Skip to main content
Patient Care

Mobile Stroke Units: Are They Worth the Cost?

1806 MSUs

They’re a hot idea and spreading rapidly: Since the U.S.’ first mobile stroke ambulance hit the streets in Houston in 2014, at least a dozen have followed in other cities, with more planned.1 There’s just one catch: There’s no data that shows they actually improve patient outcomes. And given their cost, that’s leading an increasing number of docs and other emergency leaders to question whether they really represent the best use of limited dollars. 

“It costs far too much to save a life here,” says Ohio emergency physician Howard Mell, MD, MPH, FACEP, a prominent social media skeptic of mobile stroke units (MSUs). “Even by their own data, these things have a cost and number needed to treat that is high and far outside of what we can do for EMS. There are so many other things we could do with that money—let’s focus instead on the right stroke care, getting people to the right destination. That we can do very cheaply.”

Here’s the math: MSUs cost roughly a million bucks.2,3 “That’s what’s been quoted to us in Louisville,” says Tim Price, MD, an associate professor in the University of Louisville’s Department of Emergency Medicine who argued against MSUs in a pro/con debate at the NAEMSP’s 2018 annual meeting, “and that’s without the telemedicine piece.” Yearly operation may cost half a million to a million more.3,4 

What the data has shown is that the vehicles can expedite treatments like the administration of tPA.5–7 What it hasn’t shown is any clear improvement in outcomes or mortality after that. Studies of one of the longest-running MSUs, Berlin’s STEMO, have found no improvement in patient mortality rates,6,8 and the literature on U.S. efforts is still pocked with qualifiers like “conclusive evidence…is still needed”9 and “has the potential” to minimize mortality.10 

“I think just because someone might get tPA a little earlier doesn’t necessarily translate to a better outcome,” says Price. “They’re pretty small numbers thus far with these units. And tPA is probably not the final answer to this problem. We’re seeing now that mechanical thrombectomy, for the LVOs, is probably going to be the best therapeutic option.” 

Shouldn’t earlier tPA lead to better outcomes for the smaller strokes, though? Not so fast: “That’s the problem here,” says Mell. “We’re rushing to deploy a mechanism we’ve never proven works. There is not a single study that has ever definitively shown tPA works. In fact, a lot of the studies on tPA were stopped early for harm!”

The Value of tPA

This may be startling to providers holding the impression that fast tPA is a wonder drug for stroke. At, author David Newman, MD, overviews 12 key studies that collectively depict something much different: Only two showed clear benefit. Ten were negative, and four of those indeed had to be terminated early.11  

And even those positive studies came with questions. The NINDS-2 trial, published in 1995, had its planned primary outcome, improvement in functional and stroke scores, changed post-hoc to a simple favorable/unfavorable, with the original outcome result not reported.12 It found an overall number needed to treat (NNT) of 8 for that favorable outcome. 

The ECASS-III trial, published in 2008, found no difference in mortality but more favorable outcomes with tPA—but had sicker patients in its placebo group.13 Numerous other problems with this study have led at least one doc to publicly wonder if it was “intentionally deceptive.”14 Its NNT for favorable outcomes was 14.

“The only studies that were positive,” says Mell, “were a couple where they really tortured the data; all the rest have essentially been negative. But they did this big Cochrane review—that’s where a lot of this comes from—where they pushed all these data together. And when they did that, they came up with, if they combined all these studies and massaged the heck out of the data, they believed they maybe found a trend—if you pooled them all together, you could get to a point where there was maybe some benefit. The problem is, there was a lot of heterogenicity between the trials, so trying to combine them is really tough, because you can’t.”

What’s more, while those authors concluded thrombolytic therapy seemed to reduce the proportion of patients dead or dependent in their activities of daily living after strokes, they noted, “This overall benefit was apparent despite an increase both in deaths (evident at 7–10 days and at final follow-up) and in symptomatic intracranial hemorrhages.”15 We know tPA can cause brain bleeding in about 6% of patients and death or serious long-term disability in about 1%.16 

“I think there are arguments on both sides,” says Price of the tPA question, “but there’s certainly a camp that believes the data’s not that compelling. The number needed to treat to have a significant benefit isn’t huge, but it’s not like every patient who’s getting tPA is actually benefiting, which seems to be what a lot of people believe. A lot of these people may have gotten better anyway.”

Some, in fact, may not have had strokes at all: Around 1 in 10 perceived cases of stroke are actually mimics, and decreasing door-to-needle times has been linked to increased rates of mimic treatment. As well, some strokes treated quickly and apparently successfully with tPA may also have been transient ischemic attacks that would have resolved anyway.4 

Few Administrations

And whether or not tPA works for acute ischemic strokes, early data shows that even mobile stroke units intended for its early delivery just aren’t getting tPA to that many patients. Of the first 100 patients treated by Cleveland’s MSU, just 16 ultimately received tPA—5% of those for whom it was dispatched.3,7 A study of information systems around New York Presbyterian Hospital’s unit found 16 administrations in seven months.17 Calgary’s unit gave it 16 times in its first year.18 And in Houston, now site of the most comprehensive look yet at MSUs’ impact (the BEST-MSU study), a run-in phase revealed an administration rate of 1.5 a week.19 

“When you look at these trials, the most favorable of them to tPA suggest a number needed to treat of 8,” says Mell. “But let’s give them better than that—let’s say the number needed to treat is 4 or 5. There’s no data to support that, but assume it for the sake of argument. So in a city the size of Houston, we’re going to put in this multimillion-dollar vehicle and program, and we might save one person a month. Might. If we make all these suppositions there’s no data to support. And that means anyplace smaller, you’ll have less effectiveness, because you won’t be able to support those numbers.” 

The actual NNT is likely not that low. In an analysis of some underpowered German data presented at the 2017 International Stroke Conference, investigators led by UCLA neurologist May Nour, MD, PhD, found 58 of every 1,000 patients treated there to be disability-free, for a hefty NNT of 17.20,21 Expanding their criteria to any patient who benefited to any degree trimmed the NNT to 5.5.20 But applying even that lower figure to Houston’s 1.5 weekly use rate would result in roughly one patient per month with benefit from the millions spent on its MSU.  

“Let me take those millions of dollars and put them toward CPR education, and I guarantee you I’ll save more lives,” says Mell. “Let me take those millions of dollars and put another unit on the street. Let me put them into better gear. Let’s put them into AEDs. Let’s put them into teaching all high schoolers CPR and first aid. Hell, let’s put them into buying every high school graduate a tourniquet—that’ll save more lives!”

Large-Vessel Occlusions

The emphasis in prehospital stroke care today lies less on smaller strokes than large-vessel occlusions. Is there a role for MSUs in identifying these and routing patients to comprehensive stroke centers for fast thrombectomies?

Some recent studies have concluded that’s feasible.22,23 But at NAEMSP Price explored some limitations to that data too. He looked specifically at the conclusions of a Cleveland piece that found the MSU “may help in early triage and shorten the time to IAT (intra-arterial therapy) for AIS (acute ischemic stroke).”24 

The authors of that study compared MSU patients who underwent IAT for AIS with historical controls. After 164 days of service, their MSU had transported 155 patients—of whom 5 received IAT. They showed some shorter key time intervals (door-to-initial-CT, CT-to-IAT), but after 539 dispatches the n was just 5—one IAT for every 100 dispatches. 

Price also took issue with how some of those time intervals were calculated. 

“What you really need to look at is the time from first medical contact until they get to that definitive therapy,” he says, “and on that point they kind of compared apples to oranges. If you break that down and compare apples to apples, there was very little actual difference in how long it took to get the patent into the intervention suite.” That true dispatch-to-IAT difference may be as little as 15 minutes, per his calculations. 

Additionally, Price noted, MSUs so far operate in urban areas that are well suited for direct triage, and even authors who have found MSUs feasible have noted their operational costs.23 

“If we were still just doing tPA for all these patients, perhaps continuing to study this particular treatment method of bringing the CT to the patient and giving them tPA in the field might bear out to be beneficial,” Price says. “But I think that’s not going to be the case now, given the change in the recommended therapy being interventional treatment as opposed to tPA.”

The ACS Model

“I think we have to start looking at strokes like ACS: Not every one of them is going to benefit from the currently available therapies,” says Mell. “There’s no other organ in the body where we take such a one-size-fits-all approach. But for stroke, it’s tPA—always. Doesn’t matter which vessel, doesn’t matter how big. We’re gonna align our whole system to tPA.

“Let’s redesign this thing to be like ACS. Let’s acknowledge that with the light and mild strokes, they may be strokes or they may be mimics, but they really don’t benefit from tPA. And for the big strokes, the hemispheric strokes, let’s rewrite the EMS protocols to get these folks to the regional centers that can do an intra-arterial approach and get the damn clots out. That’s actually the best stroke care.” 


1. Dobkowski D. Functional outcomes may not improve with mobile stroke units. Healio, 2018 Feb 15; 

2. Schencker L. Special ambulances will scan, treat stroke victims in western suburbs. Chicago Tribune, 2016 Oct  20;

3. Zoler ML. Mobile stroke units becoming more common despite cost effectiveness questions. MDedge, Clinical Neurology News, 2017 Mar 14; 

4. Bukata R. Are Mobile CT Units Worth the Price Tag? Emergency Physicians Monthly, 

5. Weber JE, Ebinger M, Rozanski M, et al.; STEMO-Consortium. Prehospital thrombolysis in acute stroke: results of the PHANTOM-S pilot study. Neurology, 2013 Jan 8; 80(2): 163–8.

6. Ebinger M, Winter B, Wendt M, et al.; STEMO Consortium. Effect of the use of ambulance-based thrombolysis on time to thrombolysis in acute ischemic stroke: a randomized clinical trial. JAMA, 2014 Apr 23–30; 311(16): 1,622–31.

7. Taqui A, Cerejo R, Itrat A, et al.; Cleveland Pre-Hospital Acute Stroke Treatment (PHAST) Group. Reduction in time to treatment in prehospital telemedicine evaluation and thrombolysis. Neurology, 2017 Apr 4; 88(14): 1,305–12.

8. Kunz A, Ebinger M, Geisler F, et al. Functional outcomes of pre-hospital thrombolysis in a mobile stroke treatment unit compared with conventional care: an observational registry study. Lancet Neurology, 2016 Sep; 15(10): 1,035–43. 

9. Bowry R, Grotta JC. Bringing emergency neurology to ambulances: mobile stroke unit. Semin Respir Crit Care Med, 2017 Dec; 38(6): 713–7. 

10. Rasmussen PA. Stroke management and the impact of mobile stroke treatment units. Cleve Clinic J Med, 2015 Dec; 82(suppl 2): S17–S21. 

11. Newman D. Thrombolytics for Acute Ischemic Stroke: No benefit found. The NNT, 

12. National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med, 1995 Dec 14; 333(24): 1,581–7. 

13. Hacke W, Kaste M, Bluhmki E, et al.; ECASS investigators. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med, 2008 Sep 25; 359(13): 1,317–29. 

14. Mosley M. Is tPA Study Intentionally Deceptive? Emergency Medicine News, 2009 Aug; 31(8): 5–6; 

15.Wardlaw JM, Murray V, Berge E, Del Zoppo GJ. Thrombolysis for acute ischaemic stroke. Cochrane Database Syst Rev, 2009 Oct 7; (4): CD000213. 

16. American Academy of Neurology; American College of Emergency Physicians; American Heart Association/American Stroke Association. Tissue Plasminogen Activator (tPA): What You Should Know, 

17. Kummer BR, Lerario MP, Navi BB, et al. Clinical information systems integration in New York City’s first mobile stroke unit. Appl Clin Inform, 2018 Jan; 9(1): 89–98. 

18. Paramedic Chiefs of Canada. Webinar, Canada’s First Stroke Ambulance, 2018 Apr 4. 

19. Bowry R, Parker S, Rajan SS, et al. Benefits of stroke treatment using a mobile stroke unit compared with standard management: the BEST-MSU study run-in phase. Stroke, 2015; 46: 3,370–4. 

20. Nour M, Starkman S, Sharma LK, Saver JL. Magnitude of Benefit of Prehospital Mobile Stroke Unit vs. Conventional ED Thrombolysis: Preliminary Estimate Based on PHANTOM-S Observational Registry Study. International Stroke Conference 2017,!/4172/presentation/6519. 

21. Fiore K. Limited evidence for mobile stroke unit outcomes. Medpage Today, 

22. Wendt M, Ebinger M, Kunz A, et al.; STEMO Consortium. Improved prehospital triage of patients with stroke in a specialized stroke ambulance: results of the pre-hospital acute neurological therapy and optimization of medical care in stroke study. Stroke, 2015 Mar; 46(3): 740–5. 

23. John S, Stock S, Masaryk T, et al. Performance of CT angiography on a mobile stroke treatment unit: Implications for Triage. J Neuroimaging, 2016 Jul; 26(4): 391–4. 

24. Cerejo R, John S, Buletko AB, et al. A mobile stroke treatment unit for field triage of patients for intraarterial revasculatization therapy. J Neuroimaging, 2015 Nov–Dec; 25(6): 940–5. 

John Erich is senior editor for EMS World. Reach him at 

Back to Top