Related Experiment Video
Updated: Aug 6, 2026
Optimized System for Cerebral Perfusion Monitoring in the Rat Stroke Model of Intraluminal Middle Cerebral Artery Occlusion
Published on: February 17, 2013
Clinical software for unsupervised automated net water uptake analysis predicts futile recanalization in acute
Abdallah Aburub1, Oussama Dob1, Mariana Gurschi1
1Department of Diagnostic and Interventional Neuroradiology, Philipps University of Marburg, Marburg, Germany.
Objectives:
This study aimed to determine whether automated net water uptake (NWU) measured on admission non-contrast computed tomography (CT) independently predicts futile recanalization (FR) in anterior-circulation large vessel occlusion (LVO) stroke treated with endovascular thrombectomy (EVT).
Methods:
This was a retrospective single-center cohort study that included consecutive patients presenting at a tertiary care stroke center between January 2023 and April 2025. Patients were included if they presented with an anterior circulation LVO and had successful recanalization (mTICI 2b-3). An automated platform (VEOcore/MRAY) provided the Alberta Stroke Program Early CT Score (ASPECTS), perfusion metrics, and NWU. Associations with FR were analyzed using logistic regression. Discrimination was evaluated using receiver operating characteristic (ROC) analysis and Youden's J statistic to identify an optimal NWU threshold, reporting the area under the curve (AUC), sensitivity, specificity, and predictive values.
Results:
Among 91 patients, 62 (68.1%) achieved recanalization with a 90-day modified Rankin Scale (mRS) score of 0-4, and 29 (31.9%) met the criteria for FR (mRS score 5-6). Those with FR were older (81.6 ± 8.3 vs. 76.9 ± 10.7 years; p = 0.044) and had more severe strokes at presentation (median National Institutes of Health Stroke Scale [NIHSS] score of 16 [13-20] vs. 9.5 [6-15]; p < 0.001). NWU in the core was higher in FR (21.9% [7.0-29.4]) than in non-FR cases (3.0% [0.5-7.3]; p < 0.001) and correlated with 90-day mRS (Pearson r = 0.602; 95% CI 0.452-0.719; p < 0.001). In multivariable models, NWU remained independently associated with FR (odds ratio [OR] 1.15; 95% CI 1.06-1.24; p < 0.001).
Conclusion:
Automated NWU on admission CT is an independent, strongly discriminative predictor of FR and 90-day disability. A tiered strategy using >11.5% as a risk flag and >17.5% as a high-specificity rule-in threshold may enhance early triage, prognostication, and clinical trial stratification. Prospective multicenter validation is warranted.
