- Home>
- Specialty Care>
- Neurology>
- Summary and Comment
Predicting Stroke Outcomes from CT Perfusion Data
Applying the Alberta Stroke Program Early Computed Tomography Score to these data may provide useful prognostic information.
Parsons and colleagues applied the Alberta Stroke Program Early Computed Tomography Score (ASPECTS) to CT perfusion (CTP) data to predict final infarct extent and clinical stroke outcome. ASPECTS is derived by scoring 10 predefined regions in the MCA territory as 1 (normal) or 0 (abnormal) and calculating the sum (Am J Neuroradiol 2001; 22:1534). A low ASPECTS suggests involvement of a large fraction of the MCA territory and correlates with worse clinical outcomes.
The researchers prospectively enrolled 37 consecutive patients presenting within 6 hours after suspected anterior-circulation (primarily MCA) infarction. ASPECTS was computed from the baseline noncontrast CT (NCCT); from the CTP source images at the peak of contrast arrival (CTP-SI); from the CT angiogram source images (CTA-SI); and from the cerebral blood volume (CBV), flow (CBF), and mean transit time (MTT) parametric maps. Follow-up ASPECTS was measured by diffusion-weighted imaging (DWI) at day 3, and clinical outcome was assessed at 90 days.
In the 20 patients who achieved major reperfusion, the mean baseline CBV and CTP-SI ASPECTS did not differ significantly from follow-up ASPECTS; the extent of the final infarct was underestimated by baseline CTA-SI and NCCT ASPECTS and was overestimated by baseline CBF and MTT ASPECTS. For the 17 patients without major reperfusion, final infarct ASPECTS was similar to baseline CBF and MTT ASPECTS, whereas baseline ASPECTS for NCCT, CTA-SI, CTP-SI, and CBV all underestimated the final infarct extent. Approximately 65% of patients with CTP-SI or CBV ASPECTS >6 at presentation had a favorable outcome (modified Rankin scale score 0–2), compared with about 14% for ASPECTS 
6.
Comment: Critical treatment decisions in acute stroke management depend on accurately assessing damaged tissue versus tissue at risk. CTA and CTP can be obtained rapidly, are widely available, and are becoming the studies of choice for evaluating acute stroke in many centers. Several studies have shown that early parametric CT-CBV maps and CTA-SI can be used to predict the minimal extent of final tissue damage (Stroke 2001; 32:2021, Stroke 2004; 35:2472, Stroke 2002; 33:959, and Ann Neurol 2002; 51:417) and that MTT and CBF maps help outline the territory at risk for infarction. The current findings suggest that CTP-SI ASPECTS, like CBV ASPECTS, is a more accurate early predictor of the minimal extent of irreversible damage than either NCCT or CTA-SI.
The application of these results to the management of individual patients should be tempered by the need for researchers to more accurately quantify the relation between the degree of CT signal change and the probability of underlying tissue damage (Stroke 2001; 32:2021, and Ann Neurol 2002; 51:417). Further studies are necessary to pinpoint the sensitivity and specificity of these measures and the extent to which their prognostic value depends on time elapsed from infarct onset.
— Stelios Smirnakis, MD, PhD
Dr. Smirnakis is Associate Neurologist, Brigham and Women’s Hospital; Assistant in Neurology, Massachusetts General Hospital; and Instructor, Harvard Medical School, Boston.
Published in Journal Watch Neurology March 23, 2006
Citation(s):
Parsons MW et al. Perfusion computed tomography: Prediction of final infarct extent and stroke outcome. Ann Neurol 2005 Nov; 58:672-9.
- Medline abstract (Free)