Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

DWI/FLAIR mismatch MRI to determine stroke age in wake-up strokes for tPA consideration

Three Part Question

In [patients who present with neurological deficits upon waking] can [DWI/FLAIR mismatch be used] to [determine time of stroke onset for tPA therapy]?

Clinical Scenario

A 49-year-old female is brought to the emergency department via ambulance with left-sided facial droop, right tongue deviation, reduced sensation on her left side with pronator drift. Her symptoms were first noted shortly after waking; Can DWI/FLAIR mismatch MRI be used to identify the time of stroke onset for potential tPA treatment?

Search Strategy

Pubmed Database 1966-4/2017
[tPA] AND [wake up stroke] AND [MRI]
[thrombolysis] AND [wake up stroke] AND [MRI]
[Alteplase] AND [wake up stroke] AND [MRI]

Search Outcome

Pubmed: 41 articles were identified, of which 3 were retained.
[thrombolysis] AND [wake up stroke] AND [MRI]: 31 articles / 3 unique used
[tPA] AND [wake up stroke] AND [MRI]: 7 articles / 0 unique used

Relevant Papers
Trial Name Author, date and country Patient group Study type and level of evidence Outcomes Key results Study Weakness
WAKE-UP1,2 Thomalla et al., 2014, Germany Acute ischemic stroke. MRI Confirmed. Unknown onset. Randomized, Double blinded, placebo controlled, parallel assignment, multicenter, clinical trial NIHSS, MR, BI, GOS, BDI, EQ-5D
-22-36 hrs.
-5-9 days
-90 days
Compare tPA & placebo N/A -Only 90-day f/u. No long-term outcomes. Bleeding risk up to 6 months out.

-MRI takes up to 60 minutes (waste time?)

Less hospitals have MRI compared to CT
Thomalla et al., 2017, Germany
DWI/FLAIR mismatch in 48% of participants
THAWS4 Koga et al., 2014, Japan Acute ischemic stroke. unknown onset time. Multicenter, Prospective, Randomized, Open Treatment, Blinded end-point. clinical trial. Modified Rankin scale
Compare tPA & placebo N/A

[Alteplase] AND [wake up stroke] AND [MRI]: 15 articles / 0 unique used


There are currently two well-designed, clinical trials underway that include multiple medical centers 1) throughout Europe and 2) throughout Japan that are designed to provide large sample sizes and high quality data about the efficacy of DWI/FLAIR mismatch MRI for guiding the use of tPA in WUS patients2,4. Both studies have essentially the same study design. DWI/FLAIR mismatch to estimate stroke onset time has a sensitivity estimated at 62-83%, and a specificity of 71-85% 5,6. Final results from these studies have not yet been published, however the WAKE-UP trial published preliminary results which indicated that near half of wake-up strokes1.

Clinical Bottom Line

There is currently limited published evidence detailing the benefit of using DWI/FLAIR mismatch to determine the time of onset in wake-up stroke patients for consideration of therapy. There are however, two large, well designed trials underway to test the efficacy and safety this technology. Pending final results of these studies, patients presenting with wake-up strokes should continue to be managed with clot retrieval methods following hospital protocol.

Level of Evidence

Level 2 - Studies considered were neither 1 or 3.