• 
• 

Three Part Question

[In adults] Is [BE-FAST a better stroke prediction tool] [compared to FAST]

Clinical Scenario

A family have concerns over their 56 year old family member who has presented with sudden on-set coordination/gait problems, who has also vomited twice. An ambulance is called and the patient has been assessed as FAST neg. Despite this, the gentleman was transported to the local emergency department. When in the emergency department he was further assessed and found to also have ataxia and problems with vision. CT head revealed the patient was suffering with an ischemic stroke effecting a posterior artery.

Search Strategy

Literature search on the internet, using the following databases: AHA/ASA Journals and PubMed. The following search terms were utilised.

“BEFAST”, “FAST" AND “BEFAST”

Search Outcome

Initially 20 papers found with the selected search term. Out of these, 4 papers were selected for review.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
S. Aroor, et al 2016 USA	Adult patients admitted to University of Kentucky Stroke Centre between January and December 2014 with a discharge with a diagnosis of acute ischemic stroke. 736 records met the inclusion criteria.	Case series study CEBM 4	The paper looked at patient records who were diagnosed with stroke that did not have FAST symptoms. The proportions of patients missed based on FAST was compared with the proportion missed after inclusion of gait-related (gait imbalance or lower extremity weakness) or visual (visual loss and diplopia) symptoms (i.e., balance/leg weakness [B] and vision, eyes [E]; BE-FAST).	The paper rejects the null hypothesis P=0.042. 14% of strokes did not have FAST symptoms BE-FAST could reduce this to 4.4% p=0.001. 95% CI. 70% of the patients had the symptoms of either gait/leg weakness or a visual impairment, wither alone or in combination of other non-FAST symptoms. BE-FAST could potentially capture >95% of ischemic strokes.	Design is a retrospective study. Study used symptoms recorded on their medical records may be subjective and not all symptoms may have been recorded. Study took place only in one hospital. Specificity is not able to be accurately recorded.
X. Chen, et al 2021 Switzerland	6151 participants. obtained from searching the PubMed, Embase and Cochrane libraires using the keywords “stroke,” “ischemic stroke,” and “haemorrhagic stroke”.	Meta analysis 1a including 9 studies.	Primary outcomes including true positives, false positives, true negatives with FAST and BE-FAST. The reported symptoms were cross match to FAST and BEFAST.	BE-FAST has a lower sensitivity of 0.68 and a higher specificity of 0.85 compared to FAST sensitivity 0.77 specificity 0.60. The study concluded that BEFAST has a higher diagnostic value compared to FAST.	Full data not easily interpreted for yourself, only one clear chart. Limited use of P values, may have include animal studies.
S. P. Jones, et al 2020 UK	6934 participants over the age of 18 with a hospital diagnosis of stroke that was not identified at initial assessment by EMS with included paramedics and technicians in the pre-hospital setting	Observational systemic review	Looking at patient records to identify the symptoms of patients who had a diagnosis of stroke who were not initially identified by the EMS. By adding the most common missed symptoms to the stroke screening tool, does this improve pre-hospital stroke identification.	2-52% of all stroke presentations transported by the emergency medical services (EMS) are not identified in the pre-hospital setting. The most reported missed symptom were speech problems, nausea/vomiting, changes in mental state, dizziness and visual disturbances. The study reported that adding balance (imbalance) or leg weakness and visual symptoms to the FAST symptoms would have improved stroke recognition from 86%-96% p=0.0001. However further studies are needed to evaluate the sensitivity and specificity.	The studies were carried out on patients that the EMS clinician suspected of stroke or had a hospital diagnosis of stroke. It was not clear from the studies whether the symptoms data had been reported by the EMS or whether the symptoms had been missed by the EMS and reported in hospital. Over half the Studies had high risk of selection bias.
D. Pickham 2019 California	359 adult participants 200 non stroke and 159 confirmed stokes with acute stroke symptoms less than 6 hours old in the pre-hospital setting	Prospective study case series CBEM 4	Whether adding balance / coordination and eye / diplopia to the FAST screening tool will improve stroke identification in the prehospital setting	The study found no statistically significant difference between FAST and BE-FAST in the pre hospital identification of stroke P=0.09. BE-FAST has a sensitivity of 0.91 and a specificity of 0.56 compared to FAST with a sensitivity of 0.79 and a specificity of 0.68. compared with non-stroke patients, stroke patients had a higher percentage of abnormal speech 63%, arm weakness 55%, Balance / coordination 55% facial droop 52% and eye / diplopia 31%.	Small cohort 359 of these 159 were confirmed stroke, cohort included patients with "stroke symptoms" already. Posterior strokes do not necessarily present with "stroke like" symptoms which BE-FAST may have been a beneficial assessment tool

Comment(s)

There are not that many studies that have looked into the effectiveness of adding addition symptoms to the FAST test such as balance/coordination, eyes/diplopia (BE-FAST). The studies that have been completed, many of these include participant already with stroke symptoms and retrospectively investigate whether BE-FAST would have been a superior detection tool. Due to the fact BE-FAST incorporates more symptoms than FAST, BE-FAST is statically able to identify more strokes than the stand-alone FAST test. Due to the potential difference in symptoms, BE-FAST is more inclusive of strokes effecting the posterior cerebral arteries. Potentially reducing missed strokes from 14% to <5% This being said, the greater inclusion of symptoms BE-FAST has (higher sensitivity), may lead to a larger number of false positives (lower specificity) compared to FAST.

Clinical Bottom Line

BE-FAST has the potential to identify stroke symptoms that may be missed by the stand-alone FAST test. Because of this BE-FAST may well lead to a larger proportion of patients being identified as potential strokes and pre-alerted to the emergency department. This putting extra strain on specialist acute stroke teams in hospitals. It seems BE-FAST may have its place as identification tool for the public and public education programs could be revised to incorporate the additional symptoms of BE-FAST. However clinical practice should not change to use BE-FAST. To reduce the risk of missing strokes in the pre-hospital environment, clinically trained staff should assess patients based on a neurological examination, not the FAST test to rule in or out stroke. Staff training should take place to fill any short falls in knowledge regarding the signs and symptoms of posterior circulation strokes and any knowledge gaps in their ability to complete a neurological examination.

References

1. S. Aroor, et al BE-FAST (Balance, Eyes, Face, Arm, Speech, Time) Reducing the Proportion of Strokes Missed Using the FAST Mnemonic AHA Journals 2016, 479-481
2. X. Chen, et al A Systematic Review and Meta-Analysis Comparing FAST and BEFAST in Acute Stroke Patients Frontiers in Neurology 2021, 1-12
3. S. P. Jones, et al Characteristics of patients who had a stroke not initially identified during emergency prehospital assessment: a systematic review BMJ 2020, 387-393
4. D. Pickham Prognostic value of BE-FAST to identify strokes in a pre-hospital setting Prehospital Emergency Care Journal 2019, 195-200