Simple variables predict miserable outcome after intravenous thrombolysis.

Area of Science:

• Neurology
• Clinical Medicine
• Medical Informatics

Background:

• Ischemic stroke patients receiving intravenous thrombolysis (IVT) have varying outcomes.
• Predicting a miserable outcome is crucial for treatment decisions and resource allocation.
• Existing predictive scores often require complex imaging and laboratory data.

Purpose of the Study:

• To evaluate the predictive accuracy of a Simple Variables Model (SVM) for miserable outcomes in IVT-treated ischemic stroke patients.
• To compare the SVM's performance against the established DRAGON score.
• To assess the utility of the SVM for pre-hospital patient triage.

Main Methods:

• The SVM incorporates basic parameters: age, pre-stroke independence, Glasgow coma verbal score, arm lifting ability, and walking ability.
• A derivation cohort (n=1346) and validation cohort (n=638) of IVT-treated stroke patients were analyzed.
• Area Under the Curve (AUC) with 95% confidence intervals was used to compare the predictive performance of SVM and DRAGON score.

Main Results:

• The SVM demonstrated comparable predictive accuracy to the DRAGON score in both derivation (AUC 0.807 vs 0.822) and validation cohorts (AUC 0.786 vs 0.809).
• A high SVM probability (>70%) for miserable outcome strongly correlated with actual poor outcomes (83% incidence).
• An online SVM calculator is available for estimating individual patient risk.

Conclusions:

• The SVM is a reliable and accurate tool for predicting miserable outcomes post-IVT, performing similarly to the more complex DRAGON score.
• The SVM's reliance on readily available pre-hospital variables facilitates early identification of high-risk patients.
• This model can improve pre-hospital triage, potentially accelerating access to advanced treatments like endovascular therapy.