Stroke Triage Alpha Rescue Score (STARS): Prehospital phone-based tool to identify cerebral large vessel occlusion

Area of Science:

• Neurology
• Emergency Medicine
• Medical Informatics

Background:

• Prehospital identification of large vessel occlusion (LVO) stroke is critical for timely endovascular thrombectomy (EVT).
• Existing prehospital stroke assessment tools often require specialized training and expertise.
• There is a need for a simple, accessible method for identifying LVO in prehospital settings.

Purpose of the Study:

• To develop and validate a simple, phone-based score for prehospital identification of LVO.
• To assess the accuracy and feasibility of the Stroke Triage Alpha Rescue Score (STARS) in a real-world prehospital environment.

Main Methods:

• The Stroke Triage Alpha Rescue Score (STARS) was designed using a retrospective cohort of 3452 stroke alerts.
• Item weighting was determined by multivariate analysis of neurological deficits, patient characteristics, and final diagnoses.
• STARS was prospectively validated by dispatchers on 799 suspected stroke alerts.

Main Results:

• The retrospective analysis identified six key items for the STARS score, with a cut-off of 7 points.
• In prospective validation, STARS demonstrated an area under the ROC curve of 0.83 for LVO detection.
• STARS achieved 83% accuracy, 65% sensitivity, and 85% specificity in identifying LVO.

Conclusions:

• The Stroke Triage Alpha Rescue Score (STARS) is a simple, accurate, and teachable phone-based tool for prehospital LVO detection.
• STARS can effectively identify stroke patients eligible for endovascular thrombectomy.
• The score's high accuracy supports its utility in optimizing prehospital stroke care pathways.