External validation of TICI-ASPECTS-glucose score as a predictive model for symptomatic intracranial hemorrhage following mechanical thrombectomy

Introduction: The indication for mechanical thrombectomy for acute ischemic stroke (AIS) secondary to large vessel occlusion has substantially increased in the past few years, but predictors of symptomatic intracranial hemorrhage (sICH) remain largely unstudied. A recent study assessing these predictors, led to the development of the TICI-ASPECTS-glucose (TAG) score, an internally validated model to predict sICH following thrombectomy. Methods: To externally validate this scoring system and identify other potential risk factors for hemorrhagic conversion following endovascular therapy for AIS, 420 consecutive patients treated with mechanical thrombectomy from 2014-2017 were retrospectively reviewed. Data were collected pertaining to admission factors, procedural metrics, and functional outcomes. The components comprising the TAG score consist of modified thrombolysis in cerebral infarction (mTICI) score (mTICI 0-2a=2 points; 2b-3=0 points), Alberta stroke program early CT (ASPECTS) score (<6=4 points, 6-7=2 points, ≥8=0 points), and glucose (≥150 mg/dL=1 point, <150 mg/dL=0 points). Statistical analyses including univariate analysis, logistic regression analysis, and area under the receiver-operating curve (AUROC) were performed to validate the predictive capability of the model. Results: The patients with sICH presented with lower ASPECTS (8.13±1.55 v 9.16±1.24, p < 0.001), but no significant correlation with mTICI scores and admission glucose was observed. Decreasing ASPECTS correlated with increased risk of sICH (OR 1.57, 95% CI 1.25-1.96, p < 0.001), and increasing TAG score was associated with increased sICH (OR 1.46, 95% CI 1.11-1.94, p < 0.01). AUROC of the model was 0.633. Stratifying patients into low (TAG 0-2), intermediate,^3,4 and high^5-7 risk groups identified similar results to the original study with sICH risks of 5.2%, 10.5%, and 33.3%, respectively. Conclusion: The TICI-ASPECTS-glucose (TAG) score adequately predicts sICH following mechanical thrombectomy, and appropriately stratifies individual patient risk. Further inclusion of additional predictors of sICH would likely yield a more robust model.