The new england [619247]

The new england
journal of medicinen engl j med 378;1 nejm.org January 4, 2018 11established in 1812 January 4, 2018 vol. 378 no. 1
The authors’ full names, academic de –
grees, and affiliations are listed in the Ap –
pendix. Address reprint requests to Dr.
Jovin at the University of Pittsburgh Med –
ical Center Stroke Institute, Department of Neurology, Presbyterian University Hos
–
pital, 200 Lothrop St., C-400, Pittsburgh, PA 15217, or at jovintg@ upmc . edu.
* A complete list of sites and investiga
–
tors in the DAWN trial is provided in the Supplementary Appendix, available at NEJM.org.
Drs. Nogueira and Jovin contributed equal
–
ly to this article.
This article was published on November 11,
2017, at NEJM.org.
N Engl J Med 2018;378:11-21.
DOI: 10.1056/NEJMoa1706442
Copyright © 2017 Massachusetts Medical Society.BACKGROUND
The effect of endovascular thrombectomy that is performed more than 6 hours after
the onset of ischemic stroke is uncertain. Patients with a clinical deficit that is dispro –
portionately severe relative to the infarct volume may benefit from late thrombectomy.
METHODS
We enrolled patients with occlusion of the intracranial internal carotid artery or proximal middle cerebral artery who had last been known to be well 6 to 24 hours earlier and who had a mismatch between the severity of the clinical deficit and the infarct volume, with mismatch criteria defined according to age (<80 years or ≥80 years). Patients were randomly assigned to thrombectomy plus standard care (the thrombectomy group) or to standard care alone (the control group). The coprimary end points were the mean score for disability on the utility-weighted modified Rankin scale (which ranges from 0 [death] to 10 [no symptoms or disability]) and the rate of functional independence (a score of 0, 1, or 2 on the modified Rankin scale, which ranges from 0 to 6, with higher scores indicating more severe disability) at 90 days.
RESULTS
A total of 206 patients were enrolled; 107 were assigned to the thrombectomy group and 99 to the control group. At 31 months, enrollment in the trial was stopped because of the results of a prespecified interim analysis. The mean score on the utility-weight –
ed modified Rankin scale at 90 days was 5.5 in the thrombectomy group as compared with 3.4 in the control group (adjusted difference [Bayesian analysis], 2.0 points; 95% credible interval, 1.1 to 3.0; posterior probability of superiority, >0.999), and the rate of functional independence at 90 days was 49% in the thrombectomy group as com –
pared with 13% in the control group (adjusted difference, 33 percentage points; 95% credible interval, 24 to 44; posterior probability of superiority, >0.999). The rate of symptomatic intracranial hemorrhage did not differ significantly between the two groups (6% in the thrombectomy group and 3% in the control group, P = 0.50), nor did
90-day mortality (19% and 18%, respectively; P = 1.00).
CONCLUSIONS
Among patients with acute stroke who had last been known to be well 6 to 24 hours earlier and who had a mismatch between clinical deficit and infarct, outcomes for disability at 90 days were better with thrombectomy plus standard care than with standard care alone. (Funded by Stryker Neurovascular; DAWN ClinicalTrials.gov number, NCT02142283.)abstractThrombectomy 6 to 24 Hours after Stroke with a Mismatch
between Deficit and Infarct
R.G. Nogueira, A.P. Jadhav, D.C. Haussen, A. Bonafe, R.F. Budzik, P. Bhuva, D.R. Yavagal, M. Ribo, C. Cognard,
R.A. Hanel, C.A. Sila, A.E. Hassan, M. Millan, E.I. Levy, P. Mitchell, M. Chen, J.D. English, Q.A. Shah, F.L. Silver,
V.M. Pereira, B.P. Mehta, B.W. Baxter, M.G. Abraham, P. Cardona, E. Veznedaroglu, F.R. Hellinger, L. Feng,
J.F. Kirmani, D.K. Lopes, B.T. Jankowitz, M.R. Frankel, V. Costalat, N.A. Vora, A.J. Yoo, A.M. Malik, A.J. Furlan,
M. Rubiera, A. Aghaebrahim, J.-M. Olivot, W.G. Tekle, R. Shields, T. Graves, R.J. Lewis, W.S. Smith, D.S. Liebeskind,
J.L. Saver, and T.G. Jovin, for the DAWN Trial Investigators*
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 12The new england journal of medicine
Previous randomized trials that in –
volved patients with acute stroke1-6 showed
that endovascular thrombectomy had a clin –
ical benefit when it was performed within 6 hours
after the onset of stroke symptoms7 and that the
benefit diminished as the interval between the time that the patient was last known to be well and thrombectomy increased.
8 For the purposes of
determining eligibility for thrombolysis or throm –
bectomy, the time that the patient was last known to be well has typically been considered to be the time of stroke onset, including among patients who wake up with stroke symptoms or have an uncertain time of stroke onset. There is limited information on the effect of thrombectomy that is performed more than 6 hours after the time that the patient was last known to be well, particu –
larly among patients with ischemic brain tissue that has not yet undergone infarction and may be salvaged with reperfusion. Patients with brain tissue that may be salvaged with reperfusion can be identified by the presence of a clinical deficit that is disproportionately severe relative to the vol –
ume of infarcted tissue on imaging studies (see Section S3 in the Supplementary Appendix, avail –
able with the full text of this article at NEJM.org).
9
Results of previous nonrandomized studies
have suggested that patients who have a mismatch between the volume of brain tissue that may be salvaged and the volume of infarcted tissue could benefit from reperfusion of occluded proximal anterior cerebral vessels, even when the reperfu –
sion is performed more than 6 hours after the patient was last known to be well.
10,11 In the DAWN
(DWI or CTP Assessment with Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention with Trevo) trial, we compared endovascular thrombectomy plus standard medical care with standard medi –
cal care alone for the treatment of patients with acute stroke who had last been known to be well 6 to 24 hours earlier and who had a mismatch between clinical deficit and infarct.
Methods
Trial Design
The DAWN trial was a multicenter, prospective, randomized, open-label trial with a Bayesian adap –
tive–enrichment design and with blinded assess –
ment of end points.
12 The trial protocol was ap –
proved by the institutional review board at each participating site. Enrolled patients or their sur -rogates provided written informed consent. The trial was designed and conducted by a steering committee, which was composed of independent academic investigators and statisticians, in collab –
oration with the sponsor, Stryker Neurovascular, which provided funding and the thrombectomy devices for the trial and performed regulatory monitoring at each site and central database main –
tenance. The first drafts of the manuscript were written by the first and last authors, with input from all the authors and with no writing assistance from the sponsor. The authors had unrestricted access to the data. The data analysis was per –
formed by a data-management staff from Stryker Neurovascular, with oversight from independent statisticians. All the authors vouch for the com –
pleteness and accuracy of the reported data and the fidelity of the trial to the protocol. Decisions related to safety, adaptive–enrichment techniques, and trial discontinuation were made at the recom –
mendation of an independent data and safety monitoring board.
Information on the inclusion and exclusion
criteria, interventions, and assessments has been published previously.
12 The trial protocol and sta –
tistical analysis plan are available at NEJM.org.
Patients
Patients were eligible for inclusion in the trial if they had evidence of occlusion of the intracranial internal carotid artery, the first segment of the middle cerebral artery, or both on computed tomo –
graphic (CT) angiography or magnetic resonance angiography. In addition, patients had to have a mismatch between the severity of the clinical deficit and the infarct volume, which was defined according to the following criteria: those in Group A were 80 years of age or older, had a score of 10 or higher on the National Institutes of Health Stroke Scale (NIHSS; scores range from 0 to 42, with higher scores indicating a more severe defi –
cit), and had an infarct volume of less than 21 ml; those in Group B were younger than 80 years of age, had a score of 10 or higher on the NIHSS, and had an infarct volume of less than 31 ml; and those in Group C were younger than 80 years of age, had a score of 20 or higher on the NIHSS, and had an infarct volume of 31 to less than 51 ml. Infarct volume was assessed with the use of diffu –
sion-weighted magnetic resonance imaging (MRI) or perfusion CT and was measured with the use of automated software (RAPID, iSchemaView).
Other inclusion criteria were an age of 18 years A Quick Take is
available at
NEJM.org
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 13Thrombectomy 6 to 24 Hours after Stroke
or older, an interval between the time that the pa –
tient was last known to be well and randomization
of 6 to 24 hours, a prestroke score of 0 or 1 on the modified Rankin scale (which ranges from 0 to 6, with a score of 0 indicating no disability and higher scores indicating more severe disability), no evidence of intracranial hemorrhage on CT or MRI, and no evidence of an infarct involving more than one third of the territory of the middle cerebral artery on CT or MRI at baseline. Patients either did not meet the usual criteria for treatment with intra –
venous alteplase because of a late presentation or received treatment with intravenous alteplase and had persistent occlusion of the vessel at the time that they were eligible for enrollment in the trial.
Treatment
Patients were randomly assigned, in a 1:1 ratio, to thrombectomy plus standard medical care (the thrombectomy group) or to standard medical care alone (the control group). Randomization was performed with the use of a central, Web-based procedure, with block minimization processes to balance the two treatment groups, and was strati –
fied according to mismatch criteria (Group A, Group B, or Group C), the interval between the time that the patient was last known to be well and randomization (6 to 12 hours or >12 to 24 hours), and the occlusion site (intracranial internal carotid artery or the first segment of the middle cerebral artery).
The trial was conducted at 26 centers in the
United States, Canada, Europe, and Australia; at least 40 mechanical thrombectomy procedures had been performed at each center annually. Enrolled patients were admitted to stroke units or inten –
sive care units. Patients who had not received intravenous alteplase could receive therapy with antiplatelet agents, which could be started with –
in 24 hours after randomization. Standard med –
ical care was provided in accordance with local guidelines (see Section S6 in the Supplementary Appendix).
12 Thrombectomy was performed with
the use of the Trevo device (Stryker Neurovascu –
lar), a retrievable self-expanding stent that is used to remove occlusive thrombi and restore blood flow. Rescue reperfusion therapy with other de –
vices or pharmacologic agents was not permitted. Concomitant stenting of the cervical internal ca –
rotid artery at the time of thrombectomy was not permitted, but carotid angioplasty was permitted if necessary to allow for intracranial access for the catheter to deploy the retriever device.End Points
For the coprimary end points, scores on the mod –
ified Rankin scale were obtained through in-person, formal, structured interviews with patients and caregivers that were performed by local certi –
fied assessors
13,14 who were unaware of the treat –
ment assignments.15 For the 43 patients for whom
in-person assessment was not possible, telephone interviews with patients, caregivers, or both were performed.
The first primary end point was the mean score
for disability on the utility-weighted modified Rankin scale at 90 days. To determine the utility-weighted score, the score on the modified Rankin scale is weighted according to average values calculated from patient-centered and clinician-centered studies.
16-18 The following weights are
assigned to scores 0 through 6 on the modified Rankin scale: 10.0, 9.1, 7.6, 6.5, 3.3, 0, and 0, re –
spectively. The utility-weighted modified Rankin scale ranges from 0 (death) to 10 (no symptoms or disability).
The second primary end point was the rate of
functional independence (defined as a score of 0, 1, or 2 on the modified Rankin scale) at 90 days. This end point was changed from a secondary end point to a coprimary end point at the request of the Food and Drug Administration at 30 months after the start of the trial, when the trial was still blinded.
Prespecified secondary end points were an
early therapeutic response (defined as a decrease in the NIHSS score of ≥10 from baseline or an NIHSS score of 0 or 1 on day 5, 6, or 7 of hospital –
ization or at discharge if it occurred before day 5), death from any cause at 90 days, centrally adjudi-cated infarct volume and change from baseline in the infarct volume at 24 hours, and evidence of recanalization of the occluded vessel on CT an –
giography or magnetic resonance angiography at 24 hours (see Section S3 in the Supplementary Ap –
pendix). In the thrombectomy group, a secondary end point was centrally adjudicated successful re –
canalization (on the basis of findings on postpro –
cedural conventional angiography), which was de –
fined as a grade of 2b or 3 on the modified Thrombolysis in Cerebral Infarction scale (which ranges from 0 to 3, with a grade of 2b or 3 indicat –
ing reperfusion of >50% of the affected territory). A prespecified subgroup analysis for heterogene –
ity of treatment effect was performed, with sub –
groups defined according to mismatch criteria (Group A, Group B, or Group C), the interval be –
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 14The new england journal of medicine
tween the time that the patient was last known to
be well and randomization (6 to 12 hours or >12 to 24 hours), occlusion site (intracranial internal carotid artery or the first segment of the middle cerebral artery), sex, age (<80 years or ≥80 years), baseline NIHSS score (10 to 17 or >17), type of stroke onset (on awakening, unwitnessed stroke, or witnessed stroke), and time from the first ob –
servation of symptoms to randomization (0 to 6 hours or >6 hours).
The main safety end point was stroke-related
death at 90 days. Other safety end points included neurologic deterioration (defined as an increase in the NIHSS score of ≥4 points within 5 days after stroke that was not attributed to intracranial hemorrhage or malignant cerebral edema) and symptomatic intracranial hemorrhage (defined according to European Cooperative Acute Stroke Study III criteria as the presence of extravascular blood in the cranium that was associated with an increase in the NIHSS score of ≥4 points or death and was judged to be the predominant cause of neurologic deterioration) within 24 hours after randomization.
19 Safety end points, procedure-
related complications, and serious adverse events were adjudicated by an independent clinical-events committee.
Statistical Analysis
The adaptive trial design allowed for a sample size ranging from 150 to 500 patients. During interim analyses, the decision to stop or continue enroll –
ment was based on a prespecified calculation of the probability that thrombectomy plus standard care would be superior to standard care alone with respect to the first primary end point. The enrich –
ment trial design gave us the flexibility to identify whether the benefit of the trial intervention was restricted to a subgroup of patients with relatively small infarct volumes at baseline. The interim analyses, which included patients with available follow-up data at the time of the analysis, were prespecified to test for the futility, enrichment, and success of the trial.
The first primary analysis, which evaluated the
posterior probability that thrombectomy plus stan –
dard care would be superior to standard care alone with respect to the mean score for disability on the utility-weighted modified Rankin scale at 90 days, was conducted with the use of a Bayesian statistical model with adjustment for infarct vol -ume at baseline. The threshold for significance was a one-sided posterior probability of superi –
ority of at least 0.986, which was increased from 0.975 to account for the potential for enrichment and different final sample sizes. The second pri –
mary analysis, which evaluated the posterior prob –
ability that thrombectomy plus standard care would be superior to standard care alone with respect to the rate of functional independence (a score of 0, 1, or 2 on the modified Rankin scale) at 90 days, was conducted with the use of the same statistical model (with an assumption of normal distribution) and was carried out in a nested hierarchical fashion. The trial had 86% power to detect an adjusted difference between the two treatment groups in the mean score on the utility-weighted modified Rankin scale of 1.0. No additional adjustments for multiplicity were made for analyses of the secondary end points. Bayesian multiple imputations were used for patients who had missing values for the primary analyses. De –
scriptive statistics were calculated with the use of the last-observation-carried-forward method for patients who had missing values for the subgroup analyses.
Enrollment in the trial was stopped at 31
months, because the results of an interim analysis met the prespecified criterion for trial discontinu –
ation, which was a predictive probability of supe –
riority of thrombectomy of at least 95% for the first primary end point. This was the first pre –
specified interim analysis that permitted stopping for this reason, and it was based on the enrollment of 200 patients. Because enrichment thresholds had not been crossed, the analysis included the full population of patients enrolled in the trial, regardless of infarct volume. (For details about the statistical analysis, see Section S4 in the Supple –
mentary Appendix.)
Results
Patient Characteristics
From September 2014 through February 2017, a total of 206 patients were enrolled in the trial; 107 were randomly assigned to the thrombectomy group and 99 to the control group (Fig. S1 in the Supplementary Appendix). Baseline characteristics are shown in Table 1, and in Table S1 in the Sup –
plementary Appendix. At baseline, the median NIHSS score, which indicates the severity of the
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 15Thrombectomy 6 to 24 Hours after Stroke
VariableThrombectomy Group
(N = 107)Control Group
(N = 99)
Age — yr 69.4±14.1 70.7±13.2
Age ≥80 yr — no. (%) 25 (23) 29 (29)
Male sex — no. (%) 42 (39) 51 (52)
Atrial fibrillation — no. (%) 43 (40) 24 (24)
Diabetes mellitus — no. (%) 26 (24) 31 (31)
Hypertension — no. (%) 83 (78) 75 (76)
Previous ischemic stroke or transient ischemic attack — no. (%) 12 (11) 11 (11)
NIHSS score†
Median 17 17
Interquartile range 13–21 14–21
10 to 20 — no. (%) 78 (73) 72 (73)
Treatment with intravenous alteplase — no. (%) 5 (5) 13 (13)
Infarct volume — ml
Median 7.6 8.9
Interquartile range 2.0–18.0 3.0–18.1
Type of stroke onset — no. (%)‡
On awakening 67 (63) 47 (47)
Unwitnessed stroke 29 (27) 38 (38)
Witnessed stroke 11 (10) 14 (14)
Occlusion site — no. (%)§
Intracranial internal carotid artery 22 (21) 19 (19)
First segment of middle cerebral artery 83 (78) 77 (78)
Second segment of middle cerebral artery 2 (2) 3 (3)
Interval between time that patient was last known to be well
and randomization — hr
Median 12.2 13.3
Interquartile range 10.2–16.3 9.4–15.8
Range 6.1–23.5 6.5–23.9
Time from first observation of symptoms to randomization — hr
Median 4.8 5.6
Interquartile range 3.6–6.2 3.6–7.8
* Plus–minus values are means ±SD. Percentages may not sum to 100 because of rounding. There were no significant
differences between the two treatment groups with respect to the baseline characteristics, except for a history of atrial fibrillation (P = 0.01), treatment with intravenous alteplase (P = 0.04), and the onset of stroke on awakening (P = 0.03).
† Scores on the National Institutes of Health Stroke Scale (NIHSS) range from 0 to 42, with higher scores indicating a more severe deficit.
‡ A patient with the onset of stroke on awakening had last been known to be well before going to bed and had the first observation of symptoms on awakening. In a patient with an unwitnessed stroke, the time that the patient had last been known to be well and the first observation of symptoms were different and the first observation of symptoms did not occur on awakening. In a patient with a witnessed stroke, the time that the patient had last been known to be well and the first observation of symptoms were the same; all patients with a witnessed stroke had a time from first obser-vation of symptoms to randomization of more than 6 hours.
§ Patients who had occlusion of the intracranial internal carotid artery may also have had occlusion of the first segment of the middle cerebral artery.Table 1. Characteristics of the Patients at Baseline.*
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 16The new england journal of medicine
stroke deficit, was 17 in both treatment groups;
the median infarct volume was 7.6 ml in the thrombectomy group and 8.9 ml in the control group. The median interval between the time that a patient was last known to be well and random –
ization was 12.2 hours in the thrombectomy group and 13.3 hours in the control group. Base –
line characteristics were generally balanced be –
tween the two groups, except for the percentage of patients with a history of atrial fibrillation and the percentage who had the onset of stroke symptoms on awakening, which were higher in the thrombectomy group than in the control group, and the percentage of patients who received intra –
venous alteplase, which was higher in the control group than in the thrombectomy group.
Thrombectomy was performed in 105 of the
107 patients in the thrombectomy group. Ipsilat -eral carotid angioplasty was performed in 3 of the 107 patients. In 11 patients in the thrombectomy group (10%), thrombectomy was performed while the patient was under general anesthesia. In 102 of the 105 patients who underwent thrombectomy, the procedure was performed with the use of the Trevo device only; the other 3 patients underwent treatment with alternative endovascular reperfu –
sion devices after the initial treatment with the Trevo device failed, although this approach was not permitted in the protocol.
Efficacy Outcomes
The first primary end point of the mean score for disability on the utility-weighted modified Rankin scale at 90 days was 5.5 in the thrombectomy group as compared with 3.4 in the control group (adjusted difference [Bayesian analysis], 2.0 points; OutcomeThrombectomy
Group
(N = 107)Control
Group
(N = 99)Absolute
Difference
(95% CI)†Adjusted
Difference
(95% Credible
Interval)‡Posterior
Probability
of Superiority
Primary end points
Score on utility-weighted modified Rankin scale at 90 days§ 5.5±3.8 3.4±3.1 2.1 (1.2–3.1) 2.0 (1.1–3.0) >0.999
Functional independence at 90 days — no. (%)¶ 52 (49) 13 (13) 36 (24–47) 33 (21–44) >0.999
Risk Ratio
(95% CI) P Value
Secondary end pointsEarly response — no. (%)‖ 51 (48) 19 (19) 29 (16–41) 3 (2–4) <0.001**
Recanalization at 24 hr — no. (%)†† 82 (77) 39 (39) 40 (27–52) 2 (2–4) <0.001**
Change from baseline in infarct volume at 24 hr — ml†† 0.003‡‡
Median 1 13
Interquartile range 0–28 0–42
Infarct volume at 24 hour — ml†† <0.001‡‡
Median 8 22
Interquartile range 0–48 8–68
Grade of 2b or 3 on mTICI scale — no. (%)§§ 90 (84) NA
* Plus–minus values are means ±SD. CI denotes confidence interval, and NA not applicable.
† Absolute differences are reported in percentage points, except for the absolute difference in the score on the utility-weighted modified
Rankin scale, which is reported in points.
‡ Adjusted differences were estimated with the use of a Bayesian general linear model with adjustment for infarct volume at baseline.
§ The utility-weighted modified Rankin scale ranges from 0 (death) to 10 (no symptoms or disability).
¶ Functional independence was defined as a score of 0, 1, or 2 on the modified Rankin scale, which ranges from 0 to 6, with higher scores indicating more severe disability.
‖ Early response was defined as a decrease in the NIHSS score of 10 points or more from baseline or an NIHSS score of 0 or 1 on day 5, 6, or 7 of hospitalization or at discharge if it occurred before day 5.
** The P value was calculated with the use of Fisher’s exact test.
†† For details on the assessment of this end point, see Section S2 in the Supplementary Appendix.
‡‡ The P value was calculated with the use of the nonparametric Wilcoxon test.
§§ The modified Thrombolysis in Cerebral Infarction (mTICI) scale ranges from 0 to 3, with a grade of 2b or 3 indicating reperfusion of more than 50% of the affected territory.Table 2. Efficacy Outcomes.*
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 17Thrombectomy 6 to 24 Hours after Stroke
95% credible interval, 1.1 to 3.0; posterior prob –
ability of superiority, >0.999). The second primary
end point of the rate of functional independence (a score of 0, 1, or 2 on the modified Rankin scale) at 90 days was 49% in the thrombectomy group as compared with 13% in the control group (adjusted difference, 33 percentage points; 95% credible in –
terval, 21 to 44; posterior probability of superi –
ority, >0.999) ( Table 2 and Fig. 1). In post hoc
sensitivity analyses that adjusted for between-group differences in baseline characteristics that had a P value of less than 0.10, the posterior prob –
ability of superiority of thrombectomy remained significant for both coprimary end points (see Section S7 in the Supplementary Appendix).
Among the patients who underwent throm –
bectomy, immediate reperfusion was achieved in 84% according to results of central laboratory assessments and in 82% according to results of evaluations by local interventionists; the median interval between the time the patient was last known to be well and reperfusion was 13.6 hours (interquartile range, 11.3 to 18.0). Recanalization was achieved at 24 hours in 77% of the patients in the thrombectomy group and in 36% of the patients in the control group. For all the second –
ary end points, the comparisons between the two treatment groups favored thrombectomy (Table 2). In prespecified subgroup analyses, no evidence of heterogeneity of treatment effect was detected (Fig. 2); the relatively small sample size limited
the power of some of these analyses. (For details about secondary and subgroup analyses, see Figs. S2 through S8 in the Supplementary Appendix.)
Safety Outcomes
The rates of safety end points and serious adverse events — including stroke-related death at 90 days, death from any cause at 90 days, and symptomatic intracerebral hemorrhage — did not differ sig –
nificantly between the two treatment groups (Table 3, and Table S2 in the Supplementary Ap –
pendix). The rate of neurologic deterioration was lower in the thrombectomy group than in the control group (14% vs. 26%; absolute difference, −12 percentage points; 95% confidence interval, −23 to −1; P = 0.04).
Discussion
The DAWN trial showed that, among patients with stroke due to occlusion of the intracranial internal carotid artery or proximal middle cerebral artery who had last been known to be well 6 to 24 hours earlier and who had a mismatch between the se –
verity of the clinical deficit and the infarct volume, outcomes for disability and functional indepen –
dence at 90 days were better with thrombectomy plus standard medical care than with standard medical care alone. For every 2 patients who un –
derwent thrombectomy, 1 additional patient had a better score for disability at 90 days (as com –
pared with the results in the control group); for every 2.8 patients who underwent thrombectomy, 1 additional patient had functional independence Figure 1. Distribution of Scores on the Modified Rankin Scale at 90 Days.
Shown is the distribution of scores for disability on the modified Rankin
scale (which ranges from 0 to 6, with higher scores indicating more severe disability) among patients in the thrombectomy group and the control group, both in the overall intention-to-treat population (Panel A) and in subgroups defined according to time of stroke onset (Panel B). The num –
bers in the bars are percentages of patients who had each score; the per –
centages may not sum to 100 because of rounding. For the first primary end point, scores on the modified Rankin scale were weighted according to average values calculated from patient-centered and clinician-centered studies. For the second primary end point, functional independence was defined as a score of 0, 1, or 2 on the modified Rankin scale.Thrombectomy
(N=107)
Control
(N=99)
0 50 40 30 20 10 60 70 80 90 100
Percent/uni0020of/uni0020Patients/uni0020
Percent/uni0020of/uni0020Patients/uni0020Thrombectomy
(N=50)
Control
(N=46)
0 50 40 30 20 10 60 70 80 90 100BSubgroups/uni0020According/uni0020to/uni0020Time/uni0020of/uni0020Stroke/uni0020Onset
Last/uni0020Known/uni0020to/uni0020Be/uni0020Well/uni00206/uni0020to/uni002012/uni0020Hr/uni0020before/uni0020Randomization
Thrombectomy
(N=57)
Control
(N=53)
0 50 40 30 20 10 60 70 80 90 100Last/uni0020Known/uni0020to/uni0020Be/uni0020Well/uni0020>12/uni0020to/uni002024/uni0020Hr/uni0020before/uni0020RandomizationAIntention-to-Treat/uni0020Population0 1 2 3 4 5 or 6Score/uni0020on/uni0020the/uni0020Modified/uni0020Rankin/uni0020Scale
49
522
417
1613
3413
3625
714
722
618
1110
376
3330
25
423
216
2116 19
40 3221
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 18The new england journal of medicine
at 90 days (see Section S1 in the Supplementary
Appendix). The benefit of thrombectomy was con –
sistent across prespecified subgroups that were defined according to age, stroke severity, occlusion site, time to treatment, and type of stroke onset, but the power of the trial to assess differences between subgroups was limited.
Endovascular thrombectomy in patients with Figure 2. Subgroup Analyses of the First Primary End Point.
The first primary end point was the mean score for disability on the utility-weighted modified Rankin scale at 90 days. To determine the
utility-weighted score, the score on the modified Rankin scale is weighted according to average values calculated from patient-centered and clinician-centered studies. The following weights are assigned to scores 0 through 6 on the modified Rankin scale: 10.0, 9.1, 7.6, 6.5, 3.3, 0, and 0, respectively. The utility-weighted modified Rankin scale ranges from 0 (death) to 10 (no symptoms or disability). Adjusted differences were estimated with the use of a Bayesian general linear model with adjustment for infarct volume. In the forest plots, the size of the box is proportional to the sample size. The Bayesian posterior probability of heterogeneity is the probability of an interaction be –
tween the subgroup and the treatment benefit; a probability of greater than 0.975 or less than 0.025 was considered to be a significant interaction. Subgroups for mismatch between the severity of the clinical deficit and the infarct volume were defined according to the fol –
lowing criteria: patients in Group A were 80 years of age or older, had a score of 10 or higher on the National Institutes of Health Stroke Scale (NIHSS; scores range from 0 to 42, with higher scores indicating a more severe deficit), and had an infarct volume of less than 21 ml; those in Group B were younger than 80 years of age, had a score of 10 or higher on the NIHSS, and had an infarct volume of less than 31 ml; and those in Group C were younger than 80 years of age, had a score of 20 or higher on the NIHSS, and had an infarct volume of 31 to less than 51 ml. The analysis for occlusion site did not include a subgroup with occlusion of the second segment of the middle cerebral artery because of the small number of patients in that subgroup.2 0 4 6
Thrombectomy/uni0020Better Control/uni0020BetterOverall
Mismatch criteria
Group AGroup B
Group C
Sex
Male
Female
Age
<80 yr
≥80 yr
Baseline NIHSS score
10 to 17>17
Occlusion site
Intracranial internal carotid artery
First segment of the middle
cerebral artery
Type of stroke onset
On awakening
Witnessed strokeUnwitnessed stroke
Interval between time that patient was last
known to be well and randomization
6 to 12 hr
>12 to 24 hr
Time from first observation of symptoms
to randomization0 to 6 hr
>6 hrAdjusted/uni0020Difference/uni0020between/uni0020Thrombectomy
and/uni0020Control/uni0020(95%/uni0020Credible/uni0020Interval)Subgroup
−12.4 (0.8 to 3.9)2.0 (0.9 to 3.2)1.8 (0.4 to 3.4)
2.4 (1.1 to 3.6)3.0 (0.5 to 5.9)2.3 (1.0 to 3.6)
1.4 (−0.5 to 3.2)2.0 (0.9 to 3.1)3.0 (0.8 to 5.2)1.8 (0.6 to 3.1)2.3 (0.3 to 4.2)
2.4 (1.0 to 3.7)1.9 (0.8 to 2.8)2.6 (1.3 to 4.0)1.8 (0.2 to 3.2)1.8 (0.6 to 2.9)
2.5 (−0.6 to 5.5)2.3 (0.3 to 4.2)0.47
0.14
0.42
0.71
0.77
0.21
0.22
0.70>0.99
0.99
>0.99
0.95
0.99
>0.99
>0.99
0.99
>0.99
>0.99
>0.99
>0.99
>0.99
0.99
0.93
>0.99
>0.99
>0.99
>0.992.0 (1.1 to 3.0)Posterior/uni0020Probability
Benefit Heterogeneity
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 19Thrombectomy 6 to 24 Hours after Stroke
stroke is usually performed within 6 hours after
the onset of stroke. However, the rate of func –
tional independence in the thrombectomy group in our trial, in which patients received treatment 6 to 24 hours after stroke onset, was similar to the rate reported in a pooled analysis of five trials of thrombectomy, in which patients predominantly received treatment within 6 hours after stroke on –
set (49% and 46%, respectively).
7 In contrast, the
rate of functional independence in the control group in our trial was lower than the rate in the control group in the pooled analysis (13% vs. 26%). It is possible that the worse outcomes in our con –
trol group were related to the lower rate of treat –
ment with intravenous alteplase (14% in our trial vs. 88% in the pooled analysis); patients were enrolled in our trial after the accepted window of time in which intravenous thrombolytic ther –
apy is typically administered. An additional pos –
sible determinant of the worse outcomes in our control group was a higher percentage of patients with adverse prognostic features, particularly an age of 80 years or older and an NIHSS score after stroke of 10 or higher. The rates of functional independence that were observed in our control group are similar to those reported in prospective observational studies that included older patients with occlusion of a proximal large vessel who had a severe deficit and did not receive treatment with intravenous alteplase or thrombectomy.
20-22
Other recent randomized trials of thrombectomy have used enrollment criteria that are similar to those used in our trial.
23
This trial has limitations. Randomization was
stratified according to prognostic variables that the investigators determined to be most pertinent in the patient population; these variables were bal –
anced between the two treatment groups. How –
ever, there were significant differences between the two groups in other baseline variables. In post hoc sensitivity analyses that adjusted for these dif –
ferences, the benefit of thrombectomy remained.
We found that, among patients with acute
stroke who have a mismatch between the severity of the clinical deficit and the infarct volume, the safety profile for thrombectomy performed 6 to 24 hours after the onset of stroke was similar to a previously observed safety profile for thrombec-tomy performed within 6 hours after the onset of stroke
7; the rates of death and symptomatic in –
tracerebral hemorrhage did not differ significantly from the rates seen with standard medical care. OutcomeThrombectomy
Group
(N = 107)Control
Group
(N = 99)Absolute
Difference
(95% CI)Risk Ratio
(95% CI)
no. (%)percentage
points
Stroke-related death at 90 days 17 (16) 18 (18) −2 (−13 to 8) 1 (1 to 2)
Death from any cause at 90 days 20 (19) 18 (18) 1 (−10 to 11) 1 (1 to 2)
Symptomatic intracranial hemorrhage at 24 hr† 6 (6) 3 (3) 3 (−3 to 8) 2 (1 to 7)
Neurologic deterioration at 24 hr‡ 15 (14) 26 (26) −12 (−23 to −1) 1 (0 to 1)
Procedure-related complications 7 (7) NA
Distal embolization in a different territory 4 (4) NA
Intramural arterial dissection 2 (2) NA
Arterial perforation 0 NA
Access-site complications leading to intervention 1 (1) NA
* There were no significant differences between the two treatment groups with respect to safety outcomes, except for
neurologic deterioration (P = 0.04). All safety outcomes were adjudicated by an independent clinical-events committee.
† Symptomatic intracranial hemorrhage was defined according to European Cooperative Acute Stroke Study III criteria as the presence of extravascular blood in the cranium that was associated with an increase in the NIHSS score of 4 points or more or death and was judged to be the predominant cause of neurologic deterioration.
‡ Neurologic deterioration was defined as an increase in the NIHSS score of 4 or more points within 5 days after stroke that was not attributed to intracranial hemorrhage or malignant cerebral edema.Table 3. Safety Outcomes.*
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 20The new england journal of medicine
Because our trial restricted enrollment to patients
with infarcts of a small or medium volume, our findings may be concordant with previous reports that the extent of tissue injury is a determinant of the risk of symptomatic intracerebral hemorrhage after reperfusion therapy.
24
On the basis of retrospective studies, approxi –
mately one third of the patients with occlusion of a proximal anterior cerebral vessel who present within 6 to 24 hours after the onset of stroke may meet the imaging-based eligibility criteria that were used in this trial.
25,26 Further studies are
needed to establish the prevalence of patients who would be eligible for thrombectomy among the entire population of patients with ischemic stroke. Further studies are also needed to determine whether late thrombectomy has a benefit when more widely available imaging techniques are used to estimate the infarct volume at presenta –
tion, such as assessment of the extent of hypoden –
sity on non–contrast-enhanced CT.
In conclusion, we found that outcomes for
disability were better with thrombectomy plus standard medical care than with standard medi –
cal care alone among patients with acute stroke who received treatment 6 to 24 hours after they had last been known to be well and who had a mismatch between the severity of the clinical defi –
cit and the infarct volume, which was assessed with the use of diffusion-weighted MRI or perfu –
sion CT and measured with the use of automated software.
Supported by Stryker Neurovascular.Dr. Yavagal reports receiving consulting fees from and serv
–
ing on a steering committee for Medtronic and receiving con –
sulting fees from Neural Analytics; Dr. Cognard, receiving con –
sulting fees from Medtronic, Stryker, and MicroVention; Dr. Hanel, receiving consulting fees from Stryker, MicroVention, and Codman and grant support and consulting fees from Medtronic and holding stock in Neurvana, Three Rivers Medi
-cal, and InNeuroCo; Dr. Sila, receiving honoraria and adminis –
trative analysis from Medtronic; Dr. Hassan, receiving consult –
ing fees, lecture fees, and proctor fees from Medtronic and consulting fees and lecture fees from Penumbra, MicroVention, GE Healthcare, and Stryker; Dr. Levy, receiving lecture fees and honoraria for training from Covidien, consulting fees from Pul
–
sar Vasulcar, and honoraria for training from Abbott Vascular, holding stock and ownership interest in Intratech Medical, NeXtGen Biologics, and Neuravi, serving on advisory boards for Stryker, NeXtGen Biologics, MedX, and Cognition Medical, and serving as an expert witness for Renders Medical; Dr. Mitchell, receiving grant support to his institution from Stryker and Medtronic and serving as an unpaid advisory-board member for Johnson & Johnson; Dr. Chen, receiving consulting fees from Medtronic and lecture fees from Penumbra, receiving lecture fees from and serving on an advisory board for Stryker, and serving on an advisory board for Genentech; Dr. English, receiv
–
ing consulting fees from Stryker; Dr. Baxter, receiving consult –
ing fees and fees for serving on a speakers’ bureau from Penum –
bra and consulting fees from Stryker, Medtronic, Route 92 Medical, and Pulsar and holding U.S. Patent 9526863 on devices and methods for perfusion therapy, licensed to Neuronal Protec
–
tion System; Dr. Abraham, receiving consulting fees and lecture fees from Stryker and fees for serving on a speakers’ bureau and lecture fees from Boehringer Ingelheim; Dr. Veznedaroglu, re
–
ceiving consulting fees from Stryker, Trice, and Penumbra; Dr. Lopes, receiving grant support and honoraria for training and educational activities from and serving on an advisory board for Stryker; Dr. Yoo, receiving grant support from Penumbra, and Neuravi–Cerenovus and having an equity investment in Insera Therapeutics; Dr. Olivot, receiving lecture fees from Boston Scientific, Pfizer, Bristol-Myers Squibb, and Boehringer Ingel
–
heim and consulting fees from AstraZeneca and Servier; Mr. Shields, being employed by Stryker; Dr. Graves, receiving con
–
sulting fees from Stryker; Dr. Lewis, receiving fees for serving as a senior statistical scientist from Berry Consultants; Dr. Smith, receiving fees for serving as chair of a data and safety monitor
–
ing board from Stryker; Dr. Liebeskind, receiving consulting fees from Stryker and Medtronic; Dr. Saver, receiving grant sup
–
port paid to the University of California Regents and fees for serving on steering committees from Medtronic–Abbott and Neuravi; and Dr. Jovin, receiving consulting fees from and own
–
ing stock in Silk Road Medical, owning stock in and serving on an advisory board for Anaconda BioMed, Route 92 Medical, Blockade Medical, and FreeOx Biotech, and receiving consulting fees and fees for serving on a data and safety monitoring board from Codman and consulting fees and fees for serving on a steering committee from Neuravi. No other potential conflict of interest relevant to this article was reported.
Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
Appendix
The authors’ full names and academic degrees are as follows: Raul G. Nogueira, M.D., Ashutosh P. Jadhav, M.D., Ph.D., Diogo C. Haussen, M.D., Alain Bonafe, M.D., Ronald F. Budzik, M.D., Parita Bhuva, M.D., Dileep R. Yavagal, M.D., Marc Ribo, M.D., Christophe Cognard, M.D., Ricardo A. Hanel, M.D., Cathy A. Sila, M.D., Ameer E. Hassan, D.O., Monica Millan, M.D., Elad I. Levy, M.D., Peter Mitchell, M.D., Michael Chen, M.D., Joey D. English, M.D., Qaisar A. Shah, M.D., Frank L. Silver, M.D., Vitor M. Pereira, M.D., Brijesh P. Mehta, M.D., Blaise W. Baxter, M.D., Michael G. Abraham, M.D., Pedro Cardona, M.D., Erol Veznedaroglu, M.D., Frank R. Hellinger, M.D., Lei Feng, M.D., Jawad F. Kirmani, M.D., Demetrius K. Lopes, M.D., Brian T. Jankowitz, M.D., Michael R. Frankel, M.D., Vincent Costalat, M.D., Nirav A. Vora, M.D., Albert J. Yoo, M.D., Ph.D., Amer M. Malik, M.D., Anthony J. Furlan, M.D., Marta Rubiera, M.D., Amin Aghaebrahim, M.D., Jean-Marc Olivot, M.D., Wondwossen G. Tekle, M.D., Ryan Shields, M.Sc., Todd Graves, Ph.D., Roger J. Lewis, M.D., Ph.D., Wade S. Smith, M.D., Ph.D., David S. Liebeskind, M.D., Jeffrey L. Saver, M.D., and Tudor G. Jovin, M.D.
The authors’ affiliations are as follows: the Marcus Stroke and Neuroscience Center, Grady Memorial Hospital, and the Department
of Neurology, Emory University School of Medicine, Atlanta (R.G.N., D.C.H., M.R.F.); the Stroke Institute, Departments of Neurology
(A.P.J., T.G.J.) and Neurosurgery (B.T.J.), University of Pittsburgh Medical Center, Pittsburgh, and Abington Health, Abington (Q.A.S.) — both in Pennsylvania; the Department of Neuroradiology, Hôpital Gui-de-Chauliac, Montpellier (A.B., V.C.), and the Department of Diagnostic and Therapeutic Neuroradiology (C.C.) and the Neuroimaging Center and Center for Clinical Investigations (J.-M.O.), Uni
–
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.

n engl j med 378;1 nejm.org January 4, 2018 21Thrombectomy 6 to 24 Hours after Stroke
versity Hospital of Toulouse, Toulouse — both in France; OhioHealth Riverside Methodist Hospital, Columbus (R.F.B., N.A.V.), and
University Hospitals of Cleveland, Cleveland (C.A.S., A.J.F.) — both in Ohio; Texas Stroke Institute, Dallas–Fort Worth (P.B., A.J.Y.), the Department of Neuroscience, Valley Baptist Medical Center, Harlingen (A.E.H., W.G.T.), and Berry Consultants, Austin (T.G., R.J.L.) — all in Texas; the Department of Neurology and Neurosurgery, University of Miami Miller School of Medicine–Jackson Memo
–
rial Hospital, Miami (D.R.Y., A.M.M.), Baptist Health, Jacksonville (R.A.H., A.A.), Memorial Regional Hospital, Hollywood (B.P.M.), and Florida Hospital, Orlando (F.R.H.) — all in Florida; the Stroke Unit, Hospital Vall d’Hebrón (M. Ribo, M. Rubiera), and Hospital Universitari de Bellvitge (P.C.), Barcelona, and the Department of Neuroscience, Hospital Germans Trias i Pujol, Universitat Autònoma de Barcelona, Badalona (M.M.) — all in Spain; the Department of Neurosurgery, State University of New York at Buffalo, Buffalo (E.I.L.); the Department of Interventional Neuroradiology, Royal Melbourne Hospital, Victoria (P.M.); the Departments of Neurology (M.C.) and Neurosurgery (D.K.L.), Rush University Medical Center, Chicago; California Pacific Medical Center (J.D.E.) and the Depart
–
ment of Neurology, University of California, San Francisco (W.S.S.), San Francisco, the Department of Neuroradiology, Kaiser Perma –
nente (L.F.), and the Neurovascular Imaging Research Core, Department of Neurology and Comprehensive Stroke Center (D.S.L.), David Geffen School of Medicine, University of California, Los Angeles (UCLA) (D.S.L., J.L.S.), Los Angeles, Stryker Neurovascular, Fremont (R.S.), and Los Angeles County Harbor–UCLA Medical Center, Torrance (R.J.L.) — all in California; the Departments of Medical Imaging and Surgery (F.L.S., V.M.P.) and Neurology (V.M.P.), Toronto Western Hospital, University Health Network, Univer
–
sity of Toronto, Toronto; the Department of Radiology, Erlanger Hospital at the University of Tennessee, Chattanooga (B.W.B.); the Department of Neurology, University of Kansas Medical Center, Kansas City (M.G.A.); and the Neuroscience Center, Capital Health Hospital, Trenton (E.V.), and the JFK Medical Center, Edison (J.F.K.) — both in New Jersey.
References
1. Berkhemer OA, Fransen PSS, Beumer
D, et al. A randomized trial of intraarte –
rial treatment for acute ischemic stroke. N Engl J Med 2015; 372: 11-20.
2. Campbell BCV, Mitchell PJ, Kleinig TJ,
et al. Endovascular therapy for ischemic
stroke with perfusion-imaging selection. N Engl J Med 2015; 372: 1009-18.
3. Goyal M, Demchuk AM, Menon BK, et
al. Randomized assessment of rapid en –
dovascular treatment of ischemic stroke.
N Engl J Med 2015; 372: 1019-30.
4. Jovin TG, Chamorro A, Cobo E, et al.
Thrombectomy within 8 hours after symp –
tom onset in ischemic stroke. N Engl J Med
2015; 372: 2296-306.
5. Saver JL, Goyal M, Bonafe A, et al.
Stent-retriever thrombectomy after intrave –
nous t-PA vs. t-PA alone in stroke. N Engl J Med 2015; 372: 2285-95.
6. Bracard S, Ducrocq X, Mas JL, et al.
Mechanical thrombectomy after intrave –
nous alteplase versus alteplase alone after stroke (THRACE): a randomised controlled trial. Lancet Neurol 2016; 15: 1138-47.
7. Goyal M, Menon BK, van Zwam WH,
et al. Endovascular thrombectomy after
large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet 2016; 387:
1723-31.
8. Saver JL, Goyal M, van der Lugt A, et
al. Time to treatment with endovascular
thrombectomy and outcomes from ische –
mic stroke: a meta-analysis. JAMA 2016;
316: 1279-88.
9. Dávalos A, Blanco M, Pedraza S, et al.
The clinical-DWI mismatch: a new diag –
nostic approach to the brain tissue at risk of infarction. Neurology 2004; 62: 2187-92.
10. Jovin TG, Liebeskind DS, Gupta R, et
al. Imaging-based endovascular therapy
for acute ischemic stroke due to proximal intracranial anterior circulation occlusion treated beyond 8 hours from time last seen well: retrospective multicenter anal
–
ysis of 237 consecutive patients. Stroke 2011; 42: 2206-11.
11. Lansberg MG, Cereda CW, Mlynash
M, et al. Response to endovascular reper –
fusion is not time-dependent in patients with salvageable tissue. Neurology 2015;
85: 708-14.
12. Jovin TG, Saver JL, Ribo M, et al. Dif –
fusion-weighted imaging or computer –
ized tomography perfusion assessment with clinical mismatch in the triage of wake up and late presenting strokes un
–
dergoing neurointervention with Trevo (DAWN) trial methods. Int J Stroke 2017;
12: 641-52.
13. Stryker. Welcome to the DAWN trial
campus (http://dawntrial .trainingcampus
.net).
14. NETT mRS test. Ann Arbor: Univer –
sity of Michigan (https:/ / lessons .ummu
.umich .edu/ 2k/ nett_mrs/ test).
15. Patel RD, Starkman S, Hamilton S,
et al. The Rankin Focused Assessment-Ambulation: a method to score the Modi
–
fied Rankin Scale with emphasis on walk –
ing ability. J Stroke Cerebrovasc Dis 2016;
25: 2172-6.
16. Rivero-Arias O, Ouellet M, Gray A,
Wolstenholme J, Rothwell PM, Luengo-Fernandez R. Mapping the modified Rankin scale (mRS) measurement into the generic EuroQol (EQ-5D) health out
–
come. Med Decis Making 2010; 30: 341-54.
17. Hong KS, Saver JL. Quantifying the
value of stroke disability outcomes: WHO
Global Burden of Disease Project disabil –
ity weights for each level of the modified Rankin Scale. Stroke 2009; 40: 3828-33.
18. Chaisinanunkul N, Adeoye O, Lewis
RJ, et al. Adopting a patient-centered ap –
proach to primary outcome analysis of
acute stroke trials using a utility-weighted modified Rankin scale. Stroke 2015; 46:
2238-43.19. Hacke W, Kaste M, Bluhmki E, et al.
Thrombolysis with alteplase 3 to 4.5
hours after acute ischemic stroke. N Engl J Med 2008; 359: 1317-29.
20. Hernández-Pérez M, Pérez de la Ossa
N, Aleu A, et al. Natural history of acute stroke due to occlusion of the middle ce
–
rebral artery and intracranial internal ca –
rotid artery. J Neuroimaging 2014; 24: 354-
8.
21. Lima FO, Furie KL, Silva GS, et al.
Prognosis of untreated strokes due to an –
terior circulation proximal intracranial
arterial occlusions detected by use of computed tomography angiography. JAMA Neurol 2014; 71: 151-7.
22. Janardhan V, Venizelos A, Gianatasio
RM, et al. Natural history of acute stroke from large vessel proximal occlusion:

a comparison with the IMS III patients. Stroke 2014; 45: Suppl 1: AWMP8.
23. Albers GW, Lansberg MG, Kemp S, et
al. A multicenter randomized controlled trial of endovascular therapy following imaging evaluation for ischemic stroke (DEFUSE 3). Int J Stroke 2017; 12: 896-905.
24. Mishra NK, Christensen S, Wouters
A, et al. Reperfusion of very low cerebral
blood volume lesion predicts parenchy –
mal hematoma after endovascular thera –
py. Stroke 2015; 46: 1245-9.
25. Copen WA, Rezai Gharai L, Barak ER,
et al. Existence of the diffusion-perfusion mismatch within 24 hours after onset of acute stroke: dependence on proximal ar
–
terial occlusion. Radiology 2009; 250: 878-
86.
26. Rai AT, Seldon AE, Boo S, et al. A pop –
ulation-based incidence of acute large
vessel occlusions and thrombectomy eli –
gible patients indicates significant poten –
tial for growth of endovascular stroke therapy in the USA. J Neurointerv Surg 2017; 9: 722-6.
Copyright © 2017 Massachusetts Medical Society.
The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF SZEGED on October 29, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.