Dynamic CTA-Based Whole-Brain Arterial-Venous Collateral Assessment for Predicting Futile Recanalization in Acute...

Area of Science:

• Neurology
• Interventional Radiology
• Medical Imaging

Background:

• Futile recanalization poses a significant challenge for acute ischemic stroke (AIS) patients undergoing endovascular treatment (EVT).
• Predicting this outcome is crucial for optimizing treatment strategies and patient management.
• Current predictive methods may not fully capture the complex interplay of factors influencing recanalization success.

Purpose of the Study:

• To develop and validate a predictive model for futile recanalization in AIS patients post-EVT.
• The model integrates arterial-venous collateral status with clinical parameters.
• To assess the model's performance across training, internal validation, and independent external validation cohorts.

Main Methods:

• A cohort of 392 AIS patients with anterior circulation large vessel occlusion treated with EVT was analyzed.
• Predictors were identified using Boruta algorithm and LASSO regression.
• Machine learning models were evaluated using discrimination, calibration, and decision curve analyses; SHAP analysis assessed feature importance.

Main Results:

• Age, whole-brain arterial collateral status, and whole-brain venous collateral status were identified as independent predictors.
• The model demonstrated high discrimination in training (AUC: 0.914) and internal validation (AUC: 0.918) cohorts.
• Robust performance was maintained in an independent external validation cohort (AUC: 0.755), with good calibration.

Conclusions:

• Integrating whole-brain arterial-venous collateral assessment with clinical parameters shows promise for predicting futile recanalization after EVT.
• The validated model may enhance clinical decision-making for AIS patients.
• Further research can explore refining this predictive tool for broader clinical application.