Abstract WP86: MRI-Guided Mismatch Imaging to Identify Patients for Thrombolysis in the Unknown Time Window: Insights From the WAKE-UP Trial

Abstract

Background: Two imaging paradigms, the diffusion-weighted imaging (DWI) - fluid attenuated inversion recovery (FLAIR) and perfusion-weighted imaging (PWI) - DWI mismatch, used in the Efficacy and Safety of MRI-Based Thrombolysis in Wake-Up Stroke (WAKE-UP) and Extending the Time for Thrombolysis in Emergency Neurological Deficits (EXTEND) trial respectively, can identify patients eligible for thrombolysis if stroke onset time is unknown. Objectives: We explored presence and influence of the PWI-DWI mismatch on outcome in WAKE-UP. Methods: We included patients screened for DWI-FLAIR mismatch in WAKE-UP who also underwent PWI. We defined PWI-DWI mismatch according to EXTEND as core <70 ml, mismatch ratio >1.2 and mismatch volume >10 ml. Primary efficacy end point was favorable outcome defined as modified Rankin Scale score of 0-1, adjusted for age and symptom severity. Results: The analysis included 343 screened patients of which 162 were randomized and treated. Of 343 screened patients, 162 had a DWI-FLAIR mismatch, 80 had a PWI-DWI mismatch and of those, 36 had both mismatches. Proportions of PWI-DWI mismatch did not differ in those with (22%, 36/162) vs without (24%, 44/181) DWI-FLAIR mismatch (p=0.74). PWI-DWI mismatch status did not modify treatment effect of thrombolysis (p for interaction=.68). In patients with both DWI-FLAIR and PWI-DWI mismatch, favorable outcome was present in 52% of those treated with alteplase (11/21) vs 38% (5/13) in those receiving placebo (adjusted OR 2.02; 95% CI 0.44-9.24, p=0.36). Selection based on the presence of the DWI-FLAIR mismatch identified more thrombolysis eligible patients (47%; 95% CI 42%-53%) compared to the PWI-DWI mismatch (23%; 95% CI 19%-28%, p<.0001). Screening for either one of the mismatch profiles resulted in a yield of 60% (95% CI 55%-65%) candidates for thrombolysis. Conclusion: We did not identify an association between the DWI-FLAIR and PWI-DWI mismatch. PWI-DWI mismatch status did not modify treatment effect in patients with DWI-FLAIR mismatch, but the analysis was underpowered. Screening for the presence of either one of the mismatch profiles with MRI seems the most inclusive approach to identify patients who can still benefit from thrombolysis in the unknown time window after stroke onset.