Background and Aims

The Face-Arm-Speech-Time (FAST) score to recognize stroke patients primarily detects clinical symptoms of anterior circulation strokes and is less effective in posterior circulation strokes.^1,2 The purpose of the presented study was to investigate if stroke detection in the emergency setting might be improved by extending the established FAST score by 4 additional items („Dizziness“, „Diplopic images“, „Deficit in field of view“, „Dysmetria“) to a new FAST4D score.

Methods

This prospective observational analysis was performed in a district of Hessen, Germany. Paramedics were instructed to use FAST4D instead of FAST by extending the standard operating procedure of stroke treatment. We included all patients who were admitted to the emergency department with suspected stroke and all patients with the diagnosis of stroke when being discharged from hospital. Differences between the sensitivity and specificity of FAST and FAST4D were calculated.

Results

Between 05/2019 and 06/2021 a total of 2,436 patients were screened and 1,876 were included. FAST4D detected 190 (15%) patients more with stroke then did the FAST score. This observation showed a sensitivity for FAST4D of 91% in contrast to FAST with a sensitivity of 71%. Specificity decreased from 56% using FAST to 8% with FAST4D.

Conclusions

The extension of FAST with the 4 additional items to FAST4D increased the stroke detection rate about 15% and potentially improves the time-critical primary care by correct allocation of these patients. The lower specificity should be accepted in light of the necessity to treat stroke patients immediately.

¹Jones EmergMed, 2021; ²Hoyer. FrontNeurol, 2021