Gea M, Álvarez M, Forcén S, Paré M, Sorrentino A, Zhu N, Planas-Ballvé A, Broto J, Martín-Aguilar L, Ramos-Pachón A, Hernández-Pérez M, Dorado L, Gomis M, Millán M, Dávalos A, and Pérez de la Ossa N
Aims: To determine the sensitivity of stroke detection by emergency medical services (EMS) and to analyse the clinical characteristics of unidentified patients with suspected stroke., Patients and Methods: Prospective register of patients with suspected stroke in our area (850,000 inhabitants) from 2011 to 2017. The population that notified the EMS was selected. Of this population, patients with and without stroke code activation by the EMS were compared (EMS+ versus EMS-). Demographics, time to progression, clinical characteristics of the episode and reperfusion therapy administered were recorded., Results: Of a total of 5,497 patients with suspected stroke, 2,087 alerted the EMS: 1,611 (77%) EMS+ and 476 (33%) EMS-. The EMS- patients presented lower scores on the National Institute of Health Stroke Scale (8 vs. 11) and a greater frequency of clinical features of the vertebrobasilar territory (14.1% vs. 8.7%) and partial hemispheric clinical features (23.5% vs. 18.4%), especially in the left hemisphere (78.1% vs. 48.4%). Reperfusion treatment was administered in 29% of EMS+ and 23% of EMS-. The time from symptom onset to treatment was 42 minutes longer in the EMS group (175 versus 133 minutes)., Conclusions: The sensitivity of EMS to detect stroke patients in our series is 77%. We have identified clinical features associated with lack of sensitivity, such as vertebrobasilar territory symptoms or isolated language disorder.