Validity and Practicality of TURN prior to Administration of r-tPA in Acute Ischemic Stroke Joseph Morris, Stephanie Hyche, Eric Clayton Memorial University Medical Center - Savannah, GA
Background/Purpose: Thrombolytic therapy with recombinant tissue plasminogen activator (r-tPA) used in ischemic stroke carries significant risk of intracranial hemorrhage (ICH). In 2015 TURN (Thrombolysis risk using mRS and NIHSS) was developed to quantify the risk of ICH based on neurologic exam. This study sought to determine if the tool could be reliably implemented at a primary stroke center to assist in prescriber and patient decision-making.
Methodology: This study reviewed retrospective data from 151 ischemic stroke patients to identify independent risk-factors associated with ICH. These factors were then used to develop a model whose predictive ability was compared to TURN. Additionally, observational data was collected during stroke protocols to determine the frequency at which patients present with enough medical history to employ TURN at bedside.
Results: Of the 20 parameters analyzed, hyperdense MCA (middle cerebral artery) sign, elevated blood glucose, NIHSS (National Institute of Health Stroke Scale, and the presence of visual field deficits were identified as independent predictors of ICH after univariable analysis, but only MCA sign (p < 0.0005) and blood glucose (p 0.009) remained statistically significant after multivariable analysis. When compared to TURN, the generated model had better predictive ability [AUROC 0.806 (95% CI: 0.632, 0.980) vs 0.740 (95% CI: 0.610, 0.870)].
Conclusions: Independent predictors found in this study were different than those in the TURN study. Despite this, TURN performed nearly as well as the model generated from this dataset suggesting either model may be a viable option for bedside decision-support.
Presentation Objective: Describe a tool used to predict ICH risk in stroke patients.
Self-Assessment: What determiners were found to independently predict ICH in this study?
