Canceled Code Strokes: Evaluating BE FAST Specificity and Workflow Efficiency in Emergency Stroke Triage
Karim Makhoul1, David Blihar1, Spencer McFarlane1, Pawan Puli1, Tehshina Mallick1, Angela Xia1, Isaac Lipsky1, Ting Huai Shi1, Isabella Tincher2, Allison Winter2, Joshua Assi1, Richard Libman1
1Northwell Health, 2Donald and Barbara Zucker School of Medicine at Hofstra/Northwell Health
Objective:

To evaluate the characteristics and final diagnoses of emergency department (ED) code strokes (CS) that were subsequently canceled, with focus on identifying limitations of the BE FAST triage tool.

Background:

Stroke mimics account for major portion of ED CS activations. While prior studies have described stroke mimics, little attention was given to canceled CS in literature. Understanding this group might provide insight into workflow efficiency to help improve specificity of screening tools such as BE FAST.

Design/Methods:

We retrospectively reviewed 368 canceled codes at a single center in 2024. Data included were presenting complaint, NIHSS, baseline disability (mRS) and final diagnosis. Descriptive statistics were used for analysis.

Results:

Patients with canceled CS mostly presented with mild neurologic deficits (NIHSS 0–5) and moderate baseline disability (mRS 2–3). The reasons for cancellation consisted of: patient being out of the window for acute management; having hemorrhage not requiring acute intervention; or with a nonspecific deficit alongside presence of alternate diagnosis explaining the deficit.  The leading triggers were mostly nonspecific including altered mental status (22%) and speech disturbance (14%), while motor weakness (11%) ranked 3rd to nonspecific triggers. Cancellation was highly specific and none missed an acute intervention opportunity. While 4 % had cerebral infarct on MRI, none required intervention mostly due to returning to baseline NIHSS of zero or being out of the recommended window. Most cancellation decisions (77%) were made after agreement between emergency and neurology teams, while 19% were canceled by a single provider for not meeting BE FAST. Final diagnoses included seizures, toxic encephalopathy, vestibular disorders, cardiac, respiratory, primary ophthalmologic or alternate neurologic etiology such as migraine or demyelination.

Conclusions:
Our analysis confirmed that BE FAST is highly sensitive but nonspecific. Mild neurologic deficits and patients with pre-existing disability were the major contributors. Analyzing canceled CS can help improve ED triage algorithms.
10.1212/WNL.0000000000215951
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.