Reader response: Optimizing in-hospital triage for large vessel occlusion using a novel clinical scale (GAI2AA).

Abstract

In their development of the GAI2AA scale,1 Ohta et al. identified a history of atrial fibrillation (AF) or AF on ECG as independent predictors of an anterior circulation large vessel occlusion (LVO) in patients presenting within 4.5 hours of symptom onset. Using a different case definition—all patients who had an occlusion of the common carotid, internal carotid, proximal middle cerebral, anterior or posterior cerebral arteries, or the vertebrobasilar system—and a different patient population—patients admitted to the University of Kentucky Medical Center vs 3 Japanese stroke centers—we also found that a history of AF was associated with an LVO (36.2% vs 18.5%, p < 0.001),2 although the frequency was less than reported by Ohta et al. (65% vs 24%, p < 0.001). In our cohort, the C-statistic for LVO based on a history of AF alone (0.579; 95% CI 0.540–0.617) reflected no more than moderate discriminative capacity and did not improve the accuracy of the Cincinnati Prehospital Stroke Severity scale, Los Angeles Motor Scale, Rapid Arterial Occlusion Evaluation score, or the Field Assessment Stroke Triage for Emergency Destination scale. Nonetheless, simply obtaining a history of AF can alert emergency responders that a patient with an acute focal neurologic deficit may have an LVO.