Angiographic outcome of endovascular stroke therapy correlated with MR findings, infarct growth, and clinical outcome in the DEFUSE 2 trial


  • Conflict of interest: G. Albers received consulting fees and expenses from Lundbeck for Steering Committee work and consulting fees from Concentric for serving on a DSMB. G. Albers and R. Bammer are equity shareholders in iSchemaView. G. Zaharchuk receives modest research funding support from GE Healthcare.
  • Funding sources: The National Institute for Neurological Disorders and Stroke (NINDS). R01 NS03932505.



DEFUSE 2 demonstrated that patients with magnetic resonance imaging mismatch had a favorable clinical response to tissue reperfusion assessed by magnetic resonance imaging. This study reports the endovascular results and correlates angiographic reperfusion with clinical and imaging outcomes.


Prospectively enrolled ischemic stroke patients underwent baseline magnetic resonance imaging and started endovascular therapy within 12 h of onset. Patients were classified as either target mismatch or no target mismatch using magnetic resonance imaging. The pre- and postprocedure angiogram was evaluated to determine thrombolysis in cerebral infarction scores. Favorable clinical response was determined at day 30, and good functional outcome was defined as a modified Rankin Scale 0–2 at day 90.


One-hundred patients had attempted endovascular treatment. At procedure end, 23% were thrombolysis in cerebral infarction 0–1, 31% thrombolysis in cerebral infarction 2A, 28% thrombolysis in cerebral infarction 2B, and 18% thrombolysis in cerebral infarction 3. More favorable thrombolysis in cerebral infarction-reperfusion scores were associated with greater magnetic resonance imaging reperfusion (P < 0·001). thrombolysis in cerebral infarction scores correlated with 30-day favorable clinical response (P = 0·041) and 90-day modified Rankin Scale 0–2 (P = 0·008). These correlations were significant for target mismatch patients at 30 days (P = 0·034) and 90 days (P = 0·003). Infarct growth was strongly associated with poorer thrombolysis in cerebral infarction scores in target mismatch patients (P < 0·001). Patients with thrombolysis in cerebral infarctionnfarction 2A reperfusion had less magnetic resonance imaging reperfusion (P = 0·004) and poorer clinical outcome at 90 days (P = 0·01) compared with thrombolysis in cerebral infarction 2B-3 patients.


Thrombolysis in cerebral infarction reperfusion following endovascular therapy for ischemic stroke is highly correlated with magnetic resonance imaging reperfusion, infarct growth, and clinical outcome.