A 5-Item Scale to Predict Stroke Outcome After Cortical Middle Cerebral Artery Territory Infarction
Validation From Results of the Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution (DEFUSE) Study
Background and Purpose—Various clinical, laboratory, and radiographic parameters have been identified as predictors of outcome for ischemic stroke. The purpose of this study was to combine these parameters into a validated scale for outcome prognostication in patients with a middle cerebral artery territory infarction.
Methods—We retrospectively reviewed 129 patients over a 2-year period and considered demographic, clinical, laboratory, and radiographic parameters as potential predictors of outcome. Inclusion criteria were unilateral hemispheric infarcts within the middle cerebral artery territory >15 mm in diameter. Our primary outcome measure was a favorable recovery defined as a modified Rankin Score was ≤2 at 30 days. A multivariable model was used to determine independent predictors of outcome and weighted to create a 5-item scale to predict stroke recovery. External validation of this model was done using data from the Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution (DEFUSE) study.
Results—The 5 independent predictors of outcome were as follows: age (OR, 1.09; 95% CI, 1.03 to 1.14; P=0.001), National Institutes of Health Stroke Scale score (OR, 1.17; 95% CI, 1.06 to 1.30; P=0.003), infarct volume (OR, 1.01; 95% CI, 1.00 to 1.02; P=0.03), admission white blood cell count (8.5×103/mm3; OR, 1.16; 95% CI, 1.03 to 1.27; P=0.04), and presence of hyperglycemia (OR, 4.2; 95% CI, 1.1 to 16.4; P=0.04). Combining these variables into a point scale significantly improved prediction over the individual variables accounted alone as evidenced by the area underneath the receiver operating curve (OR, 0.91; 95% CI, 0.87 to 0.96; P=0.0001). When applied to the DEFUSE study population for validation, the model achieved a sensitivity of 83% and specificity of 86%.
Conclusions—With validation from a prospective study of similar patients, this model serves as a useful clinical and research tool to predict stroke recovery after cortical middle cerebral artery territory infarction.
- Received July 8, 2010.
- Revision received September 27, 2010.
- Accepted October 18, 2010.
- © 2011 American Heart Association, Inc.