BACKGROUND: Door-to-needle time of 20 minutes to stroke patients with intravenous tissue plasminogen activator (iv-tPA) is feasible when computed tomography (CT) is used as first-line of brain imaging. Magnetic resonance imaging (MRI)-based assessment is more time-consuming but superior in detecting acute ischemia. The certainty with which stroke physicians prescribe or refrain from giving iv-tPA treatment to CT- versus MRI-examined patients has not previously been studied. The aim of the present study was to determine the effect of a primary imaging strategy of CT or MRI on clinicians' certainty to prescribe or refrain from giving iv-tPA to patients with suspected acute stroke.

METHOD: Consecutive patients with suspected stroke were quasi-randomized to either CT- or MRI-based assessment before potential iv-tPA treatment. The influence of (1) the clinical findings and (2) the image findings, and (3) the certainty with which the stroke physician prescribed or refrained from giving iv-tPA treatment were assessed with visual analog scales (VAS). Predictors of treatment certainty were identified with a random-effect model.

RESULTS: Four-hundred forty-four consecutive patients were quasi-randomized. MRI influenced the final treatment decision more than CT (P = .002). Compared with CT-examined patients (mean VAS score 8.6, SD ±1.6) stroke physicians were significantly more certain when prescribing or refraining from giving iv-tPA to MRI-examined patients (mean VAS score 9.0, SD ±1.2) (P = .014). No differences in modified Rankin scale or mortality were detected at 3 months in CT- versus MRI-examined iv-tPA-treated patients.

CONCLUSIONS: Stroke physicians were significantly more certain when prescribing iv-tPA to MRI-examined stroke patients, and MRI influences the final treatment decision significantly more compared with CT, although no difference in mortality and functional outcome at 3 months was detected between CT- and MRI-examined patients treated with iv-tPA.