Automated ASPECTS for multi-modality CT predict infarct extent and outcome in large-vessel occlusion stroke

Area of Science:

• Neuroimaging
• Stroke Medicine
• Radiology

Background:

• Large-vessel occlusion (LVO) stroke requires accurate assessment of infarct extent for effective treatment.
• The Alberta Stroke Program Early CT Score (ASPECTS) is a standard tool, but its application with various CT modalities needs further evaluation.

Purpose of the Study:

• To evaluate the predictive value of automated Alberta Stroke Program Early CT Score (ASPECTS) using different CT angiography (CTA) phases for infarct extent and clinical outcomes in LVO stroke.
• To compare the performance of non-contrast CT, CTA-arterial, CTA-venous, and mismatch-ASPECTS in predicting stroke outcomes.

Main Methods:

• Retrospective analysis of 135 patients with LVO stroke who received reperfusion therapy.
• Correlation analysis using Spearman rank coefficient between different CT-ASPECTS modalities and follow-up scores/outcomes.
• Receiver operating characteristic (ROC) curve analysis to assess the predictive accuracy of CT-ASPECTS for good clinical outcomes.

Main Results:

• CTA-venous-ASPECTS demonstrated an almost-perfect correlation with follow-up CT-ASPECTS (r=0.92).
• CTA-venous-ASPECTS showed a substantial correlation with 90-day modified Rankin Scale (mRS) scores (r=-0.64).
• CTA-venous-ASPECTS yielded the highest area under the ROC curve (AUC=0.82) for predicting good outcomes, outperforming mismatch-ASPECTS (AUC=0.75).

Conclusions:

• CTA-venous-ASPECTS is a reliable automated tool for predicting infarct extent and clinical outcomes in LVO stroke.
• Mismatch-ASPECTS, reflecting different angiographic phases, also shows prognostic value and high sensitivity for identifying patients with good outcomes.