Dispatch sends you to a truck stop for a 58-year-old long-haul trucker who "can't walk straight." He is sitting on the step of his cab. Face is symmetric, both arms hold, speech is clear, and glucose is 112 mg/dL. He tells you he sees two of everything and drifted into the fuel island twice. His last normal moment was 50 minutes ago, when he finished a phone call with his company's dispatcher.
Bottom line: a negative or low stroke screen does not make this a minor call. At the hospital, the lytic decision for a low-score stroke turns on whether the deficit is disabling and how much of the 4.5-hour window is left, and you are the person on the call who can document both.
Why scene screens miss low-score strokes
Face, arm, and speech findings cover the anterior circulation well and posterior strokes poorly. In a review of 736 consecutive ischemic stroke admissions at the University of Kentucky stroke center, 14.1% had no FAST symptom at presentation [1]. Of those missed patients, 42% had gait imbalance or leg weakness and 40% had visual symptoms. Adding balance and eyes to the screen cut the missed proportion to 4.4% [1]. That was a single-center retrospective study, and the authors asked for prospective validation before changing public education [1].
Large vessel occlusion screens have their own ceiling. PRESTO followed 1,039 FAST-positive suspected strokes transported by ambulance to eight hospitals in the Netherlands within 6 hours of onset [2]. Paramedics scored eight prehospital scales on a phone app. 120 patients (12%) had an anterior circulation large vessel occlusion. RACE performed best (area under the curve 0.83), followed by G-FAST and CG-FAST (0.80 each), LAMS and Cincinnati (0.79), and FAST-PLUS (0.72) [2]. The NIHSS done by clinicians in the emergency department did slightly better at 0.86 [2]. PRESTO enrolled only FAST-positive patients and counted only anterior occlusions, so it tells you nothing about the trucker above. The 2026 AHA/ASA guideline calls prehospital scales effective for detection and still nondiagnostic [3].
What disabling means at the receiving hospital
The guideline says the NIHSS alone does not decide treatment. Its own example is a leg deficit that prevents walking, scoring only 2 and still disabling [3]. For patients scoring 0 to 5, the hospital asks one question: if this deficit persists, can the patient still do basic activities of daily living or return to work [3]? The guideline's mnemonic is BATHE: bathing and dressing, ambulating, toileting, hygiene, eating. It tells clinicians to assess walking and swallowing and to decide with the patient and available family [3].
Deficits it treats as typically disabling include complete loss of half the visual field, severe aphasia, severe neglect, and any weakness that cannot hold against gravity [3]. Isolated facial droop, numbness on one side, mild clumsiness with preserved walking, and mild word-finding trouble with meaningful communication often do not meet that bar [3]. The PRISMS trial, which tested alteplase in patients judged not clearly disabling, used the same daily-activities test and required that patients walk unassisted [4].
This is where your history matters. The family member who knows the patient's baseline is often on scene and not at the hospital. Ask what the patient could do this morning that he cannot do now. Record whether he walked to the stretcher or needed two people. Write down the work he returns to. For the trucker, a driver who cannot hold a lane is not a minor stroke to the person who has to sign his return-to-work form.
Time, glucose, and the radio report
The guideline lists the prehospital stroke metrics that track with faster hospital treatment and more discharges home: dispatch within 90 seconds, a documented stroke screen, a glucose check, a last known well time, scene time of 15 minutes or less, prenotification with stroke-specific details, and an IV line [3]. Glucose matters for eligibility. Severe hypoglycemia (under 50 mg/dL) or hyperglycemia (over 400 mg/dL) is corrected first, and deficits that remain disabling afterward stay eligible [3].
Last known well sets the window. Inside 4.5 hours, eligible patients with disabling deficits receive tenecteplase 0.25 mg/kg (maximum 25 mg) or alteplase 0.9 mg/kg, and the guideline recommends treating as quickly as possible without waiting for advanced imaging [3]. When a patient cannot consent and no representative is present, the guideline supports treating an otherwise eligible patient with disabling deficits [3]. A witness who can confirm last known well still saves the team minutes, so bring one or get a callback number.
Destination and occlusion
The guideline recommends direct transport to the closest thrombectomy-capable hospital where well-functioning rapid transfer systems are absent [3]. Where local hospitals give lytics efficiently and transfer quickly, direct transport of suspected occlusions to a center 45 to 60 minutes away did not improve 3-month outcomes in the RACECAT trial, and the guideline rates that bypass as no benefit [3]. Follow your regional triage plan, because the guideline places the screening tool choice inside that plan [3].
A low-score patient with an occlusion is not low risk. TEMPO-2 enrolled 886 patients with NIHSS 0 to 5 and a proven intracranial occlusion or perfusion deficit within 12 hours at 48 hospitals in ten countries [5]. Tenecteplase 0.25 mg/kg did not improve return to baseline function (72% versus 75% with standard care), and deaths were 5% versus 1% [5]. The authors concluded these patients should not routinely receive thrombolysis [5]. The guideline's strongest thrombectomy recommendation covers ICA or M1 occlusions with NIHSS 6 or higher within 6 hours [3]. Low-score patients with occlusion sit outside that recommendation, which is why the receiving team needs your exam trend and timeline.
Mobile stroke units carry a strong recommendation where available [3]. BEST-MSU, a US alternating-week observational trial, enrolled 1,515 patients. Among 1,047 lytic-eligible patients, onset-to-treatment was 72 minutes with a mobile unit versus 108 with standard EMS, and modified Rankin 0 to 1 at 90 days was 55.0% versus 44.4% [6].
Key Takeaways
- Add balance and eyes to your screen: in the Kentucky series, 14.1% of ischemic strokes had no FAST sign, and adding gait and vision cut misses to 4.4% [1].
- Document the BATHE answer, whether the patient walked, and who confirmed baseline. An NIHSS 2 leg deficit that prevents walking is disabling [3].
- Hit the guideline metrics: glucose, last known well, stroke screen, scene time of 15 minutes or less, and prenotification with stroke details [3].
- A low-score patient with occlusion is high risk: in TEMPO-2, routine tenecteplase did not help and deaths were 5% versus 1%, so transport by your regional plan and report the exam trend [3,5].
FAQ
Can a patient with a negative Cincinnati or FAST screen be having a stroke?
Yes. In one stroke center review, 14.1% of ischemic strokes had no face, arm, or speech sign, most often presenting with gait imbalance, leg weakness, or visual symptoms [1].
What should EMS document for a low-score stroke patient?
Last known well, glucose, the stroke screen result, whether the patient could walk, and what daily activities or work the deficit prevents, with the name and number of the person who confirmed baseline [3].
Should a low-score suspected occlusion bypass the nearest stroke center?
Follow your regional triage plan. The guideline favors the closest thrombectomy-capable hospital when rapid transfer systems are not working well, and finds no benefit from bypass where local centers treat and transfer efficiently [3].
Dr. Chet's Take
I have served as a HEMS and critical care transport medical director long enough to know which line matters on a low-score stroke chart. It is the line where the crew wrote down what the patient could not do. This piece gets the important thing right. The decision at the hospital is a disability decision, and EMS holds most of the evidence for it. The core argument holds. My addition is about the radio report. Crews are trained to give a scale score, and a score of 1 or 2 sounds like good news to the person receiving it. Lead with the function instead. "Cannot stand without two people" changes the room before the doors open. The score still goes in the report. It just should not be the headline.
That being said, the evidence for treating low-score strokes is thinner than many protocols suggest. The honest answer is that nobody has shown that a lytic helps a patient with a small, disabling deficit and an occlusion, and the trial that came closest pointed the other way. That makes the EMS job more precise. The thrombectomy team needs to know whether the deficit is getting worse on your watch, because a low-score patient who fluctuates is a different patient. Repeat your exam at the door and say whether it changed. The opposite trap matters just as much. A patient whose symptoms fade in the back of the truck is still a stroke patient, and early recurrence is the reason the hospital starts antiplatelet therapy fast in non-disabling strokes. Improvement is not a reason to accept a refusal without a hard conversation or to choose a slower destination.
If you are running this call, get last known well from a witness, check glucose, and test walking if the patient is safe to stand. Write the BATHE answer in plain words on the chart, and give the hospital a callback number for the person who knows baseline. If you are flying, do your second exam in the air and report the change. Ask your medical director whether the regional stroke plan has a pathway for posterior strokes that screen negative. If it does not, that is a protocol gap and it belongs on the next agenda. Then run one low-score posterior stroke at your next skills day, with the patient sitting up and talking, and see how many crews call the alert. Function goes first in every stroke report you give.
— Chester Shermer, MD, FACEP | Emergency Medicine, 25+ Years Clinical Experience | State Surgeon
Continue Your Training
If you're an EMT, paramedic, flight medic, or CCTP (or any clinician treating patients daily) trying to understand how AI will actually impact your clinical practice — not just the hype — I put together a free practical guide. You can download the AI in EM Survival Guide here.
Browse All Courses at Global MedOps Command
Relevant Reading on Global MedOps Command:
How to Avoid Becoming an AI Casualty — Dr. Shermer's guide to navigating AI tools in clinical and operational settings without compromising judgment or patient outcomes.
Emergency Department Efficiency Playbook — Practical systems for throughput, triage optimization, and operational efficiency.
Read more from Dr. Shermer on Medium
Connect with Dr. Shermer: LinkedIn — Chester "Chet" Shermer, MD, FACEP.
References
- Aroor S, Singh R, Goldstein LB. BE-FAST (Balance, Eyes, Face, Arm, Speech, Time): Reducing the Proportion of Strokes Missed Using the FAST Mnemonic. Stroke. 2017;48(2):479-481. PubMed
- Duvekot MHC, Venema E, Rozeman AD, et al. Comparison of eight prehospital stroke scales to detect intracranial large-vessel occlusion in suspected stroke (PRESTO): a prospective observational study. Lancet Neurol. 2021;20(3):213-221. PubMed
- Prabhakaran S, Gonzalez NR, Zachrison KS, et al. 2026 Guideline for the Early Management of Patients With Acute Ischemic Stroke: A Guideline From the American Heart Association/American Stroke Association. Stroke. 2026;57(8):e316-e436. PubMed
- Khatri P, Kleindorfer DO, Devlin T, et al. Effect of Alteplase vs Aspirin on Functional Outcome for Patients With Acute Ischemic Stroke and Minor Nondisabling Neurologic Deficits: The PRISMS Randomized Clinical Trial. JAMA. 2018;320(2):156-166. PubMed
- Coutts SB, Ankolekar S, Appireddy R, et al. Tenecteplase versus standard of care for minor ischaemic stroke with proven occlusion (TEMPO-2): a randomised, open label, phase 3 superiority trial. Lancet. 2024;403(10444):2597-2605. PubMed
- Grotta JC, Yamal JM, Parker SA, et al. Prospective, Multicenter, Controlled Trial of Mobile Stroke Units. N Engl J Med. 2021;385(11):971-981. PubMed
Related simulation training: EM-Sim · Books by Dr. Shermer