Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

1

 From the Institute for Health Policy, Management and Evaluation (K.R.B., M.K.K., A.L.), Department of Medical Imaging (K.R.B., R.I.A., A.R.M.), Institute of Medical Sciences (R.I.A., A.R.M.), and Department of Medicine (M.K.K., A.L.), University of Toronto, 263 McCaul St, 4th Floor, Toronto, ON, Canada M5T 1W7; Department of Medical Imaging, Fraser Health Authority, Vancouver, British Columbia, Canada (D.D.); Department of Medical Imaging, University of British Columbia, Vancouver, British Columbia, Canada (D.D.); Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada (M.K.K.); and Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Toronto, Ontario, Canada (A.L.). Received April 11, 2014; revision requested May 21; revision received June 11; accepted June 26; final version accepted July 18. Address correspondence to K.R.B. (e-mail: [email protected]).

Purpose:

To determine rates of death, disability, and symptomatic intracranial hemorrhage (SICH) among patients with acute ischemic stroke selected for thrombolytic therapy by using perfusion computed tomography (CT) by conducting a systematic review and meta-analysis.

Materials and Methods:

A search of the literature up to July 2012 was performed by using MEDLINE, EMBASE, the Cochrane Library, PubMed, and Google Scholar on terms including “brain ischemia” and “perfusion imaging.” The search was unrestricted by language of publication. Two reviewers extracted study data and independently assessed the risk of study bias. Outcomes of patients selected by using perfusion CT, including case-fatality rate, favorable outcome (modified Rankin Scale [mRS] score, 2), and rates of SICH, were estimated.

Results:

Thirteen experimental or observational studies that included patients who received intravenous thrombolytic treatment after perfusion CT were identified. The methodologic quality of the small studies was generally good. Overall, 90-day mortality was 10.0% (95% confidence interval [CI]: 5.4%, 15.9%). Among patients treated within 3 hours of symptom onset, mortality was 12.5% (95% CI: 6.7%, 19.7%), a favorable outcome (mRS score, 2) was seen in 42.5% of patients (95% CI: 16.6%, 70.9%), and the SICH rate was 3.3% (95% CI: 0.7%, 7.7%). Among patients treated more than 3 hours after symptom onset, mortality was 2.9% (95% CI: 0.0%, 12.7%), 69.9% of patients (95% CI: 0%, 83.5%) had a favorable outcome, and the SICH rate was 3.9% (95% CI: 0.8%, 9.2%).

Conclusion:

The outcomes (mortality, morbidity, and SICH rates) for patients selected with perfusion CT to receive intravenous thrombolytic treatment more than 3 hours after symptom onset appear favorable. q RSNA, 2014

Online supplemental material is available for this article.

 RSNA, 2014

q

Radiology: Volume 274: Number 1—January 2015  n  radiology.rsna.org

103

Practice

Kirsteen R. Burton, MBA, MSc, MD Deljit Dhanoa, MD Richard I. Aviv, MBChB Alan R. Moody, MBBS Moira K. Kapral, MD, MSc Andreas Laupacis, MD, MSc

Original Research  n  Evidence-Based

Perfusion CT for Selecting Patients with Acute Ischemic Stroke for Intravenous Thrombolytic Therapy1

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis

S

troke is a common disease that produces substantial long-term health and economic burdens (1– 3). In Canada, the annual cost of care within the 1st 6 months of stroke is estimated at $2.5 billion Canadian (4). Acute ischemic stroke (AIS) constitutes approximately 85% of all strokes, with optimal management requiring timely access to neuroimaging (computed tomography [CT] or magnetic resonance [MR] imaging) and thrombolytic therapy (5,6). The introduction of thrombolytic therapy for patients with AIS has increased the importance of the use of diagnostic imaging technologies to guide the selection of patients most likely to benefit from thrombolytic medications. CT performed without contrast material enhancement (noncontrast CT) has been the mainstay of imaging in patients suspected of having a stroke because of its wide availability, relatively low cost, and high sensitivity in the detection of intracranial hemorrhage (ICH) (7). However, several studies have suggested that the use of advanced imaging techniques (perfusion CT or

Advances in Knowledge nn The patients selected by using perfusion CT included in this systematic review had acceptable rates of morbidity, mortality, and symptomatic intracranial hemorrhage (SICH) after thrombolysis. nn Among perfusion CT–selected patients treated within 3 hours of symptom onset, mortality was 12.5% (95% confidence interval [CI]: 6.7%, 19.7%), a favorable outcome (modified Rankin Scale score, 2) was seen in 42.5% of patients (95% CI: 16.6%, 70.9%), and the SICH rate was 3.3% (95% CI: 0.7%, 7.7%). nn Among perfusion CT–selected patients treated more than 3 hours after symptom onset, mortality was 2.9% (95% CI: 0.0%, 12.7%), 69.9% of patients (95% CI: 0.0%, 83.5%) had a favorable outcome, and the SICH rate was 3.9% (95% CI: 0.8%, 9.2%). 104

Burton et al

MR imaging) may help identify a larger proportion of patients who will benefit from thrombolytic therapy. Unlike noncontrast CT, both perfusion CT and MR imaging can depict salvageable brain tissue and provide data about mismatched diffusion and perfusion characteristics of brain tissue. This information has been shown to be effective in influencing the selection of patients beyond 3 and 4.5 hours of symptom onset for thrombolytic therapy (8,9). However, MR imaging is more time consuming and costly than both noncontrast CT and perfusion CT (3,10). Perfusion CT imaging enables noninvasive, rapid evaluation of cerebral perfusion, ischemia, and infarction (11). Furthermore, perfusion CT may be performed with most modern CT scanners (11), adds only a few minutes to the noncontrast CT protocol, and is more accessible and rapidly available than MR imaging (12). Our goal is to review the evidence for the effectiveness of perfusion CT in the selection of patients with AIS who are eligible for thrombolytic therapy. Although systematic reviews (13– 17) have been performed to estimate the outcomes of patients with AIS selected for thrombolytic therapy by using noncontrast CT or MR imaging, to our knowledge, none currently exist that provide estimates of outcomes for perfusion CT–selected patients treated within defined treatment windows. The main objective of this systematic review and meta-analysis was to determine rates of death, disability, and symptomatic ICH (SICH) in patients with AIS selected for thrombolytic therapy by using perfusion CT. Specifically, the primary outcomes were as follows: (a) the modified Rankin Scale (mRS) score at 90 days, (b) the in-hospital SICH rate, and (c) 90-day all-cause mortality.

Implication for Patient Care nn These preliminary estimates drawn from our systematic review suggest that perfusion CT may be a useful neuroimaging modality for selecting patients for thrombolytic therapy.

Materials and Methods We included all randomized controlled trials (RCTs) that used perfusion CT neuroimaging to select patients with AIS for thrombolytic therapy. Perfusion CT is a relatively new imaging modality and has been evaluated in few RCTs. Therefore, we included all quasi-randomized studies; observational studies with matched, unmatched, or historical controls; longitudinal studies; case-control studies; and case series. Each of the included studies was required to report at least one of the outcomes of interest: mRS score, frequency of SICH, or all-cause mortality. The patient group was composed of adults (.18 years of age) of either sex with imaging-confirmed AIS. Furthermore, the stroke could have occurred in any cerebral territory, and the patients may or may not have had a prior stroke of any type. Patients had no contraindications to thrombolytic therapy and could have presented to any type of health care setting. All studies must have used perfusion CT to select patients for intravenous thrombolytic therapy. Studies that used any type of CT scanner to produce perfusion CT images were included in this review. All such studies, irrespective of their definition of perfusion deficit or Published online before print 10.1148/radiol.14140728  Content code: Radiology 2015; 274:103–114 Abbreviations: AIS = acute ischemic stroke CI = confidence interval ICH = intracranial hemorrhage mRS = modified Rankin Scale NIHSS = National Institutes of Health Stroke Scale RCT = randomized controlled trial SICH = symptomatic ICH Author contributions: Guarantor of integrity of entire study, K.R.B.; study concepts/study design or data acquisition or data analysis/ interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, K.R.B., D.D., R.I.A.; clinical studies, K.R.B., D.D., R.I.A.; statistical analysis, K.R.B.; and manuscript editing, all authors. Conflicts of interest are listed at the end of this article.

radiology.rsna.org  n Radiology: Volume 274: Number 1—January 2015

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis

the type, dose, and time of intravenous thrombolytic therapy, were included. A library scientist with experience in the conduct of systematic reviews independently conducted electronic searches in MEDLINE (from January 1946 to July 2012) and EMBASE (from January 1947 to July 2012) using the search strategy provided in Appendix E1 (online). To optimize the sensitivity of the review, a second independent search of MEDLINE and EMBASE was performed by one investigator (K.R.B.), and the results were amalgamated (18). Medical Subject Headings, or MeSH, terms used in the database search included the following: “stroke”; “brain ischemia”; “tomography, spiral computed”; “tomography, x-ray computed”; “perfusion imaging”; “magnetic resonance imaging”; “hemorrhage”; “cerebral hemorrhage”; “intracranial hemorrhage, hypertensive”; “death”; and “mortality.” Explosion terms were used for “stroke.” An additional electronic search was performed in the Cochrane Library (including the Cochrane Central Register of Controlled Trials and Technology Assessments) to July 22, 2012. The following supplementary electronic searches were conducted: (a) a search in PubMed (to July 18, 2012) using the terms “stroke” or “acute ischemic stroke” and “perfusion CT” or “multimodal imaging,” (b) a search in Google Scholar (to July 18, 2012) using the terms “acute ischemic stroke” and “perfusion CT” or “multimodal imaging,” and (c) a search in the stroke clinical trials registry (to July 23, 2012). The search of the “gray literature”—that is, literature that reports study results but does not appear within peer-reviewed journals and is presented elsewhere (eg, research meeting abstracts, technical reports)—included a list of theses, patents, health technology reviews, and conference proceedings. Bibliographies of studies that were short listed for inclusion in the meta-analysis were searched by hand, as were the bibliographies of relevant review articles. No restriction was placed on the language of the published studies.

Data Collection and Management Each of the abstracts of the citations identified by the search strategy was

independently reviewed for inclusion by two reviewers (K.R.B. and D.D.). A standardized abstract review tool (Table E1 [online]) was used to assess abstracts for inclusion. Reasons for exclusion were noted during the evaluation of each study (Figure). All data from each of the studies included in the final review were extracted separately by two reviewers (K.R.B. and D.D.) using a standardized data extraction table. Information extracted included study outcomes and relevant study details. Corresponding authors were contacted if data from any of the outcomes were not reported or were reported in a form that was not usable in the meta-analysis. Data were recorded in Excel: Mac 2008 (Microsoft, Redmond, Wash); RevMan, version 5.1 (the Cochrane Collaboration, Copenhagen, Denmark); and StatsDirect, version 2.7.9 (StatsDirect, Cheshire, England). Two reviewers assessed the risk of bias of each included study using the Cochrane risk of bias tool for RCTs (18) and the Newcastle-Ottawa Scale (NOS) for observational studies (19). The Cochrane tool estimates bias within five domains and across eight items, and the NOS estimates bias within three domains and across eight items. Bias was assessed for each outcome reported (18,20). Summary assessments of study bias for both RCTs and observational studies were made according to the Cochrane Handbook guidelines as follows: A low risk of bias was assigned when there was a low risk of bias for all domains; a moderate risk of bias, when there was an unclear risk of bias for one or more domains; and a high risk of bias, when there was a high risk of bias for one or more domains (21).

Meta-Analysis We followed published guidelines for the meta-analysis of RCTs and observational studies (18,22). Our analyses focused on generating pooled proportions of patients who experienced each of the clinical outcomes. Outcomes were extracted from each study, and pooled estimates of proportions and 95% confidence intervals (CIs) were calculated. Statistical heterogeneity among studies was tested by

Radiology: Volume 274: Number 1—January 2015  n  radiology.rsna.org

Burton et al

using the Cochrane Q test, and inconsistency was quantified with the I2 statistic.

Results Our systematic review identified 98 articles that were studied in detail (Figure). Two articles required translation into English (23,24). Thirty-one articles met our inclusion criteria, and 18 were excluded after full-text review (Table E2 [online]). The 13 included studies were published between 2000 and 2012, with seven published in 2011 or 2012. Study designs were a mixture of RCTs and prospective and retrospective cohorts (Table 1). Three RCTs contributed 21.5% of patients to the meta-analysis, while seven prospective cohort studies contributed 36.1% of patients and three retrospective cohort studies contributed 42.4% of patients. None of the RCTs included in this metaanalysis specifically compared perfusion CT with any other imaging modality in the selection of patients for thrombolysis. Instead, the RCTs compared one of the following: (a) desmoteplase and alteplase in the treatment of AIS (8), (b) alteplase and placebo in patients with unknown time of stroke onset (31), or (c) tenecteplase and alteplase in the treatment of AIS (33). Two RCTs demonstrated low to moderate risk of bias (31,33) (Table E3 [online]). Of the 18 outcomes reported by the 10 observational studies, 10 outcomes had low to moderate risk of bias (Table E4 [online]). All studies were performed in tertiary care centers, and the CT scanners used ranged from 16-section to 64-section scanners (Table 2). All study centers confirmed radiographically that the patients did not have an ICH at the start of the study. Five studies (25,26,28,30,31) used neuroradiologists to interpret study images, while the remaining studies either used radiologists without specialty neuroradiology training or did not report the expertise of those reading the images. Different perfusion CT mismatch definitions were used across the studies. Alteplase was the most commonly used thrombolytic drug, but 105

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis

Burton et al

to 6 (death). One study reported the proportion of patients within each mRS category, while the others reported scores in collapsed mRS categories. The most common method of reporting mRS outcomes was an mRS score of 2 or lower (widely considered to be a “good outcome” or a “mild stroke” [36]) and an mRS score greater than 2. Therefore, we reported pooled morbidity outcomes as mRS scores of 2 or lower or greater than 2. The majority of studies reported a 90-day mRS score, except for one (35), which reported an in-hospital score. Five studies contributed a total of 171 patients to the SICH analysis. The point estimate and 95% CI of the I2 statistic determined the appropriate pooling method: A point estimate greater than 0.30 led to the use of a random-effects model approach, whereas a point estimate of 0.30 or less led to the use of a fixed-effects model approach.

Mortality Across all studies with low or moderate chance of bias, the pooled 90-day mortality rate was 10.0% (95% CI: 5.4%, 15.9%; I 2 = 0) (Table 3). The mortality rate for patients receiving thrombolytic therapy within 3 hours was 12.5% (95% CI: 6.7%, 19.7%; I 2 = 0). For patients who received therapy beyond 3 hours, the mortality rate was 2.9% (95% CI: 0%, 12.7%; I 2 = 0).

Flowchart of selection of articles. CTP = perfusion CT, IA = intraarterial, t PA = tissue plasminogen activator.

in two studies, desmoteplase and tenecteplase were used, the latter as an alternative treatment arm to alteplase in an RCT (8,33). Time to thrombolytic treatment ranged from within 3 hours to a maximum of 9 hours after symptom onset. 106

For the calculation of the 90day mortality rate, 118 patients from four studies were eligible. Four studies containing a total of 151 patients contributed to the primary morbidity analysis. The mRS contains seven categories ranging from 0 (no symptoms)

Morbidity Among all studies with a low to moderate chance of bias that reported a 90day mRS score, 100 (66.2%) of 151 patients had a score of 2 or lower (Table 3). When these patients were categorized according to time to treatment, 42.5% (95% CI: 16.6%, 70.9%; I2 not estimable) of patients treated within 3 hours had a favorable outcome (mRS score, 2). Of the patients treated beyond 3 hours, 69.9% (95% CI: 0.0%, 83.5%; I2 = 45.1) had a favorable outcome. Among patients who received treatment within 3 hours, the proportion with an mRS score greater than 2 was 57.5% (95% CI, 29.2%, 83.4%; I2 not estimable). The proportion of patients treated beyond 3 hours with an

radiology.rsna.org  n Radiology: Volume 274: Number 1—January 2015

Radiology: Volume 274: Number 1—January 2015  n  radiology.rsna.org

2011

2012

2012

2012

2009

2007

2000

2012

2009

2012

2010

2012

2002

Eckert (25)

Eyding (26)

García-Bermejo (27)

Gentile (28)

Hacke (8)

Kloska (29)

Lee (30)

Michel (31)

Miteff (32)

Parsons (33)

Popiela (23)

Sillanpaa (34)

Wintermark (35)

Prospective  cohort Retrospective  cohort Prospective  cohort

Prospective  cohort RCT

Retrospective  cohort Prospective  cohort RCT

Prospective  cohort Retrospective  cohort Prospective  cohort Prospective  cohort RCT

Study Design

Multimodal CT (noncontrast   CT plus CT angiography)

Multimodal CT (noncontrast   CT plus CT angiography) Multimodal CT (noncontrast   CT plus CT angiography) Multimodal CT (noncontrast   CT plus CT angiography) Noncontrast CT in some   patients, not all Multimodal CT (noncontrast   CT plus CT angiography) Multimodal CT (noncontrast   CT plus CT angiography) Multimodal CT (noncontrast   CT plus CT angiography) Multimodal CT (noncontrast   CT plus CT angiography) Noncontrast CT

Multimodal CT (noncontrast   CT plus CT angiography) Multimodal CT (noncontrast   CT plus CT angiography) Noncontrast CT

Adjuvant Imaging Modality

mRS score (in hospital), mortality   (in hospital)

mRS score (at 90 d)

mRS score (at 90 d), SICH (in hospital),   mortality (at 90 d) SICH (in hospital), mortality (at 30 d)

mRS score (at 90 d), SICH (at 90 d),   mortality (at 90 d) mRS score (at 90 d), SICH (at 24 h and 4 d),   mortality (at 7 d and 90 d) mRS score (at 90 d)

mRS score (at 90 d), mortality (at 90 d)

mRS score (at 90 d)

SICH (in hospital)

mRS score (at 90 d), SICH (in hospital)

SICH (in hospital), mortality (in hospital)

mRS score (at 90 d), SICH (in hospital)

Outcome(s) Examined and Time Points

2

140

4

75

60

6

16

39

39

25

43

57

51

Sample Size (All Patients)*

2

140

4

n/a

0

0

9

39

0

0

0

47

40

Sample Size (Patients Treated within  3 h)

0

0

0

n/a

0

6

6

0

39

0

0

10

11

Sample Size (Patients Treated after . 3 h)

3

3

,2

6

6

.3

3, 7

3

.3

,5

.4.5

3, .3

3, 3–6

Reported Times to Treatment (h)

* The sample size of all patients within each study who met our inclusion criteria. Given that not all studies reported all of our outcomes of interest, in most instances, subsets of these patients contributed data to each outcome of interest.

Publication Year

First Author and Reference No.

General Characteristics of Included Studies

Table 1

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis Burton et al

107

108

June 2005–  March  2007

Not noted

Hacke (8) 2009

Kloska (29) 2007

Munster,  Germany

Multicenter  (DIAS-2  trial)

Tertiary hospital  (University   of Munster)

Tertiary hospital  (Temple  University   School of  Medicine) Tertiary  hospitals  (multicenter)

January–  October  2008

Philadelphia

Philips  Brilliance 40

CT Scanner Type

Yes

Yes

Unknown

Siemens No  Somatom,   16 section

Varied

3 h: mean,   62 min 6  31; .3 h:  mean,   332 min 6  250 NA

NA

Time from Symptom Onset to Presentation

“…tissue at risk as   defined by mean   transit time-CBV   mismatch greater   than 20%.” Not specifically stated NA   but assumed to be   qualitative evaluation,   on the basis of   Figure 1. “Mismatches were NA   evaluated qualitatively   at each site, in   accordance with local   practice, as a visually   apparent mismatch of   core and perfusion   lesions.” “…quantitative   mismatch assessments   were done  retrospectively….” “The ASPECT score NA   was applied to NECT   and PCT maps to   assess the extent of  ischemia.”

“Mismatch between   the ASPECTS of MTT   lesion and the extent   of CBV/EIH lesion.” “Mismatch was defined   if impaired rCBF   exceeded impaired   rCBV 20% or more.”

Penumbra or Mismatch Neurora­diologist? Parameters

GE Unknown  LightSpeed 64 or Toshiba Quilion 40 Siemens Yes  Sensation 16 or 64

Tertiary hospital Siemens  (University  Somatom  Hospital, 64  Knappschaftsk­ rankenhaus)

Gentile (28) 2012

Bochum,  Germany

Tertiary hospital  (Asklepios   Klinik Altona)

Setting

Tertiary hospital  (Hospital  Clinico  Universitario)

January–  December  2010

Eyding (26) 2012

Hamburg,  Germany

Location

GarcíaJanuary 2008– Valladolid, Bermejo  December  Spain (27) 2012  2010

2007–2008

Data Accrual Period

Eckert (25) 2011

First Author, Reference No., and Publication Year

Clinical Characteristics of Included Studies

Table 2

NA

Mean =  14.5 6  5.5

NA

NA

69.6 y 6  11.2

NA

NA

59.0

NA

NA

Alteplase   (0.9 mg/kg)

Alteplase   (0.9 mg/kg)

MCA

Alteplase   (0.9 mg/kg)

Desmoteplase   (90 and  125 mg/kg)  and  placebo

Table 2 (continues)

Various

Not Alteplase  stated   (0.9 mg/kg)

MCA

Not Alteplase  stated   (0.9 mg/kg)

Various

Percentage Thrombolytic of Male Vessel(s) Drug(s) and Patients Studied Dose

3 h: 3 h: 49;   73 y 6   .3 h:  10; .3 h:  40   71 y 6  12

NA

Mean Age of Patient Cohort

.4.5 h: .4.5 hrs: .4.5 h:   median =   69.02 y 6  55.8   9 (IQR,  13.2  5–18)

3 h:   median =  7; .3 h:   median =  11

NA

Baseline NIHSS Score

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis Burton et al

radiology.rsna.org  n Radiology: Volume 274: Number 1—January 2015

Radiology: Volume 274: Number 1—January 2015  n  radiology.rsna.org

June 2004–  December  2007

July 2006–   June 2008

Michel (31) 2012

Miteff (32) 2009

Parsons (33) 2008–2011 2012

September  1997–  October  1998

Data Accrual Period

Lee (30) 2000

First Author, Reference No., and Publication Year

Table 2 (continued)

Tertiary hospital  (Samsung  Medical  Center)

Setting

Newcastle,  Australia

Newcastle,  Australia

Unknown

GE Yes  LightSpeed  16  Advantage  (until  November  2005);  GE  LightSpeed   VCT 64  thereafter Philips No  Mx8000,   16 section

GE HighYes  speed  Advantage

Tertiary hospitals: 16 Or 64   three Australian  section,   stroke centers   per site

Tertiary hospital   (John Hunter  Hospital)

Lausanne, Tertiary hosptial  Switzerland  (Center  Hospitalier  Universitaire   Vaudois and   University of  Lausanne)

Seoul,  Korea

Location

CT Scanner Type

NA

3 h: mean,   2.2 h;   .3 h:  mean,   5.4 h

Time from Symptom Onset to Presentation

“…ratio of the MTT NA   lesion/CBV lesion.”   Using thresholds: MTT   delay of . 145%;   ,2.0 mL/100 g   for baseline infarct core  volume. “Perfusion lesion at NA   least 20% greater   than the infarct core.”

Not specifically stated,   but assumed to be   qualitative: “According   to the perfusion   deficit and collateral   blood flow on TPCT….” “MTT.145% of the   contralateral side   values and  CBV.2.0 mL/100 g.”

Penumbra or Mismatch Neurora­diologist? Parameters

NA

NA

.3 h:   median =   17 (IQR,  13–21)

At patient level, l3 h:  15.7;   .3 h:  12.3

Baseline NIHSS Score 3 h: 66.7; Various   .3 h:  33.3

NA

NA

NA

NA

Presumably  alteplase   but not  specifically  stated

Table 2 (continues)

Not Alteplase  stated   (0.9 mg/kg)  or  tenecteplase  (0.1–0.25  mg/kg)

Various

Alteplase   (0.9 mg/kg)

Alteplase   (0.9 mg/kg)

Percentage Thrombolytic of Male Vessel(s) Drug(s) and Patients Studied Dose

.3 h: .3 h: 50.0 MCA  median,   69.5 y  (IQR,   85–131 y)

3 h:   58.7 y;   .3 hrs:   69.9 y

Mean Age of Patient Cohort

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis Burton et al

109

110

3 h: 2.0 h rCBV threshold =   2.5 mL per 100 g. GE Unknown  LightSpeed Lausanne, Tertiary  Switzerland  hospital  (CHUV) Wintermark March 2000– (35) 2002  December  2000

GE No  LightSpeed   16 or  Philips  Brilliance  64 Tertiary hospital  (Tampere  University  Hospital) Tampere,  Finland Sillanpaa (34) January 2012  2006–  December  2007

Note.—ASPECTS = Alberta Stroke Program Early CT score, CBV = cerebral blood volume, EIH = early infarct hypodensity, IQR = interquartile range, MCA = middle cerebral artery, MTT = mean transit time, NA = not applicable, NIHSS = National Institutes of Health Stroke Scale, rCBF = regional cerebral blood flow, rCBV = regional cerebral blood volume, NECT = nonenhanced CT, PCT = perfusion CT, TPCT = triphasic CT.

Alteplase   (0.9 mg/kg)

Alteplase   (0.9 mg/kg)

Median with 64-row 64-row Various  64-row  scanner:  scanner: scanner: mean, 67.1 45; 8.4; y; 16-row 16-row median scanner: scanner: with 70.0 y 64 16-row scanner: 8.5 NA 68 y 100.0 Various

Not Alteplase  stated   (0.9 mg/kg) NA NA NA

“…in PCT the relationship NA  of the penumbra to the completed stroke area was greater than 1.” ASPECTS scoring in PCT NA  maps Tertiary hospitals: Not noted   three Polish  hospitals Krakow,  Poland Popiela (23) Not noted 2010

No

Setting Location Data Accrual Period First Author, Reference No., and Publication Year

Table 2 (continued)

CT Scanner Type

Penumbra or Mismatch Neurora­diologist? Parameters

Time from Symptom Onset to Presentation

Baseline NIHSS Score

Mean Age of Patient Cohort

Percentage Thrombolytic of Male Vessel(s) Drug(s) and Patients Studied Dose

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis

Burton et al

mRS score greater than 2 was not estimable, as only one study was eligible to contribute data to this outcome.

Rates of SICH From studies of low to moderate bias, the proportion of all patients with SICH detected after thrombolysis during the in-hospital period was 3.6% (95% CI: 1.4%, 6.8%; I2 = 0) (Table 3). Of the patients who were treated within 3 hours, 3.3% (95% CI: 0.7%, 7.7%; I2 = 0) had a SICH, compared with 3.9% (95% CI: 0.8%, 9.2%; I2 = 0) of patients treated beyond 3 hours. Sensitivity analyses.—We generated additional estimates of patient mortality, morbidity, and SICH rates for patient groups according to study quality and time to treatment. When mortality rates were estimated for patients from studies of any quality and at any time to treatment, the point estimates for proportions of patients for each mortality outcome did not change substantially (Table 4). Likewise, when we estimated rates of favorable morbidity (mRS score, 2) for all patient subgroups using studies with any quality, the proportions of patients experiencing this outcome did not differ substantially from the proportions of patients experiencing these morbidity outcomes in higher-quality studies. We estimated the SICH rate among patients within studies of any quality. Within this group, we found that 5.2% (95% CI: 3.0%, 8.1%) of patients experienced a SICH, which was similar to the rate for all patients in the primary analysis. Last, where feasible, we estimated outcomes for patients treated within 4.5 hours and those treated beyond 4.5 hours of symptom onset, and these results were not significantly different from the results reported for the other subgroups (Table 4). Comparison of Perfusion CT Outcomes to Noncontrast CT and MR Imaging Outcomes Two studies (37,38) published outcomes for patients with AIS selected by using noncontrast CT or MR imaging to receive intravenous thrombolysis. Both studies reported end points identical to radiology.rsna.org  n Radiology: Volume 274: Number 1—January 2015

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis

Table 3 Results of Analyses of Outcomes according to Timing of Delivery of Intravenous Thrombolysis Therapy Outcome 90-Day mortality   All patients   Patients treated  3 hours   after symptom onset   Patients treated . 3 hours   after symptom onset mRS score  2 at 90 days   All patients   Patients treated  3 hours   after symptom onset   Patients treated . 3 hours   after symptom onset mRS score . 2 at 90 days   All patients   Patients treated  3 hours   after symptom onset   Patients treated . 3 hours   after symptom onset SICH in hospital   All patients   Patients treated  3 hours   after symptom onset   Patients treated . 3 hours   after symptom onset

Percentage of Patients 10.0 12.5

95% CI (%) 5.4, 15.9 6.7, 19.7

No. of Patients/No. of Studies

I 2 (%)

118/4 95/3

0 (0, 61.0) 0 (0, 72.9)

2.9

0, 12.7

23/3

0 (0, 72.9)

59.4 42.5

38.2, 78.9 16.6, 70.9

100/4 47/2

73.1 (0, 87.3) NA

69.9

0, 83.5

53/3

45.1 (0, 83.5)

38.4 57.5

6.6, 77.5 29.1, 83.4

51/2 47/2

82.4 (0, 92.5) NA

n/a

NA

NA

NA

3.6 3.3

1.4, 6.8 0.7, 7.7

171/5 94/3

0 (0, 56.3) 0 (0, 72.9)

3.9

0.8, 9.2

77/5

0 (0, 64.1)

Note.—NA = not applicable. Data in parentheses are 95% CIs. All studies were considered to have low evidence quality and unclear risk of bias for all outcomes.

ours (90-day mortality rate, 90-day mRS score  2, and SICH rate). Each study estimated these outcomes for three patient groups—those who underwent noncontrast CT within 3 hours, those who underwent MR imaging within 3 hours, and those who underwent MR imaging beyond 3 hours. Our outcome estimates were statistically similar to each of those of the three imaging groups included in these two studies except in two instances. In one study (37), the point estimate for the rate of SICH in patients selected by using noncontrast CT within 3 hours was significantly higher (9.0%) than our estimate of the rate of SICH in patients selected by using perfusion CT at the same time point (3.3%; 95% CI: 0.7%, 7.7%), while the mortality rate for patients selected by using MR imaging beyond 3 hours was significantly higher (13.3%)

than our estimate for patients selected by using perfusion CT at the same time point.

Discussion This systematic review and meta-analysis estimated the pooled outcomes from studies where perfusion CT imaging was used to select patients with AIS for intravenous thrombolysis who were treated within or beyond 3 hours from symptom onset. We found that patients treated beyond 3 hours had higher rates of SICH than patients treated within 3 hours, although these differences were small and not statistically significant. Surprisingly, we found that patients treated beyond 3 hours had lower mortality rates and more favorable outcomes (defined as an mRS score  2) compared with those in patients

Radiology: Volume 274: Number 1—January 2015  n  radiology.rsna.org

Burton et al

treated within 3 hours. Although these latter results were not statistically significant, they warrant discussion. Outcomes of patients with AIS may be influenced by many factors, including patient age and stroke severity at presentation (39). Owing to differences in the way patient ages were reported, we were unable to test the significance of differences in patient ages between the two time groups. However, mean patient ages (69–71 years) appeared similar in each of the studies included in our primary analyses, except for that in one patient group (in the study of Lee et al [30], the mean age of patients treated within 3 hours was 58.7 years). Stroke severity at presentation could have influenced our findings, because patients in the earlier treatment groups tended to have higher baseline NIHSS scores than patients in the later treatment groups. The type of artery involved in the stroke may have influenced our results, particularl the relatively high mortality rate among patients treated within 3 hours. Of the studies that contributed patients to this outcome, two (29,30) restricted their samples to patients with middle cerebral artery territory (severe) strokes. Both studies reported high mortality rates. Other factors may have influenced our paradoxical findings. Studies in the greater than 3-hour period tended to have been conducted more recently, and in those that provided data about the outcome of favorable morbidity, a greater proportion of patients were selected for thrombolysis by using more advanced perfusion CT scanners (64 section vs 16 section). Perfusion CT penumbra or mismatch parameters also differed among studies. Perfusion CT mismatch definitions and thresholds have been shown to be important determinants of stroke treatment risks, and there has recently been a call for standardization of perfusion CT mismatch definitions and thresholds for use in perfusion CT studies (40,41). Study quality differed, with the patients treated beyond 3 hours who experienced a more favorable outcome (mRS score, 2) tending to be 111

112

3 h .3 h 4.5 h .4.5 h

1.1, 8.6 NA 0.3, 12.0 0, 32.6

3.0, 8.1

5.2

4.0 NA 2.2 8.6

14.0, 77.6 NA

18.8, 59.1

37.8

44.6 NA

39.1, 83.1 34.9, 73.3 35.8, 97.8

52.5 54.4 73.6

41.6, 66.4

54.1

Note.—NA = not applicable. Data in parentheses are 95% CIs.

       

  3 h   .3 h SICH in hospital   Any time

  3 h   .3 h   .4.5 h mRS score . 2   Any time

7.9, 20.9 NA 0.2, 15.2

6.9, 15.3

95% CI (%)

13.8 NA 5.0

10.7

Mortality   Any time

  3 h   .3 h   Up to 4.5 h mRS score  2   at 90 days   Any time

Percentage

Outcome

Results of Sensitivity Analyses

Table 4

98/4 NA 21/2 48/2

275/8

49/3 NA

128/4

187/3 92/4 46/2

465/9

101/5 NA 28/3

199/7

No. of Patients/No. of Studies

NA NA NA

0 (0, 67.9)

0 (0, 49.8)

71.4 (0, 89.5) NA

79.3 (43.3, 88.9)

89.4 (72.8, 94.2) 68.6 (0, 85.7) NA

85.8 (77.2, 90.1)

10.5 (0, 67.7) NA 1.8 (0, 73.4)

0 (0, 52.7)

I2 (%)

Low, unclear/high NA Low, unclear Low

Low, unclear/high

Unclear/high NA

Unclear/high

Unclear/high Low, unclear/high Low/unclear

Low, unclear/high

Unclear/high NA Unclear/high

Low, unclear/high

Evidence Quality/Risk of Bias

Includes: patients from any time group (including those treated 6 hours);   studies of any quality; any type of thrombolytic medication. Includes: studies of any quality; alteplase only. No additional patients to add to sensitivity analysis. Includes: studies of any quality; alteplase only. Includes: studies of any quality; alteplase only.

Includes: patients from any time group (including those treated  6 hours);   studies of any quality; any type of thrombolytic medication. Includes: studies of any quality; alteplase only. No additional patients to add to sensitivity analysis.

Includes: patients from any time group (including those treated  6 hours);   studies of any quality; any type of thrombolytic medication. Includes: studies of any quality; alteplase only. Includes: studies of any quality; any type of thrombolytic medication. Includes: studies of any quality; alteplase only.

Includes: patients from any time group (including those treated  6 hours);   studies of any quality; any type of thrombolytic medication. Includes: studies of any quality; alteplase only. No additional patients to add to sensitivity analysis. Includes: studies of any quality; alteplase only.

Comments

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis Burton et al

radiology.rsna.org  n Radiology: Volume 274: Number 1—January 2015

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis

included in studies that were of higher quality. Finally, the small sample sizes of the studies included in this systematic review mean that some of the differences among subgroups occurred by chance. Overall, our findings suggest that thrombolyis in patients selected by using perfusion CT is associated with outcomes that are comparable with those in studies that used other imaging modalities and that outcomes in selected patients treated with intravenous thrombolysis more than 3 hours after symptom onset appear good. Our study had a number of strengths. We conducted an exhaustive review of the literature, which included a gray literature search with no restriction on the language of publication. Data abstraction bias was minimized, as data abstraction was performed by two individuals independently. Finally, the studies included in the systematic review were similar in terms of the thrombolytic type and the type of hospital to which the patients presented. Limitations of our study included the heterogeneity observed among these small studies, including the different definitions of perfusion CT mismatch, different methods of recording the mRS, and differences in the age and sex of the patients. The studies varied in methodologic quality, although our primary analysis was conducted in studies with low to moderate risk of bias. We were unable to include a number of studies because detailed study data were unavailable, were presented in an unusable format, or were not provided by authors even when contacted (42). Because of the small number of patients, the results must be interpreted with caution. Unfortunately, all studies did not report the most important predictor of outcome, which is the severity of stroke at presentation (commonly assessed by using the NIHSS score). Most of the studies in our primary analysis used an adjuvant neuroimaging modality to assess the patients at presentation (in the majority of studies, noncontrast CT in conjunction with CT angiography). Two studies (26,31) explicitly noted that CT angiography

was not used to inform the thrombolysis decision, one study (26) noted that CT angiography was also used to inform decisions about mechanical revascularization, and another study (29) specified that CT angiography was used to assess the site of vessel occlusion and vessel recanalization status. One study (25) noted that CT angiography information was used to inform the thrombolysis decision, and we could not separately assess the effect of the CT angiography information. We assumed that the thrombolysis decision was based largely on information garnered from noncontrast CT and perfusion CT. Finally, no studies randomized patients to thrombolytic therapy guided by perfusion CT compared with another imaging modality. Head-to-head RCTs yield the highest-quality evidence, and such studies should be done in the future. Current thrombolysis guidelines have expanded the treatment window to 4.5 hours after symptom onset for patients who undergo penumbral imaging (43). Penumbral imaging (perfusion CT or MR imaging) is increasingly attractive because of its ability to help select patients for thrombolysis among patients presenting beyond 4.5 hours. Therefore, outcomes should be generated for patients treated between 3.0 and 4.5 hours and for those treated beyond 4.5 hours. In sensitivity analyses, we were able to create some subgroups of patients who were treated within these time windows. Unfortunately, most of the estimates were based on data from two or three studies, and some estimates were not calculable because of a paucity of studies (26,27,30). Future studies will hopefully include more patients treated beyond 4.5 hours. In conclusion, we generated estimates of mortality, morbidity, and SICH rates in patients in whom thrombolytic therapy was guided by perfusion CT. We studied patients treated within, as well as after, 3 hours of symptom onset. Patients treated after 3 hours of symptom onset had acceptable outcomes. However, these estimates should be viewed with caution because of the small sizes of the studies and heterogeneity among

Radiology: Volume 274: Number 1—January 2015  n  radiology.rsna.org

Burton et al

studies. Future research should include head-to-head comparisons of perfusion CT and MR imaging in the selection of patients with AIS for thrombolytic therapy within adequately powered RCTs. Disclosures of Conflicts of Interest: K.R.B. disclosed no relevant relationships. D.D. disclosed no relevant relationships. R.I.A. disclosed no relevant relationships. A.R.M. disclosed no relevant relationships. M.K.K. disclosed no relevant relationships. A.L. disclosed no relevant relationships.

References 1. Gillum RF, Wilson JB. The burden of stroke and its sequelae. Dis Manag Health Outcomes 1997;1(2):84–94. 2. Lindgren P, Glader EL, Jönsson B. Utility loss and indirect costs after stroke in Sweden. Eur J Cardiovasc Prev Rehabil 2008;15(2): 230–233. 3. Donnan GA, Baron JC, Ma H, Davis SM. Penumbral selection of patients for trials of acute stroke therapy. Lancet Neurol 2009;8(3): 261–269. 4. Cost of caring for stroke patients double that of earlier estimates, study finds. Canadian Stroke Network. http://www.canadianstrokenetwork.ca/index.php5/news/. Published 2010. Accessed March 2014. 5. Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. N Engl J Med 1995; 333(24):1581–1587. 6. Stroke Unit Trialists’ Collaboration. Organised inpatient (stroke unit) care for stroke. Cochrane Database Syst Rev 2007;(4): CD000197. 7. Muir KW, Buchan A, von Kummer R, Rother J, Baron JC. Imaging of acute stroke. Lancet Neurol 2006;5(9):755–768. 8. Hacke W, Furlan AJ, Al-Rawi Y, et al. Intravenous desmoteplase in patients with acute ischaemic stroke selected by MRI perfusiondiffusion weighted imaging or perfusion CT (DIAS-2): a prospective, randomised, double-blind, placebo-controlled study. Lancet Neurol 2009;8(2):141–150. 9. Mishra NK, Albers GW, Davis SM, et al. Mismatch-based delayed thrombolysis: a meta-analysis. Stroke 2010;41(1):e25–e33. 10. Kidwell CS, Alger JR, Saver JL. Beyond mismatch: evolving paradigms in imaging the ischemic penumbra with multimodal magnetic resonance imaging. Stroke 2003;34(11): 2729–2735.

113

EVIDENCE-BASED PRACTICE: Perfusion CT for Selecting Patients with Stroke for Thrombolysis

11. Hoeffner EG, Case I, Jain R, et al. Cerebral perfusion CT: technique and clinical applications. Radiology 2004;231(3):632–644. 12. Parsons MW. Perfusion CT: is it clinically useful? Int J Stroke 2008;3(1):41–50. 13. Lansberg MG, Albers GW, Wijman CA. Symptomatic intracerebral hemorrhage following thrombolytic therapy for acute ischemic stroke: a review of the risk factors. Cerebrovasc Dis 2007;24(1):1–10. 14. Wardlaw JM, Mielke O. Early signs of brain infarction at CT: observer reliability and outcome after thrombolytic treatment—systematic review. Radiology 2005;235(2):444–453. 15. Wardlaw JM, Warlow CP, Counsell C. Systematic review of evidence on thrombolytic therapy for acute ischaemic stroke. Lancet 1997;350(9078):607–614. 16. Charidimou A, Kakar P, Fox Z, Werring DJ. Cerebral microbleeds and the risk of intracerebral haemorrhage after thrombolysis for acute ischaemic stroke: systematic review and meta-analysis. J Neurol Neurosurg Psychiatry 2013;84(3):277–280. 17. Kane I, Sandercock P, Wardlaw J. Magnetic resonance perfusion diffusion mismatch and thrombolysis in acute ischaemic stroke: a systematic review of the evidence to date. J Neurol Neurosurg Psychiatry 2007;78(5): 485–491. 18. Higgins JP, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions. 5th ed. http://handbook.cochrane. org/. Accessed February 18, 2013. 19. Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in metaanalyses. Ottawa Hospital Research Institute. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed March 2014. 20. Higgins JP, Altman DG. Assessing risk of bias in included studies. In: Higgins JP, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions. Chichester, England: Wiley, 2008; 187–242. 21. Higgins JP, Altman DG, Sterne JA. Assessing risk of bias in included studies. In: Cochrane Handbook for Systematic Reviews of Interventions. http://handbook.cochrane. org/chapter_8/8_assessing_risk_of_bias_in_ included_studies.htm. Accessed September 30, 2013. 22. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies

114

Burton et al

in Epidemiology (MOOSE) group. JAMA 2000;283(15):2008–2012.

ischaemic stroke. Brain 2009;132(Pt 8): 2231–2238.

23. Popiela TJ, Urbanik A, Słowik A. Results of thrombolyses procedures in acute ischemic cerebral stroke realized in Kraków 2004-2007. Grant Ministry of Science and Information [in Polish]. Przegl Lek 2010;67(4):284–288.

33. Parsons M, Spratt N, Bivard A, et al. A randomized trial of tenecteplase versus alteplase for acute ischemic stroke. N Engl J Med 2012;366(12):1099–1107.

24. Cortijo E, Calleja AI, Garcia-Bermejo P, et al. Perfusion computed tomography makes it possible to overcome important SITSMOST exclusion criteria for the endovenous thrombolysis of cerebral infarction [in Spanish]. Rev Neurol 2012;54(5):271–276. 25. Eckert B, Küsel T, Leppien A, Michels P, Müller-Jensen A, Fiehler J. Clinical outcome and imaging follow-up in acute stroke patients with normal perfusion CT and normal CT angiography. Neuroradiology 2011;53(2):79–88. 26. Eyding J, Wiebringhaus R, Klein FG, et al. Multimodal CT imaging and recanalizing therapy in acute ischemic stroke: retrospective analysis of a one-year single-center experience. Eur Neurol 2012;67(4):193–199. 27. García-Bermejo P, Calleja AI, Pérez-Fernández S, et al. Perfusion computed tomographyguided intravenous thrombolysis for acute ischemic stroke beyond 4.5 hours: a casecontrol study. Cerebrovasc Dis 2012;34(1): 31–37. 28. Gentile NT, Cernetich J, Kanamalla US, et al. Expedited computed tomography perfusion and angiography in acute ischemic stroke: a feasibility study. J Emerg Med 2012; 43(2):308–315.

34. Sillanpaa N, Rusanen H, Saarinen JT, Dastidar P, Soimakallio S. Comparison of 64-row and 16-row multidetector CT in the perfusion CT evaluation of acute ischemic stroke patients receiving intravenous thrombolytic therapy. Neuroradiology 2012;54(9):957–963. 35. Wintermark M, Reichhart M, Thiran JP, et al. Prognostic accuracy of cerebral blood flow measurement by perfusion computed tomography, at the time of emergency room admission, in acute stroke patients. Ann Neurol 2002;51(4):417–432. 36. Weisscher N, Vermeulen M, Roos YB, Haan RJ. What should be defined as good outcome in stroke trials; a modified Rankin score of 0-1 or 0-2? J Neurol 2008;255(6):867–874. 37. Köhrmann M, Jüttler E, Fiebach JB, et al. MRI versus CT-based thrombolysis treatment within and beyond the 3 h time window after stroke onset: a cohort study. Lancet Neurol 2006;5(8):661–667. 38. Schellinger PD, Thomalla G, Fiehler J, et al. MRI-based and CT-based thrombolytic therapy in acute stroke within and beyond established time windows: an analysis of 1210 patients. Stroke 2007;38(10):2640–2645. 39. Saposnik G. The art of estimating outcomes and treating patients with stroke in the 21st century. Stroke 2014;45(6):1603–1605.

29. Kloska SP, Dittrich R, Fischer T, et al. Perfusion CT in acute stroke: prediction of vessel recanalization and clinical outcome in intravenous thrombolytic therapy. Eur Radiol 2007;17(10):2491–2498.

40. Collins PD, Dani KA, Moreton F, et al. Large CT perfusion-defined mismatch predicts early improvement after IV thrombolysis in acute ischaemic stroke. J Neurol Neurosurg Psychiatry 2013;84(11):e2.

30. Lee KH, Lee SJ, Cho SJ, et al. Usefulness of triphasic perfusion computed tomography for intravenous thrombolysis with tissuetype plasminogen activator in acute ischemic stroke. Arch Neurol 2000;57(7):1000–1008.

41. Wintermark M, Albers GW, Broderick JP, et al. Acute stroke imaging research roadmap II. Stroke 2013;44(9):2628–2639.

31. Michel P, Ntaios G, Reichhart M, et al. Perfusion-CT guided intravenous thrombolysis in patients with unknown-onset stroke: a randomized, double-blind, placebo-controlled, pilot feasibility trial. Neuroradiology 2012;54(6):579–588. 32. Miteff F, Levi CR, Bateman GA, Spratt N, McElduff P, Parsons MW. The independent predictive utility of computed tomography angiographic collateral status in acute

42. Furtado AD, Smith WS, Koroshetz W, et al. Perfusion CT imaging follows clinical severity in left hemispheric strokes. Eur Neurol 2008;60(5):244–252. 43. Del Zoppo GJ, Saver JL, Jauch EC, Adams HP Jr; American Heart Association Stroke Council. Expansion of the time window for treatment of acute ischemic stroke with intravenous tissue plasminogen activator: a science advisory from the American Heart Association/American Stroke Association. Stroke 2009;40(8):2945–2948.

radiology.rsna.org  n Radiology: Volume 274: Number 1—January 2015

Perfusion CT for selecting patients with acute ischemic stroke for intravenous thrombolytic therapy.

To determine rates of death, disability, and symptomatic intracranial hemorrhage ( SICH symptomatic ICH ) among patients with acute ischemic stroke se...
354KB Sizes 0 Downloads 7 Views