Defining acute ischemic stroke tissue pathophysiology with whole brain CT perfusion

Area of Science:

• Neurology
• Radiology
• Medical Imaging

Background:

• Acute stroke management relies on accurate identification of ischemic core and penumbra.
• Perfusion computed tomography (CTP) is crucial for assessing brain tissue viability.
• Existing CTP thresholds require validation and refinement for optimal clinical use.

Purpose of the Study:

• To identify and validate whole brain CTP thresholds for ischemic core and penumbra in acute stroke.
• To develop a probability-based model for enhanced accuracy in tissue pathophysiology measurements.
• To compare the performance of single thresholds versus a probability model.

Main Methods:

• 183 acute stroke patients underwent multimodal stroke CT and 24-hour MRI (DWI, PWI).
• Pixel-based receiver operator curve analysis determined optimal CTP parameter thresholds using coregistered acute CTP and 24-hour DWI.
• A sigma curve fitted probability-based model was developed incorporating multiple perfusion parameters.

Main Results:

• Time to peak (TTP) +5s best identified penumbra (AUC 0.79), while cerebral blood flow (CBF) <50% best identified the ischemic core (AUC 0.73).
• The probability model demonstrated higher accuracy for ischemic core (AUC 0.80) and penumbra (AUC 0.85).
• The probability model's volume measurements were significantly closer to reference DWI findings (P=0.031).

Conclusions:

• Whole brain CTP can accurately delineate penumbra and ischemic core with thresholds comparable to previous studies.
• A novel probability-based model offers improved accuracy in defining the ischemic core and penumbra compared to single thresholds.
• This probability model enhances the precision of CTP in acute stroke assessment.