Objective: To examine the effect of serial neuroimaging studies on the diagnosis, therapy, and outcome of patients with acute stroke.
Design: Retrospective case series.
Setting: Tertiary care teaching hospital.
Patients: 206 adult patients (mean age ±SD, 66.0 ± 10.8 years) hospitalized with a diagnosis of acute stroke between 1990 and 1993.
Measurements: Strokes were retrospectively assigned to five categories (large-vessel, small-vessel, cardioembolic, other, or unknown) using standardized criteria based on the history, physical examination, ancillary test results, and first computed tomographic (CT) or magnetic resonance imaging (MRI) study of the head. Strokes were reclassified after the results of further neuroimaging studies, if any, were reviewed. The type and timing of therapy and the patient outcome at hospital discharge were documented.
Results: The additional studies changed stroke classification in only 20.0% of the 140 patients who had two or more neuroimaging studies. All classification changes were from the unknown cause category to a category with a specific cause. In most patients receiving treatment (93.2%), therapy began before an additional CT or MRI study was obtained. In patients who had one neuroimaging study, 70.1% went home, 24.0% went to a skilled nursing facility, and 5.9% died; the corresponding percentages in persons who had multiple studies were 73.3%, 24.4%, and 2.2% (P > 0.1).
Conclusions: Serial neuroimaging studies did not alter the classification of strokes for which an initial diagnosis had already been made. However, they were useful in determining the cause of strokes initially classified as having an unknown cause. Therapy was almost always begun immediately after the first CT or MRI study was obtained. Outcome at hospital discharge was not significantly related to the number of neuroimaging studies obtained.