Details

Autor(en) / Beteiligte

• Ali, Syed F
• Singhal, Aneesh B
• Feske, Steven K
• Davis, Lawrence W
• Forducey, Pamela G
• Nystrom, Karin
• Schindler, Joseph L
• Schlegel, Sherene
• Likosky, William H
• Solenski, Nina J
• Southerland, Andrew M
• Schwamm, Lee H

Titel

Abstract 197: The TeleStroke Mimic (TM) Score: A Prediction Rule for Identifying Stroke Mimics Evaluated in a Telestroke Network

Ist Teil von

• Stroke (1970), 2014-02, Vol.45 (suppl_1)

Erscheinungsjahr

2014

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Abstract only Intro: Up to 30% of acute stroke evaluations are deemed stroke mimics (SM). SMs are likely common in telestroke as well, and a model to help a priori identify these patients might be clinically useful. Methods: We used 829 consecutive patients from 01/04 to 04/13 in our internal New England based Partners TeleStroke Network for a derivation cohort and 332 cases for internal validation. External validation was performed on 226 cases from 01/08-08/12 in our Partners National TeleStroke Network. Performance of a prediction rule developed with stepwise logistic regression was characterized by ROC curve analysis. Result: There were 23% SM in the derivation, 24% in the internal and 22% in external validation cohorts based on final clinical diagnosis. Compared to those with ischemic cerebrovascular disease (CVD), SM had lower mean age, fewer vascular risk factors, more often prior seizure and a different profile of presenting symptoms (Table 1). The TM-Score (Figure 1) was based on factors independently associated with SM status including age, medical history (atrial fibrillation, hypertension, seizures), facial weakness and NIHSS >14. The TM-Score performed well on ROC curve analysis (derivation cohort AUC=0.753, internal validation AUC=0.710, external validation AUC=0.770). Conclusion: As telestroke consultation expands, increasing numbers of SM patients are being evaluated. These patients differ substantially from their ischemic CVD counterparts in their vascular risk profiles and other characteristics. Decision-support tools based on predictive models, like the one we propose, may help highlight these differences during complex, time-critical telestroke evaluations.