Quantitative Cerebrovascular Analysis for Improved Prediction of Post-Stroke Complications

Clinical Snapshot

Key Findings

• P Value: p<0.05 for AUROC improvements in END, sICH, and mRS outcomes; MBE and TC/PEG significance not explicitly stated

• Effect Size: AUROC improvements: END 0.81 (MI) vs 0.73 (BC); sICH 0.68 (MI) vs 0.56 (BC); MBE 0.67 (MI) vs 0.56 (BC); TC/PEG 0.66 (MI) vs 0.58 (BC)

• Primary Outcome: Discrimination of post-EVT complications by morphology-informed (MI) vs baseline-clinical (BC) ML models, assessed by AUROC

• Nnt Or Sensitivity: Sensitivity, specificity, PPV, NPV, and decision curve analysis not reported in abstract; AUROC for END (MI model) = 0.81 represents the strongest discriminative performance; calibrated probability profiles generated for individualised risk stratification

• Confidence Interval: Not reported in abstract

Clinical Application

Automated vascular segmentation and collateral grading from existing CT angiography or MR angiography data are technically feasible in centres with appropriate computational infrastructure. However, clinical deployment requires prospective validation, regulatory approval of the software pipeline, integration with existing neuroimaging workflows, and clinician training. The tool is not yet ready for routine clinical use without external validation. Endovascular thrombectomy is funded and performed at comprehensive stroke centres across Australia, with EVT volumes increasing following landmark trials. The RACGP and Stroke Foundation Australia guidelines emphasise early risk stratification post-EVT. Automated vascular morphometry tools would require TGA registration as a medical device software (SaMD) under the TGA's Software as a Medical Device framework before clinical deployment. No PBS implications at this stage. Australian comprehensive stroke centres (e.g., Royal Melbourne Hospital, Princess Alexandra Hospital, Royal Prince Alfred Hospital) would be appropriate sites for external validation studies. The tool's scalability aligns with Australia's telehealth and hub-and-spoke stroke network model, potentially supporting remote risk stratification. Adult patients with acute ischemic stroke due to large-vessel occlusion who are candidates for or have undergone endovascular thrombectomy, particularly those at risk of early neurological deterioration, symptomatic intracranial haemorrhage, malignant brain oedema, or requiring tracheostomy/gastrostomy

Abstract

Endovascular thrombectomy (EVT) has transformed the treatment of acute ischemic stroke (AIS). However, a substantial proportion of AIS patients experience poor outcomes despite successful recanalization, often due to severe neurological deterioration or life-threatening complications. Early identification of these high-risk patients remains a major unmet need. In this study, we developed and validated machine-learning (ML) models that integrate automated quantitative brain arterial morphology and collateral grading with demographic, clinical, laboratory, and imaging variables to predict major post-EVT complications and early neurological outcomes. Using a prospectively collected database of 727 AIS patients that underwent EVT, we developed ML models to incorporate patient-specific vascular morphometry with conventional clinical, laboratory, and imaging data to predict emergence of early neurological deterioration (END), symptomatic intracranial hemorrhage (sICH), malignant brain edema (MBE) requiring surgical decompression, and neurogenic respiratory failure and dysphagia requiring tracheostomy/gastrostomy (TC/PEG). Our analysis of morphological features, including increased tortuosity and reduced vessel diameter, showed strong associations with complications. Morphology-informed (MI) models consistently outperformed baseline-clinical (BC) models for patients with END (AUROC 0.81 for MI model vs. 0.73 for BC), sICH (AUROC 0.68 MI vs. 0.56 BC model), MBE (AUROC 0.67 MI model vs. 0.56 BC), or patients who underwent TC/PEG (AUROC 0.66MI vs. 0.58 BC model). Statistical testing confirmed significant AUROC improvements for END, sICH and mRS (p < 0.05), Finally, patient-specific calibrated probability profiles enabled individualized, multidimensional risk stratification, revealing distinct complication-specific risk patterns across patients. These findings demonstrate that cerebrovascular structure-an often overlooked yet physiologically fundamental determinant of ischemic injury and reperfusion dynamics-provides significant predictive information that is not captured by standard clinical or visual imaging assessments. Automated vascular segmentation and collateral grading techniques enable rapid and objective integration of cerebrovascular metrics into prognostic models, offering a scalable tool for precision risk stratification, supporting earlier intervention, targeted monitoring, and improved post-EVT management.

References

1. 1.Deshpande, A., Wang, J., Altaweel, L. R., Yi, S., Bahiru, Z., Leiphart, T. J., Tahsili-Fahadan, P., & Laksari, K. (2026). Quantitative cerebrovascular analysis for improved prediction of post-stroke complications. Translational Stroke Research. https://doi.org/10.1371/journal.pone.0120801