The aim of the study is to validate of the use of plasma B-type natriuretic peptide (BNP) point-of-care test platform in preliminary recognition of cardioembolic stroke patients in the emergency department (ED). In our ED, emergency physicians prospectively assessed consecutive adult patients with acute phase of ischemic stroke and measured plasma BNP by point-of-care test platform on admission. The included patients with plasma BNP concentration more than 66.50 pg/mL were presumed to be classified as the cardioembolism (CE) subtype and were then followed up. Stroke neurologists evaluated patients' functional outcome at hospital discharge and also made discharge diagnosis and stroke etiologic subtypes according to Trial of ORG 10172 in Acute Stroke Treatment criteria: large artery atherosclerosis, CE, small artery occlusion, stroke of other determined etiology, and stroke of other undetermined etiology. In this study, 172 of 262 acute ischemic stroke patients met the study criteria (mean age, 71.18 ± 11.65 years; 53.49% female). Of the 172 patients, 38.95% were diagnosed with large artery atherosclerosis at discharge; 26.16%, with CE; 24.42%, with small artery occlusion; and 10.47%, with stroke of other determined etiology or stroke of other undetermined etiology. Age, previous cardiac disease, atrial fibrillation, length of hospital stays, Scandinavian Stroke Scale score on admission less than or equal to 25, and modified Rankin Scale greater than or equal to 3 or death at discharge were all significantly higher in the CE patients compared to other subtypes (P < .01). The mean BNP concentration was significantly higher in the CE group than in other 3 subtypes (P < .01). The plasma BNP level greater than 66.50 pg/mL had good corresponding diagnostic performance in preliminary recognition of cardioembolic stroke patients (sensitivity, 75.56%; specificity, 87.40%). In this study, we found that the plasma BNP level greater than 66.50 pg/mL as a reference index had good corresponding diagnostic performance in preliminary recognition of cardioembolic stroke patients. However, the single BNP biomarker panel cannot be used to confidently identify CE subtype as a diagnosis and must be taken in context with clinical assessment and judgment before making management decisions. Copyright © 2015 Elsevier Inc. All rights reserved.