Introduction: The use of MRI may alter the target population for intravenous recombinant tissue plasminogen activator (rtPA) treatment relative to conventional CT. If selection changes, it remains crucial to demonstrate safety and efficacy of rtPA for the overall population, as well as in subpopulations hypothesized to benefit from MRI.
Materials and methods: Clinical outcome and incidence of symptomatic intracerebral hemorrhage (ICH) was recorded in 112 consecutive patients treated with intravenous rtPA (0-3 h) with MRI as first-choice imaging modality. According to the responder analysis, favorable outcome was separately defined for mild (NIHSS <8; n = 51), moderate (NIHSS 8-14; n = 30) and severe (NIHSS >14; n = 31) stroke.
Results: Eighty-three patients were treated with rtPA after MRI, and 29 after CT. Adjusted for baseline severity, 42% of all patients had a favorable outcome, compared to 37% in NINDS. Among patients with severe stroke, MR-selected patients showed a good outcome in 52% of patients compared to 29% in NINDS (p < 0.05). Symptomatic ICH occurred in 2 patients (1.9 %), and 7 patients died during hospitalization (6.3%).
Conclusion: MRI-based rtPA is safe and time-efficient. Outcome data compares well with NINDS data. Diagnostic information obtained from multimodal MRI may affect the target group. Our data support the hypothesized benefit of MRI in patients with severe stroke.