Related Experiment Video
Updated: Mar 22, 2026
Endovascular Perforation Model for Subarachnoid Hemorrhage Combined with Magnetic Resonance Imaging MRI
Published on: December 16, 2021
Quantitative MRA Feature Prediction of Post-Aneurysmal Subarachnoid Hemorrhage Delayed Cerebral Ischemia and
Mohamad H Mosi1, Mona Kharaji1, Mehmet Aksakal1
1From the Department of Radiology (M.H.M., M.K., M.A., C.W., D.C., M.R.L., G.C., C.Y., N.B., M.M.-B.), Bioengineering (K.Z.), Neurosurgery (M.R.L.), Mechanical Engineering (M.R.L.), University ofWashington, Seattle, WA, USA; Department of Radiology (M. A.), University of Colorado, Aurora, CO, USA and Department of Radiology (M.M.-B.), University of Alabama-Birmingham, Birmingham, AL, USA.
Background And Purpose:
Complications of aneurysmal subarachnoid hemorrhage (aSAH), including vasospasm and delayed cerebral ischemia (DCI) substantially contribute to morbidity and mortality. We investigated the correlation between quantitative measurements on MRA performed immediately post-endovascular embolization and subsequent development of DCI or vasospasm.
Materials And Methods:
We included consecutive patients scanned between 9/1/2016 and 1/14/2022 with ruptured intracranial aneurysms (IAs) after endovascular treatment. We recorded clinical data including sex, age, BMI, smoking history, vascular risk factors, baseline CTA/CT features, modified Fisher score, and Glasgow Coma Score. We extracted quantitative MRA features, including total arterial length, total branch number, and average tortuosity, using a custom semi-automated software (VesselVoyager). Correlations between baseline quantitative MRA, other imaging and clinical features with DCI and vasospasm were determined using logistic regression analysis. Receiver operating characteristic (ROC) curves were calculated for clinical-only models and for models augmented with quantitative MRA features.
Results:
78 patients with ruptured IAs were included, of whom 48 developed vasospasm and 24 developed DCI. Multivariable logistic regression analysis showed that only average tortuosity was significantly and independently associated with subsequent DCI (OR 0.62, 95% CI 0.39-0.91, p=0.02). For vasospasm, total arterial length (0.87, 0.78-0.95, p=0.005), average tortuosity (0.37, 0.18-0.62, p=0.001), and age (0.92, 0.86-0.98, p=0.02) were independently associated with vasospasm incidence on multivariable regression. Adding quantitative MRA features improved ROC performance for both DCI (AUC 0.63 [95% CI 0.51-0.75])vs 0.75 [0.64-0.87]) and vasospasm (0.67 [0.54-0.79] vs 0.85 [0.77-0.94]).
Conclusion:
In patients with aSAH, average tortuosity is independently inversely associated with future development of DCI. Total branch number and average tortuosity are independently inversely associated with future vasospasm. With further validation, quantitative MRA features may serve as predictive markers for DCI and vasospasm after aSAH.
Insights
Quantitative MRA measurements, specifically arterial tortuosity, may predict delayed cerebral ischemia (DCI) and vasospasm after aneurysmal subarachnoid hemorrhage (aSAH). These findings could improve patient outcomes by enabling early intervention for high-risk individuals.
Area of Science:
- Neurology
- Radiology
- Medical Imaging
Background:
- Aneurysmal subarachnoid hemorrhage (aSAH) complications like vasospasm and delayed cerebral ischemia (DCI) significantly increase morbidity and mortality.
- Current prediction methods for these complications are limited.
Purpose of the Study:
- To investigate the correlation between quantitative MRA measurements taken immediately post-embolization and the subsequent development of DCI or vasospasm.
- To assess the predictive value of MRA features for post-aSAH complications.
Main Methods:
- Retrospective analysis of 78 patients with ruptured intracranial aneurysms treated with endovascular embolization.
- Quantitative MRA features (arterial length, branch number, tortuosity) extracted using semi-automated software.
- Logistic regression and ROC curve analysis to determine correlations and predictive performance.
Main Results:
- Average arterial tortuosity was independently associated with a lower risk of DCI.
- Total arterial length and average tortuosity were independently associated with a lower risk of vasospasm.
- Incorporating quantitative MRA features improved the predictive accuracy for both DCI and vasospasm.
Conclusions:
- Quantitative MRA features, particularly average tortuosity, show potential as independent predictors of DCI and vasospasm after aSAH.
- These imaging biomarkers may aid in identifying patients at higher risk for complications, guiding further management.
- Further validation is warranted to establish these MRA metrics in clinical practice for predicting post-aSAH outcomes.
