Background: We aimed to design a prehospital scale to predict candidates for endovascular thrombectomy (CET) in patients with acute ischemic stroke (AIS).
Materials and methods: In the Fukuoka Stroke Registry, we identified 3,470 patients with AIS who were transferred by emergency medical service within 24 hours of stroke onset and underwent intracranial vessel evaluation on admission from September 2007 to December 2015. CET were defined as patients with causative occlusion of internal carotid artery, middle cerebral artery, or basilar artery, and National Institute of Health Stroke Scale (NIHSS) score ≥ 6. The Fukuoka Acute Stroke ThrombEctomy pRediction (FASTER) scale was developed with NIHSS items based on the predictive importance derived from random forest analysis to predict CET. The discriminative performance was compared with other published scales for large vessel occlusion.
Results: The FASTER scale was designed comprising of 4 NIHSS items : one point each was given for extinction and inattention (NIHSS subscore ≥ 1), best gaze (≥ 1), best language (≥ 1), and motor arm (≥ 2). Receiver operator curves demonstrated that the area under the curve of the FASTER scale was significantly larger than that of the Cincinnati Prehospital Stroke Severity Scale (0.907 vs 0.881, p<0.001), and not significantly different from that of the Rapid Arterial oCclusion Evaluation scale (vs 0.910, p= 0.68). The FASTER scale score ≥2 showed sensitivity of 88.4%, specificity 82.3%, positive predictive value 51.3%, and negative predictive value 97.1% for detecting CET.
Conclusion: The FASTER scale is a simple and promising tool that can identify CET in the prehospital setting.