Resident Eagle: The Failed Promise of LVO Stroke Scales

Resident Eagle is a monthly column profiling the work of top EMS physicians and medical directors from the Metropolitan EMS Medical Directors Global Alliance (the "Eagles"), who represent America’s largest and key international cities. Tentative dates for Gathering of Eagles 2021: June 14–18, Hollywood, Fla. For more see useagles.org.

Over the past few years, the landscape of acute ischemic stroke (AIS) treatment has undergone a significant transformation. It has evolved from TPA as the only viable emergent intervention into endovascular mechanical thrombectomy therapies for AIS with large vessel occlusion (LVO). Five landmark trials published in 2015 not only demonstrated significant outcome advantages with endovascular intervention for LVO-AIS (with just 2–4 patients needing treatment to have one survive with minimal to no disability) but also highlighted the time-dependent nature of endovascular intervention outcome benefits.^1–5 

With these treatment advances, coupled with the fact that not all hospitals have neuroendovascular intervention capabilities, there was the obvious desire to develop LVO “presentation markers” in the prehospital environment so EMS providers could determine which patients would potentially require endovascular intervention. Enter the LVO scales, developed in hopes of providing accurate prehospital identification of LVOs. Despite great initial enthusiasm, they are not proving to be quite the definitive tool we wanted. In all there are a half-dozen widely used LVO scales (RACE/RACE+, C-STAT [CPSSS], VAN, LAMS, FAST-ED, NIHSS) and many more smaller configurations, which in and of itself suggests they are not working as hoped. 

These scales largely have similar sensitivity, specificity, and positive and negative predictive values. Therefore, the choice of which scale to use, or if to use one at all, is usually based on idiosyncrasies of an individual community and what works best for its stroke systems of care.

Scales and Their Limitations

A June 2020 survey of 40 members of the Eagles Coalition found no consensus on whether to use an LVO scale or which one to use. Of those responding, 74% utilize an LVO scale. Their specific scale usage is 20% VAN; 20% RACE or a variation; 17% C-STAT; 17% LAMS; 10% FAST-ED; 7% CPSS; and 9% locally derived scales. Of those utilizing LVO scales, 60% use them to influence transport destinations. 

One fundamental issue with LVO scales is that they are by and large derived retrospectively and validated on the same data sets used to derive them. In these retrospective cohort studies, common clinical presentation features are identified from a cohort of patients who already have a CT-verified AIS-LVO, and then these features are taken into the prehospital setting and prospectively applied to all potential stroke patients to presumptively identify AIS-LVO and help determine who likely needs to bypass closer hospitals for one that can perform neuroendovascular intervention. 

Because these clinical characteristics were derived from patients with known AIS-LVO, many other “undifferentiated” neurodeficit-presenting illnesses that can mimic strokes (e.g., non-LVO AIS, TIA, intracranial bleed, and more) were not considered in the evaluation data set. So when EMS providers apply LVO scales to an undifferentiated patient population in the field, many non-LVO patients will ultimately receive a stroke assessment and potentially score positive on an LVO scale. This results in the failure of the LVO scales to perform as well as they do in their derivation and validation studies.

Hindering, Not Helping

El Paso County, Colo. has a population of 720,000, with a consistently and unusually high incidence of strokes. In 2017 all EMS agencies in El Paso County were trained in and implemented the C-STAT LVO scale, with a change in our destination guidelines to bypass our closer primary stroke centers (PSCs) for thombectomy-capable centers (TCCs). 

Within a few months we recognized our C-STAT data showed moderately lower sensitivity and specificity than its derivation and validation studies suggested. Believing this must indicate system deficiencies, we implemented changes, including LVO scale training directly from local neurointerventional specialists; standardization of stroke assessment tools; and an overhaul on coordinated EMS-to-hospital communication. 

The subsequent two years of data collection showed no significant movement in C-STAT specificity, indicating that the use of the LVO scale was actually hindering, rather than helping, decisions for appropriate destination guidelines and resource utilization.

In retrospective removal of C-STAT data from non-AIS neurodeficit-presenting patients, our C-STAT sensitivity and specificity almost exactly matched that of the LVO scale’s derivation and validation studies. Since we felt there was little way to prospectively remove non-AIS patients from the prehospital undifferentiated neurodeficit patient cohort, in late 2019 El Paso County ceased utilization of an LVO scale as a tool in acute stroke systems of care. 

Another issue with LVO scale use to determine destinations is that the patients who qualify for and receive neurointerventional thrombectomy represent an exceedingly small proportion of all stroke patients: Only an estimated 4%–9% of all stroke patients receive the procedure.^6–8

One study suggests preferential triage of prehospital suspected stroke patients using C-STAT would increase the number of patients transported to TCCs by 11% within six hours and an additional 10% from 6–24 hours. For every patient with LVO as final diagnosis, approximately six additional non-LVO patients would be triaged to a TCC, potentially bypassing other hospitals.9 

Conclusion

Systems must consider whether bypassing hospitals and prolonging ground transport times may pull resources from communities and leave areas with delayed 9-1-1 responses. We all anxiously await the RCT that produces and validates an LVO scale or other tool that identifies AIS-LVO patients with the needed sensitivity and specificity to clearly justify its ramifications on the entire prehospital system. Until then, prehospital stroke notification systems utilizing EMS impressions and stroke severity tools such as LVO scales lack the specificity required for modern acute stroke systems of care. Better solutions must be explored so prehospital notification can keep pace with advances in acute stroke treatment.¹⁰  

Sidebar: 2020 Eagles Award Winners

Each year the Metropolitan EMS Medical Directors Global Alliance, better known as the Eagles, bestows four key awards honoring excellence in aspects of the EMS/emergency medicine field. This year’s winners are:

Ron J. Anderson Award: Jon Krohmer, MD, director, NHTSA Office of EMS—Named for the legendary CEO of Texas’ Parkland Health & Hospital System, the Anderson Award honors commitment to public service. 

Michael K. Copass Award: Libby Char, MD, medical director, AMR Hawaii, Honolulu Fire, Kauai Fire, Maui Fire, Honolulu Ocean Safety—The Copass Award is given to an EMS medical director who has demonstrated longstanding service, contributions, and leadership in out-of-hospital emergency care and served as a role model for EMS personnel and fellow 9-1-1 system medical directors.

Paul E. Pepe Award: Michael Osterholm, MD, director, Center for Infectious Disease Research and Policy—The Pepe Award recognizes outstanding national contributions to emergency medical services. 

Corey Slovis Award: Ken Scheppke, MD, medical director, state of Florida, West Palm Beach, Palm Beach County—The Slovis Award rewards excellence in education. 

References

1. Berkhemer OA, Fransen PSS, Beumer D, et al.; MR CLEAN investigators. A Randomized Trial of Intraarterial Treatment for Acute Ischemic Stroke. N Engl J Med, 2015 Jan 1; 372(1): 11–20.

2. Goyal M, Demchuk AM, Menon BK, et al.; ESCAPE Trial investigators. Randomized Assessment of Rapid Endovascular Treatment of Ischemic Stroke. N Engl J Med, 2015 Mar 12; 372(11): 1,019–30.

3. Saver JL, Goyal M, Bonafe A, et al.; SWIFT PRIME investigators. Stent-Retriever Thrombectomy After Intravenous t-PA vs. t-PA Alone in Stroke. N Engl J Med, 2015 Jun 11; 372(24): 2,285–95.

4. Jovin TG, Chamorro A, Cobo E, et al.; REVASCAT Trial investigators. Thrombectomy Within 8 Hours After Symptom Onset n Ischemic Stroke. N Engl J Med, 2015 Jun 11; 372(24): 2,296–306.

5. Campbell BCV, Mitchell PJ, Kleinig TJ, et al.; EXTEND-IA investigators. Endovascular Therapy for Ischemic Stroke With Perfusion-Imaging Selection. N Engl J Med, 2015 Mar 12; 372(11): 1,009–18.

6. Zaidat OO, Lazzaro M, McGinley E, et al. Demand-supply of neurointerventionalists for endovascular ischemic stroke therapy. Neurology, 2012; 79(13 Suppl 1): S35–41.

7. Saver JL, Levy E, McDougall CG, et al. Planning for nationwide endovascular acute ischemic stroke care in the united states: report of the interventional stroke workforce study group. The Stroke Interventionalist, 2012; 1(1): 19–24.

8. Jadhav AP, Desai SM, Kenmuir CL, et al. Eligibility for Endovascular Trial Enrollment in the 6- To 24-Hour Time Window: Analysis of a Single Comprehensive Stroke Center, Stroke, 2018 Apr; 49(4): 1,015–7.

9. Li JL, McMullan JT, Sucharew H, et al. Potential Impact of C-STAT for Prehospital Stroke Triage Up to 24 Hours on a Regional Stroke System. Prehosp Emerg Care, 2020 Jul–Aug; 24(4): 500–4.

10. English SW, Rabinstein AA, Mandrekar J, Klaas JP. Rethinking Prehospital Stroke Notification: Assessing Utility of Emergency Medical Services Impression and Cincinnati Prehospital Stroke Scale. J Stroke Cerebrovasc Dis, 2018 Apr; 27(4): P919–25.

E. Stein Bronsky, MD, is  co-medical director for the Colorado Springs Fire Department and Colorado Springs and El Paso County AMR in Colorado, as well as medical director for the El Paso–Teller County 9-1-1 Authority. He is an emergency medicine physician with Centura Health/U.S. Acute Care Solutions. 

Matthew Angelidis, MD, is an ER physician at UCHealth and medical director for UCHealth EMS, as well as co-medical director for the Colorado Springs Fire Department and Colorado Springs AMR.