Novel Algorithm to Help Identify Stroke Mimics

Background: Stroke is a major cause of disability in the United States. A portion of patients presenting with stroke-like symptoms in the emergency room who receive tissue plasminogen activator (tPA) do not end up having a true stroke, leading to unnecessary health-care costs. The aim of our study is to identify those patients who have a high likelihood of experiencing a stroke mimic using a novel stroke mimic score and to identify a cutoff point with a high specificity of ruling in stroke mimics. Methods: We reviewed literature on stroke mimics and the various associated risk factors. We devised a 9-point scoring system and applied it retrospectively to patients who received tPA from 2010 to 2014 to calculate a score for each patient. Results: The final sample size was 105 patients, out of which 25% turned out to be patients with stroke mimics. Patients with stroke mimic were significantly younger and more likely to have history of seizure, migraine, or prior psychiatric illness. History of atrial fibrillation had the highest correlation with true stroke. We found approximately 100% specificity in ruling in a stroke mimic if a patient scored more than 5 points. Conclusions: Our stroke mimic scoring system along with a basic neurologic examination could be a useful tool in the identification of stroke mimics with a high specificity in the emergency room setting. These patients may require further studies such as rapid magnetic resonance imaging, which would decrease unnecessary tPA administration and hospital admissions.