EMS Systems Should Lower the Threshold for Stroke Alert Activation


Oostema JA, Konen J, Chassee T, et al. Clinical predictors of accurate prehospital stroke recognition. Stroke. 2015;46(6):1513—1517.


Recognizing that prehospital stroke recognition and stroke code activation result in better outcomes for patients, the authors of the study examined all EMS records of suspected stroke transported over a 12-month period.

They compared this to the patient’s final diagnosis. They also examined all patients diagnosed with stroke in the ED who weren’t recognized by EMS to be having a stroke.

There were 441 eligible cases. Of those, 371 (84.1%) were “EMS-suspected” strokes and 70 (15.9%) were “EMS-missed” strokes. Of the EMS-suspected cases, 264 (59.9%) were confirmed as either an ischemic stroke or transient ischemic attack (TIA).

All EMS cases were reviewed to determine if a Cincinnati Prehospital Stroke Scale (CPSS) was documented. When it was, EMS sensitivity to recognizing a patient with a stroke was 84.7%, but was only 30.9% when it wasn’t. Documenting a CPSS resulted in a positive predictive value of 56.2% compared to 30.4% without.

The most common EMS impression for “EMS-missed” strokes was generalized weakness, altered mental status and dizziness.

The most common final diagnoses for EMSsuspected strokes that weren’t strokes (false positives) were miscellaneous specific diagnosis (24.3%), nonspecific diagnosis (27.1%), infection (12.4%), seizure (11.3%) and syncope/ hypotension (10.2%).

EMS-suspected strokes made it to the CT scan on average at 34 minutes, while EMSmissed strokes took an average of 84 minutes. Consequently, 14.9% of EMS-suspected stroke patients received fibrinolytics, compared to only 4.4% of EMS-missed stroke patients.

Author conclusions note that EMS recognized 75% of patients with a stroke or TIA; however, half of those suspected with strokes were false positives.


I found this to be a very interesting and important study. It concludes that we, as prehospital providers, are missing a large number of stroke diagnoses. How can we improve our skills and communications to assure better care for patients experiencing a stroke?

The 2009 study “Lack of impact of paramedic training and use of the CPSS on stroke patient identification and on-scene time,” by Frendle DM, Strauss DG, Underhill BK, et al, in the journal Stroke, concluded that even after an hour of training on the specific topic of stroke recognition, the number of missed strokes didn’t improve by EMS. The conclusion was that education alone isn’t the answer.

I think back on my own education. I never saw any emphasis placed on the subtle signs of stroke. For a disease like this to be detected, we must utilize multimedia in the classroom to demonstrate the variety of physical findings seen in different types of strokes.

The study also remarks there was a higher level of accuracy when the CPSS was used. Why isn’t it being used as often as it should? The CPSS should be performed on every patient with an altered level of consciousness, weakness and other symptoms suspicious for stroke.

We need to look at ourselves first and try to improve our knowledge base. But secondly, services need to look at performance of the CPSS as a benchmark for quality. It would be a great opportunity to review what we know, and emphasize what we’re missing.

We have the knowledge and expertise to make a difference. Use the CPSS so your stroke patients receive the care they need.


As with an ST elevation myocardial infarction, the data clearly shows that patients identified by EMS-initiated stroke alerts have better outcomes. The challenge continues to be reliably recognizing patients with stroke symptoms.

In this study, half of the patients suspected of having a stroke didn’t have one. While this results in a 50% over triage, the consequences are probably negligible. My experience is that almost every over-triaged patient I saw in the ED received a CT scan to help determine the cause of their presenting symptoms.

I’m more concerned with how EMS failed to suspect stroke in the EMS-missed group with patients they thought were only weak, altered or dizzy. Unfortunately, this study doesn’t shed light on this other than to indicate these patients had a significantly lower National Institutes of Health Stroke Scale (NIHSS) score. The NIHSS score is an in-depth stroke scoring system performed by the physician.

I suspect the CPSS fails to assist EMS providers in detecting less dramatic findings. Another shortcoming of the CPSS is that it doesn’t test for findings due to strokes in the occipital lobes and cerebellum. These strokes often present with dizziness.

If you do anything, have a low threshold for initiating an ED stroke activation. Once in the ED, the patient can be rapidly assessed by the physician while still on the stretcher. If the doc agrees, then onward to the CT scan.

EMS Today

Learn more from David Page at the EMS Today Conference & Expo, Feb. 25—27, in Baltimore, Md. EMSToday.com

Previous articleLarge-Scale Study Examines Continuous vs. Interrupted CPR
Next articleTeach Providers to Appreciate the Small Parts of the Job

No posts to display