The Lancet Neurology OCTOBER 01, 2012

MRI profile and response to endovascular reperfusion after stroke (DEFUSE 2): a prospective cohort study

Maarten G. Lansberg, Matus Straka, Stephanie Kemp, Michael Mlynash, et al. (DEFUSE 2 Study Investigators)

Bottom Line

The DEFUSE 2 study demonstrated that patients with an acute ischemic stroke and a 'Target Mismatch' MRI profile (indicative of salvageable tissue) derived significant clinical and radiographic benefits from successful endovascular reperfusion, whereas those without this profile did not.

Key Findings

1. In patients with the Target Mismatch profile, early reperfusion was strongly associated with a favorable clinical response (adjusted OR 8.8, 95% CI 2.7–29.0; p=0.003 for difference in ORs compared to No Target Mismatch).

2. Target Mismatch patients who achieved early reperfusion had a significantly higher likelihood of good functional outcome at 90 days (OR 5.2, 95% CI 1.4–19.0).

3. Successful early reperfusion in the Target Mismatch group significantly attenuated infarct growth by day 5, with a median infarct growth of 30 mL versus 73 mL in those without reperfusion (p=0.01).

4. Patients lacking the Target Mismatch profile showed no statistically significant association between reperfusion and favorable clinical outcomes (adjusted OR 0.2, 95% CI 0.0–1.6).

Study Design

Design

Prospective Cohort Study

N/A

Sample

138

Patients

Duration

90 days

Median

Setting

Multicenter, US and Austria

Population Consecutive patients with acute ischemic stroke and proximal cerebral artery occlusion scheduled for endovascular therapy within 12 hours of symptom onset.

Intervention Endovascular revascularization therapy.

Comparator Patients were stratified by baseline MRI profile (Target Mismatch vs. No Target Mismatch) and assessed based on the achievement of early reperfusion (defined as >50% reduction in perfusion-weighted MRI lesion volume).

Outcome Favorable clinical response, defined as an improvement of 8 or more points on the NIH Stroke Scale (NIHSS) between baseline and day 30, or an NIHSS score of 0–1 at 30 days.

Study Limitations

• The study was an observational prospective cohort rather than a randomized controlled trial, which precludes definitive conclusions regarding the efficacy of endovascular therapy compared to medical management.

• The sample size of the 'No Target Mismatch' subgroup was relatively small, limiting the power to characterize the response to reperfusion in that specific population.

• The study findings apply to a highly selected population undergoing specific MRI-based screening, which may limit generalizability to broader stroke populations.

• Infarct volume assessment at 5 days may be influenced by edema, potentially affecting the precision of infarct growth calculations.

Clinical Significance

DEFUSE 2 provided the foundational imaging framework for patient selection in modern stroke care. By identifying the 'Target Mismatch' profile, it validated the concept that neuroimaging can identify salvageable brain tissue (the penumbra), which subsequently became a cornerstone for successful randomized trials of endovascular therapy in the extended time window.

Historical Context

At the time of this study, the utility of endovascular therapy for acute stroke was debated due to mixed results in early trials. DEFUSE 2 followed the initial DEFUSE and EPITHET studies, further refining the imaging-based selection criteria and reinforcing the crucial link between successful mechanical reperfusion and improved clinical outcomes in patients with viable at-risk tissue.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

Explain the physiological significance of the mismatch between Diffusion-Weighted Imaging (DWI) and Perfusion-Weighted Imaging (PWI) and how it relates to the concept of the 'ischemic penumbra' studied in DEFUSE 2.

Key Response

DWI identifies areas of cytotoxic edema (irreversibly damaged tissue), while PWI identifies areas of hypoperfusion. A PWI lesion significantly larger than the DWI lesion suggests a 'penumbra'—tissue that is ischemic but still viable. DEFUSE 2 used this 'Target Mismatch' to identify patients likely to benefit from reperfusion, as these patients have salvageable tissue that has not yet infarcted.

Resident

A patient presents 7 hours after stroke onset. Based on the DEFUSE 2 findings, why might you prioritize an MRI with perfusion (or CTP) over simple non-contrast CT even if the patient is outside the traditional 4.5-hour IV tPA window?

Key Response

DEFUSE 2 demonstrated that clinical benefit from endovascular reperfusion is more closely tied to the presence of a 'Target Mismatch' profile (salvageable tissue) than to the strict time from symptom onset. It provided the evidence base for treating patients in an extended window if they have a favorable imaging profile, showing that those without the mismatch do not benefit and may even be harmed by late reperfusion.

Fellow

The DEFUSE 2 'Target Mismatch' profile specifically used a Tmax threshold of >6 seconds to define the volume of hypoperfused tissue. What are the clinical implications of using this specific hemodynamic threshold compared to a more liberal Tmax >4s or a more restrictive Tmax >10s?

Key Response

Tmax >6s has been identified as the optimal threshold for identifying tissue at risk of infarction without reperfusion. A more liberal threshold (Tmax >4s) might over-calculate the penumbra by including areas of benign oligaemia that would not infarct anyway, while a more restrictive threshold (Tmax >10s) might underestimate the salvageable tissue, potentially denying treatment to patients who could still benefit.

Attending

Reflecting on the evolution of stroke trials, how did the observational nature of DEFUSE 2 resolve the apparent contradictions seen in earlier trials like IMS III, and how does it change your approach to the 'malignant' DWI profile?

Key Response

Earlier trials like IMS III failed largely due to poor patient selection based on time rather than imaging. DEFUSE 2 showed that patients with a 'malignant profile' (large baseline DWI lesions) have high rates of symptomatic intracranial hemorrhage and poor outcomes regardless of reperfusion. This teaches us that success in thrombectomy depends as much on 'who not to treat' (excluding those with large core/no mismatch) as it does on 'who to treat'.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

Evaluate the use of 'infarct growth' as a surrogate endpoint in DEFUSE 2. To what extent does this imaging biomarker meet the criteria for a surrogate endpoint that reliably predicts long-term functional outcome (mRS 0-2)?

Key Response

While DEFUSE 2 showed a strong correlation between successful reperfusion, reduced infarct growth, and improved mRS, imaging surrogates can be problematic. Infarct growth is a continuous biological variable that captures the mechanism of the intervention (salvaging tissue), but it doesn't account for 'eloquent' vs 'non-eloquent' brain regions, meaning small amounts of growth in critical areas can lead to disproportionate clinical disability.

Journal Editor

As a reviewer, how would you address the potential for selection bias in DEFUSE 2, given that the decision to undergo endovascular therapy was not randomized but left to the treating physician's discretion?

Key Response

The lack of randomization is a significant threat to internal validity; patients who clinicians felt were 'too good' or 'too bad' might have been excluded, potentially inflating the perceived benefit of the 'Target Mismatch' profile. However, the study's strength lies in its prospective, multicenter design and central blinded core-lab imaging analysis, which provided the necessary proof-of-concept to justify subsequent landmark RCTs like DEFUSE 3.

Guideline Committee

How did the results of DEFUSE 2 influence the shift in AHA/ASA guidelines from a 'time-is-brain' paradigm to a 'tissue-is-brain' paradigm for mechanical thrombectomy in the 6-24 hour window?

Key Response

DEFUSE 2 provided the foundational evidence that the 'Target Mismatch' profile (Tmax >6s / DWI core ratio) could identify responders in late windows. This led directly to the DEFUSE 3 trial, which resulted in the AHA/ASA Level I, Evidence A recommendation for mechanical thrombectomy in patients 6-16 hours from last known well who meet specific MRI/CTP criteria, effectively moving clinical practice away from rigid time cutoffs.

Clinical Landscape

Noteworthy Related Trials

2015

MR CLEAN Trial

n = 500 · NEJM

Tested

Intra-arterial treatment (mechanical thrombectomy)

Population

Patients with acute ischemic stroke and occlusion of the proximal anterior circulation

Comparator

Usual care alone

Endpoint

Score on the modified Rankin scale at 90 days

Key result: Intra-arterial treatment was effective and safe for patients with acute ischemic stroke caused by a proximal intracranial arterial occlusion in the anterior circulation.

2015

SWIFT PRIME Trial

n = 196 · NEJM

Tested

Solitaire stent retriever thrombectomy plus IV t-PA

Population

Patients with acute ischemic stroke and large-vessel occlusion

Comparator

IV t-PA alone

Endpoint

Functional independence (mRS 0-2) at 90 days

Key result: The addition of stent-retriever thrombectomy to IV t-PA resulted in better functional outcomes and higher rates of revascularization.

2018

DEFUSE 3 Trial

n = 182 · NEJM

Tested

Endovascular therapy between 6 and 16 hours after stroke onset

Population

Patients with acute ischemic stroke due to large-vessel occlusion in the anterior circulation

Comparator

Standard medical therapy

Endpoint

Ordinal score on the modified Rankin scale at 90 days

Key result: Endovascular thrombectomy plus standard medical therapy resulted in better functional outcomes than standard medical therapy alone in patients with a stroke caused by large-vessel occlusion.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis