Presenting symptoms and diagnostic accuracy of prehospital stroke scales for patients with suspected mild minor stroke

Introduction: Identifying patients with minor stroke is challenging in the prehospital setting due to subtle symptoms. The majority of studies evaluating prehospital stroke scales include patients with high median NIHSS at admission. ParaNASPP, a stepped-wedge cluster-randomized controlled trial found that prehospital NIHSS identified more patients with minor symptoms. Further knowledge on presenting symptoms of patients with suspected minor stroke, and the accuracy of prehospital stroke scales on minor stroke is needed. Methods: A post-hoc analysis of data from the ParaNASPP trial describes prehospital presenting signs and symptoms of patients with suspected mild minor stroke. We defined mild minor stroke as NIHSS 0–2 at hospital admission. Furthermore, we reconstructed and evaluated nine prehospital stroke scales (NIHSS, FAST/CPSS, BE-FAST, LAPSS, MASS, MedPacs, PreHAST, and sNIHSS-EMS) in patients with mild minor stroke. Results: Four hundred and thirty-one patients in the ParaNASPP trial had NIHSS 0–2 at hospital admission. Of these, 152 (35%) were discharged from hospital with a stroke diagnosis. When examined by paramedics, stroke patients presented with speech disturbance, facial palsy, and motor weakness in arm or leg, while stroke mimics presented with dizziness, headache, and nausea/vomiting. NIHSS had the highest sensitivity (95%) and lowest specificity (16%), while LAPSS had the lowest sensitivity (42%) and highest specificity (80%) in the patients with suspected mild minor stroke. The remaining scales had sensitivity between 67% and 93%, and specificity between 23% and 67%. Conclusions: In patients with mild minor stroke, substantial overlap in presentation between stroke and stroke mimics makes triage challenging. Prehospital stroke scales provide either high sensitivity or specificity. Competence and training of paramedics in when and how to use, and interpret, these scales is key for recognizing and correctly triaging stroke patients. The ParaNASPP trial was registered at Clinicaltrials.gov with registration number NCT04137874.