Impact of CT Stroke Window Settings on Acute Stroke Detection.

Non-contrast CT is the most important imaging modality in the evaluation of suspected acute stroke by excluding intracranial haemorrhage and directly visualizing early ischemic changes. These changes are challenging to detect on non-contrast CT due to the small reduction in the attenuation value of ischemic tissue from normal. The study’s objective was to assess the use of stroke window settings for improving the detection of acute stroke. This retrospective study included forty-nine patients in whom non-contrast CT were performed for suspected acute stroke within 24 hours from symptom onset. Images were reviewed in two reading sessions with different window settings used: default brain window (80/40 [window width/window level]) and stroke window (40/40 [window width/window level]). Both windows were evaluated for their ability to detect early ischemic changes with the final diagnosis as the reference standard. Twenty-nine patients had a final diagnosis of acute stroke. The sensitivity and specificity of non-contrast CT for acute stroke detection were 79.3% and 100% respectively at the default brain window. Both windows were comparable for detecting acute stroke (P=0.2). The CT sensitivity increased to 86.2% after adding the stroke window review to the default brain window. The resultant improvement in CT diagnostic performance by stroke window review was not statistically significant (P=0.5). Conclusion, the superior sensitivity of applying stroke window settings after the default window review is small with modern generation CT scanner. These findings should increase the confidence in routine radiology reporting that uses the standard brain window in the assessment of acute stroke.