Machine learning based outcome prediction in stroke patients with MCA-M1 occlusions and early thrombectomy.

BACKGROUND Clinical outcome varies substantially between individuals with large vessel occlusion (LVO) stroke. A small infarct core and large mismatch were found to be associated with good recovery. We investigated if those imaging variables improve individual prediction of functional outcome after early (< 6h) endovascular treatment (EVT) in LVO stroke. METHODS We included 222 patients with acute ischemic stroke due to middle cerebral artery (MCA)-M1 occlusion who received EVT. As predictors, we used clinical variables and region of interest (ROI) based magnetic resonance imaging (MRI) features. We developed different machine learning models and quantified their prediction performance by the area under the curve (AUC) of receiver operator characteristics (ROC) curves and the Brier score. RESULTS Successful recanalization rate was 78%, with 54% patients having a favorable outcome (modified Rankin scale, mRS 0-2). Small infarct core was associated with favorable functional outcome. Outcome prediction improved only slightly when imaging was added to patient variables. Age was the driving factor, with a sharp decrease of likelihood for favorable functional outcome beyond 78 years of age. CONCLUSIONS In patients with MCA-M1 occlusion strokes referred to EVT within 6 hours of symptom onset, infarct core volume was associated with outcome. However, ROI based imaging parameters led to no significant improvement in outcome prediction on individual patient level when added to a set of clinical predictors. Our study is in concordance with the current practice, where mismatch imaging or collateral readouts are not recommended for excluding patients with MCA-M1 occlusion for early EVT.