Clinical decision tool to predict functional outcome after successful endovascular recanalization in anterior-circulation large vessel

Abstract

<p>Background: </p><p>The treatment effect of intra-arterial thrombectomy (IAT) for intracranial large-vessel occlusion (LVO) can vary substantially between individuals with different background characteristics. A predictive tool that provides an evidence-based framework to forecast a patient’s functional outcome is valuable. </p><p>Objective:</p><p>To develop and validate a clinical decision tool based on an Asian cohort of LVO-stroke patients </p><p>Methods:</p><p>A consecutive series of patients treated with IAT at our institution were investigated. The inclusion criteria were : 1) Patients with anterior circulation LVO; 2) IAT performed between April 2017 to April 2023; 3) Successful recanalization defined by a modified treatment in cerebral infarction (mTICI) score of 2b to 3; and 4) clinical follow-up performed at 3 months after operation. 247 patients fulfilled the criteria. </p><p>Clinical and radiological characteristics that are expected to interact with or predict functional outcome were included in a forward stepwise logistic regression model. Included patients were randomly divided into 2 groups, with 70% assigned as the derivation cohort, and the rest as the validation cohort. </p><p><br></p><p>Results:</p><p>On univariate analysis based on the training set, the following predictors were associated with good 3-month mRS<=2: age, occlusion segment, admission NIHSS (The National Institutes of Health Stroke Scale), ASPECTS, rCS , and onset-to-perfusion time. On multivariate analysis, age, occlusion segment, admission NIHSS, ASPECTS, rCS and onset-to-perfusion time emerged as significant. The resultant binary predictive regression equation was validated using receiving operator characteristic (ROC) curve with area under curve being 0.807 (95% CI 0.735-0.879, p< 0.001).</p><p>Conclusions:</p><p>We have presented a precise predictive model for 3-month functional outcome in an Asian cohort of patients with anterior-circulation LVO-stroke post-IAT. This 5-variable simple clinical tool showed good discrimination and calibration in the validation set. </p>