonance imaging profiles predict clinical response to early reperfusion: the Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution (DEFUSE) study.  Ann Neurol. 2006;60(5):508-517.PubMedGoogle ScholarCrossref
4.
Lansberg  MG, Straka  M, Kemp  S,  et al; DEFUSE 2 Study Investigators.  MRI profile and response to endovascular reperfusion after stroke (DEFUSE 2): a prospective cohort study.  Lancet Neurol. 2012;11(10):860-867.PubMedGoogle ScholarCrossref
5.
Lansberg  MG, Lee  J, Christensen  S,  et al.  RAPID automated patient selection for reperfusion therapy: a pooled analysis of the Echoplanar Imaging Thrombolytic Evaluation Trial (EPITHET) and the Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution (DEFUSE) study.  Stroke. 2011;42(6):1608-1614.PubMedGoogle ScholarCrossref
6.
Olivot  JM, Mlynash  M, Thijs  VN,  et al.  Relationships between infarct growth, clinical outcome, and early recanalization in Diffusion and Perfusion Imaging for Understanding Stroke Evolution (DEFUSE).  Stroke. 2008;39(8):2257-2263.PubMedGoogle ScholarCrossref
7.
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(8):2205-2211.PubMedGoogle ScholarCrossref
8.
Parsons  MW, Christensen  S, McElduff  P,  et al; Echoplanar Imaging Thrombolytic Evaluation Trial (EPITHET) Investigators.  Pretreatment diffusion- and perfusion-MR lesion volumes have a crucial influence on clinical response to stroke thrombolysis.  J Cereb Blood Flow Metab. 2010;30(6):1214-1225.PubMedGoogle ScholarCrossref
9.
Mlynash  M, Lansberg  MG, De Silva  DA,  et al; DEFUSE-EPITHET Investigators.  Refining the definition of the malignant profile: insights from the DEFUSE-EPITHET pooled data set.  Stroke. 2011;42(5):1270-1275.PubMedGoogle ScholarCrossref
10.
Sanák  D, Nosál’  V, Horák  D,  et al.  Impact of diffusion-weighted MRI-measured initial cerebral infarction volume on clinical outcome in acute stroke patients with middle cerebral artery occlusion treated by thrombolysis.  Neuroradiology. 2006;48(9):632-639.PubMedGoogle ScholarCrossref
11.
Yoo  AJ, Verduzco  LA, Schaefer  PW, Hirsch  JA, Rabinov  JD, González  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(6):2046-2054.PubMedGoogle ScholarCrossref
12.
Kidwell  CS, Jahan  R, Gornbein  J,  et al; MR RESCUE Investigators.  A trial of imaging selection and endovascular treatment for ischemic stroke.  N Engl J Med. 2013;368(10):914-923.PubMedGoogle ScholarCrossref
13.
Campbell  BC, Mitchell  PJ, Kleinig  TJ,  et al; EXTEND-IA Investigators.  Endovascular therapy for ischemic stroke with perfusion-imaging selection.  N Engl J Med. 2015;372(11):1009-1018.PubMedGoogle ScholarCrossref
14.
Saver  JL, Goyal  M, Bonafe  A,  et al; SWIFT PRIME Investigators.  Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke.  N Engl J Med. 2015;372(24):2285-2295.PubMedGoogle ScholarCrossref
15.
Jovin  TG, Chamorro  A, Cobo  E,  et al; REVASCAT Trial Investigators.  Thrombectomy within 8 hours after symptom onset in ischemic stroke.  N Engl J Med. 2015;372(24):2296-2306.PubMedGoogle ScholarCrossref
16.
Berger  C, Fiorelli  M, Steiner  T,  et al.  Hemorrhagic transformation of ischemic brain tissue: asymptomatic or symptomatic?  Stroke. 2001;32(6):1330-1335.PubMedGoogle ScholarCrossref
17.
Dehkharghani  S, Bammer  R, Straka  M,  et al.  Performance and predictive value of a user-independent platform for CT perfusion analysis: threshold-derived automated systems outperform examiner-driven approaches in outcome prediction of acute ischemic stroke.  AJNR Am J Neuroradiol. 2015;36(8):1419-1425.PubMedGoogle ScholarCrossref
18.
Mandava  P, Kalkonde  YV, Rochat  RH, Kent  TA.  A matching algorithm to address imbalances in study populations: application to the National Institute of Neurological Diseases and Stroke Recombinant Tissue Plasminogen Activator acute stroke trial.  Stroke. 2010;41(4):765-770.PubMedGoogle ScholarCrossref
19.
Goyal  M, Menon  BK, van Zwam  WH,  et al; HERMES Collaborators.  Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials.  Lancet. 2016;387(10029):1723-1731.PubMedGoogle ScholarCrossref
20.
Schaefer  PW, Souza  L, Kamalian  S,  et al.  Limited reliability of CT perfusion acute infarct volume measurements compared to DWI in anterior circulation stroke.  Stroke. 2015;46(2):419-424.PubMedGoogle ScholarCrossref
21.
Albers  GW, Goyal  M, Jahan  R,  et al.  Ischemic core and hypoperfusion volumes predict infarct size in SWIFT PRIME.  Ann Neurol. 2016;79(1):76-89.PubMedGoogle ScholarCrossref
22.
Tisserand  M, Turc  G, Charron  S,  et al.  Does diffusion lesion volume above 70 mL preclude favorable outcome despite post-thrombolysis recanalization?  Stroke. 2016;47(4):1005-1011.PubMedGoogle ScholarCrossref
23.
Kruetzelmann  A, Köhrmann  M, Sobesky  J,  et al.  Pretreatment diffusion-weighted imaging lesion volume predicts favorable outcome after intravenous thrombolysis with tissue-type plasminogen activator in acute ischemic stroke.  Stroke. 2011;42(5):1251-1254.PubMedGoogle ScholarCrossref
24.
Kimura  K, Iguchi  Y, Shibazaki  K,  et al.  Large ischemic lesions on diffusion-weighted imaging done before intravenous tissue plasminogen activator thrombolysis predicts a poor outcome in patients with acute stroke.  Stroke. 2008;39(8):2388-2391.PubMedGoogle ScholarCrossref
25.
Ribo  M, Tomasello  A, Lemus  M,  et al.  Maximal admission core lesion compatible with favorable outcome in acute stroke patients undergoing endovascular procedures.  Stroke. 2015;46(10):2849-2852.PubMedGoogle ScholarCrossref
26.
Gilgen  MD, Klimek  D, Liesirova  KT,  et al.  Younger stroke patients with large pretreatment diffusion-weighted imaging lesions may benefit from endovascular treatment.  Stroke. 2015;46(9):2510-2516.PubMedGoogle ScholarCrossref
27.
Noorian  AR, Rangaraju  S, Sun  CH,  et al.  Endovascular therapy in strokes with ASPECTS 5-7 may result in smaller infarcts and better outcomes as compared to medical treatment alone.  Interv Neurol. 2015;4(1-2):30-37.PubMedGoogle ScholarCrossref
×