Composite small vessel disease scores predict hemorrhagic transformation after thrombectomy: a machine learning study
Thiago Oscar Goulart1, Rui Kleber do Vale Martins-Filho2, Millene Rodrigues Camilo2
1Department of Medicine, Temerty Faculty of Medicine, University of Toronto, Toronto, ON, Canada; Department of Neurology, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada; Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA, USA; Department of Neuroscience and Behavioral Sciences, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, São Paulo, Brazil.
Composite small vessel disease (CSVD) scores, especially the modified SVD (mSVD) score, effectively predict hemorrhagic transformation (HT) risk after mechanical thrombectomy (MT). This aids in better patient risk stratification and treatment decisions for acute ischemic stroke.
Area of Science:
- Neurology
- Radiology
- Cardiovascular Research
Background:
- Hemorrhagic transformation (HT) is a significant complication following acute ischemic stroke (AIS) treatments like mechanical thrombectomy (MT).
- Identifying predictive markers for HT and symptomatic intracranial hemorrhage (SICH) is crucial for patient management.
Purpose of the Study:
- To evaluate the predictive capability of composite small vessel disease (CSVD) scores, specifically the modified SVD (mSVD) score and Brain Frailty Score (BFS), for HT and SICH post-MT.
- To compare the predictive performance of CSVD scores against individual imaging markers.
Main Methods:
- Retrospective analysis of AIS patients undergoing MT, with or without intravenous thrombolysis (IVT).
- Quantification of CSVD burden using mSVD and BFS scores based on white matter hypodensities, brain atrophy, and lacunes.
- Logistic regression and XGBoost models were used to assess predictive performance (AUC) and feature importance (SHAP).
Main Results:
- The mSVD score demonstrated the highest predictive accuracy for HT (AUC=0.913) and SICH (AUC=0.862), outperforming BFS and individual imaging markers.
- Key predictors identified by SHAP analysis included NIHSS, glycemia, systolic BP, age, and the mSVD score.
- Intravenous thrombolysis (IVT) was not significantly associated with increased HT risk after adjusting for CSVD scores.
Conclusions:
- CSVD scores, particularly the mSVD score, are independent predictors of HT risk after MT, even when considering acute clinical factors.
- These scores can enhance early risk stratification and inform treatment decisions for AIS patients.
- Further validation in multicenter cohorts using advanced imaging techniques is recommended.