Article Text


Thrombolysis in patients with WAKE-UP or unknown time of stroke onset: ready for prime time?
  1. Ashutosh P Jadhav1,2,
  2. Shashvat M Desai1,
  3. Joshua A Hirsch3
  1. 1 Department of Neurology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
  2. 2 Department of Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
  3. 3 NeuroInterventional Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
  1. Correspondence to Dr Ashutosh P Jadhav, University of Pittsburgh, Stroke Institute, Pittsburgh, PA 15213, USA; jadhavap{at}

Statistics from

Endovascular therapy for emergent large vessel occlusion has proven benefit in the treatment of patients with acute ischemic stroke.1–3 This benefit is time dependent, with higher rates of good outcomes in patients treated earlier after symptom onset. The efficacy of chemical or mechanical recanalization disappears after 4.5 hours4 or 7.3 hours5 in unselected patients. Results of the recently published trials Diffusion Weighted Imaging or Computerized Tomography Perfusion Assessment with Clinical Mismatch in the Triage of Wake Up and Late Presenting Strokes Undergoing Neurointervention with Trevo (DAWN) and Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3 (DEFUSE-3)6 7 demonstrated superior outcomes in selected patients presenting beyond 6 hours of symptoms.8–11 Advanced imaging was applied to measure baseline core infarct on MRI head (diffusion weighted imaging (DWI)) or CT perfusion to identify patients with clinical or perfusion mismatch in the presence of a proximal anterior circulation large vessel occlusion. Similar efforts to identify thrombolysis candidates beyond conventional time windows have focused on the use of perfusion and/or core imaging to identify tissue at risk.12 13 No studies to date have shown the benefit of intravenous thrombolysis over medical therapy using this method14 although further efforts are ongoing.

An alternative approach using MRI relies on the observation that patients demonstrating ischemic lesions on DWI, in the absence of fluid attenuated inversion recovery (FLAIR) hyperintense signal, are likely within 4.5 hours of stroke onset. Utilizing this DWI-FLAIR mismatch to identify patients that fell within this time frame, investigators in the WAKE-UP trial then treated stroke patients with either alteplase or placebo.15 The investigators reasoned that the WAKE-UP MRI signature allows reclassification of patients presenting with ‘wake-up’ strokes or unwitnessed strokes as actually early time window (<4.5 hours) patients. The WAKE-UP trial was a multicenter, randomized, double-blind, placebo controlled clinical trial of acute ischemic stroke patients presenting with symptoms of unknown time of onset harboring the WAKE-UP MRI signature. The planned sample size of 800 patients was not reached due to lack of continued funding and so the analysis was performed after 503 patients were enrolled.

In the WAKE-UP trial, patient selection was based strictly on pretreatment MRI head and therefore a subset of patients was excluded based on MRI incompatibility or poor MRI image quality.15 Of 1362 patients screened, 64% were excluded. Of those that were excluded, 74% did not harbor the target MRI signature due to extensive FLAIR hyperintensity, extensive DWI lesion, or no DWI lesions. Most of the patients did not harbor a large vessel occlusion (80%) and planned thrombectomy was considered an exclusion criterion. Nearly all of the patients presented as ‘wake-up’ strokes with a median baseline National Institutes of Health Stroke Scale (NIHSS) score of 6. The primary efficacy endpoint was favorable clinical outcome at 90 days (modified Rankin Scale score of 0–1) and the primary safety endpoint was death or dependence at 90 days. Good outcomes occurred at a higher rate in the treatment arm (53.3% vs 41.8%, P=0.02). Interestingly, there was no significant difference in infarct volume between the two groups (3 mL vs 3.3 mL) on post-treatment MRI. In the alteplase group, there was a statistically higher rate of parenchymal hemorrhage type 2 (4% vs 0.4%, P=0.03), a trend towards higher mortality (P=0.07), and numerically more symptomatic hemorrhages (NS).

This study represents the first report of benefit with intravenous thrombolysis in patients with wake-up or symptoms of unknown onset. However, there are safety concerns raised in the WAKE-UP trial regarding mortality and hemorrhage. As the investigators point out, the trend towards higher mortality may have become significant with a larger sample size. The risk of hemorrhage may be abrogated in the future by the use of alternative lytics.16 For example, tenecteplase is a genetically modified variant of alteplase with higher fibrin specificity and lower rates of systemic hemorrhage when used in the setting of ST segment elevation myocardial infarction. A recent trial of alteplase versus tenecteplase before thrombectomy demonstrated higher rates of recanalization with tenecteplase (22% vs 10%, P=0.002) as well as higher rates of good outcomes (median modified Rankin Scale score at 90 days of 2 vs 3, cOR 1.7; P=0.04).17 The challenges of intravenous thrombolysis including the expense of alteplase and its many exclusions have been explored previously in JNIS.18–20

The implication of the WAKE-UP trial in patients meeting DAWN or DEFUSE-3 criteria is unclear. Specifically, the WAKE-UP trial excluded patients with emergent large vessel occlusion with planned mechanical thrombectomy. Furthermore, only 20% of patients enrolled in the study harbored large vessel occlusions and no benefit was demonstrated in patients with an NIHSS score of 10 or higher. Together, these findings suggest further studies are necessary to understand the utility of thrombolysis as an adjunctive treatment to mechanical thrombectomy in patients presenting with large vessel occlusion after 4.5 hours of symptom onset.

Much like the late time window thrombectomy trials, patient selection in the WAKE-UP trial relied on advanced imaging. The use of MRI in patient selection introduces potential logistical and other challenges. Future trials might incorporate CT based selection criteria. In the case of DAWN and DEFUSE-3, the pathophysiological rationale for patient selection and treatment was to use advanced imaging to calculate ischemic core in patients presenting with documented proximal large vessel occlusion. Recanalization was then pursued if salvageable brain tissue was felt to be at risk based on clinical or radiographic mismatch. Patients with viable tissue at late time windows were thought to have slow infarct growth (‘slow’ progressors) likely related to favorable collaterals in addition to other tissue preserving mechanisms.21 22 As such, follow-up infarct was smaller in patients who underwent thrombectomy compared with those who were managed medically.

In contrast, what is the pathophysiological mechanism by which thrombolysis leads to improved outcomes in the WAKE-UP trial? Patient selection was not predicated on identifying tissue at risk or demonstrating a large vessel occlusion. Several deliberations should be made when interpreting these results. First, the assumption that the WAKE-UP MRI signature reflects ischemia occurring within 4.5 hours of symptoms onset is based on patients with known time of symptoms onset. The DWI to FLAIR ratio in patients with known symptom onset beyond 4.5 hours is less well understood and it is less clear whether this signature can be universally applied. In fact, in the DAWN and DEFUSE-3 trials, it was observed that a subset of patients presenting at late time windows indeed had no hyperintensity on DWI or FLAIR. Such a patient clearly has salvageable brain tissue based on clinical/perfusion mismatch and would benefit from reperfusion therapy but would not meet the WAKE-UP MRI signature criteria.13 Second, the authors argue that they are selecting for patients that are presenting within 4.5 hours of symptoms based on MRI and yet treatment is initiated 25 min after MRI completion, suggesting that thrombolysis was delivered closer to 5 hours of symptom onset. Third, as the authors indicate, the outcomes observed are more robust than expected in a 0–4.5 hour treatment time window patient and more akin to a 0–3 hour treatment time window patient. This incongruency in outcomes suggests that the WAKE-UP MRI signature may not be truly identifying the proposed 0–4.5 hour time window population. Fourth, there was no significant difference in infarct growth rate between the treatment and placebo arms (as would be expected if reperfusion was achieved and tissue was rescued with alteplase). This raises further skepticism about the mechanistic underpinnings of outcomes differences. Fifth, no data are provided on recanalization rates in the large vessel occlusions and so it is unclear whether the therapeutic goal of reperfusion was achieved. Sixth, no data are provided on the location of the infarct in the treatment and placebo groups. This may be a relevant confounder as, for example, the placebo group may be enriched in patients with infarcts in eloquent territory compared with the treatment arm. Finally, while ‘time’ may be a physiological surrogate for tissue at risk, time alone does not ensure that thrombolysis will lead to improved outcomes, as evidenced by the recently presented PRISMS (A Study of the Efficacy and Safety of Alteplase in Participants With Mild Stroke), in which patients with a low NIHSS score did not experience improved outcomes when treated with alteplase versus placebo in the 0–3 hour time window.23

Triage of stroke with advanced imaging appears to move the discussion beyond the blunt instrument currently used—that is, time since onset.24 Emerging and new drugs may improve efforts at chemical thrombolysis. The WAKE-UP trial provides promising support for thrombolysis in the treatment of patients with ‘wake-up’ strokes or unknown time of symptom onset. While further confirmatory trials need to occur, the future continues to get brighter for ischemic stroke patients.


View Abstract


  • Contributors All authors contributed to the drafting and final approval of the manuscript.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.