Statewide Emergency Medical Services Protocols for Suspected Stroke and Large Vessel Occlusion

Prehospital triage is critically important in the care of patients with stroke caused by large vessel occlusion (LVO), both because of the time-sensitive nature of acute interventions and the need to appropriately use available resources.¹ However, it is unclear if standardized LVO-specific triage protocols exist among emergency medical services (EMS) nationwide.

We performed a cross-sectional analysis of publicly available statewide EMS protocols in December 2020 using online searches cross-referenced to previous literature² to characterize prehospital LVO transport algorithms across the US. We included states with mandated or recommended protocols as well as those with relevant state department of health–issued guidelines. Other states were excluded, even if they had protocols on a regional level.

After identifying 35 states with publicly available adult stroke EMS protocols, we determined the frequency of LVO-specific transport algorithms, use of LVO screening tools, criteria for determining transport to an alteplase-capable center (ACC) vs a thrombectomy-capable center (TCC), and specific time-based cutoffs considered in transport decision-making (ie, time from last known well and maximum acceptable delay to bypass an ACC for a TCC). All protocols were independently reviewed by C.C.C. and T.J.M., and there were no cases of disagreement.

Only 16 states included specific transport considerations addressing suspected LVO cases, with 5 others suggesting following regional algorithms (Figure). Five different LVO screening tools were recommended: Rapid Arterial Occlusion Exam (RACE; 4 states), Field Assessment Stroke Triage for Emergency Destination (FAST-ED; 4 states), Cincinnati Stroke Triage Assessment Tool (C-STAT; 3 states), Los Angeles Motor Scale (LAMS; 2 states), and visual, aphasia, and neglect assessment (VAN; 1 state). Two states did not specify an LVO screening tool.

Among the 16 states with LVO transport algorithms, 9 recommended bypassing ACCs in favor of TCCs for potentially alteplase-eligible patients, while 6 recommended transport of alteplase-eligible patients to the nearest ACC and alteplase-ineligible patients to a TCC; 1 state defers transportation decisions of alteplase-eligible patients to potential receiving hospitals (Table). Among states that recommended bypassing ACCs for TCCs, 7 did not indicate a maximum acceptable transportation delay, while the others allowed maximum delays ranging from 15 to 60 minutes. Three states (Kentucky, Nevada, and West Virigina) explicitly prioritized comprehensive stroke centers over other TCCs, and 5 states (Iowa, Idaho, Nevada, Washington, and West Virginia) suggested air medical transport to expedite hospital arrival.

According to a recent consensus statement, it is now recommended that patients with suspected LVO should be preferentially triaged to a TCC if within a certain range, depending on geographical setting.^3,4 Specific prehospital procedures to identify such patients are also recommended. Nevertheless, our study found that most states do not have specific prehospital guidelines on the transport of patients with suspected LVO and that there is substantial variability in the states that do have such protocols.

Part of this variability may be because of the relatively unreliable nature of existing LVO screening tools,⁵ although they may still improve outcomes with appropriate training and implementation. Our study also found substantial variability in last known well cutoffs and acceptable destination delays for suspected LVO cases, some of which conflict with American Heart Association recommendations.³ Given the importance of EMS triage in the early care of patients with stroke,⁶ there is an urgent need for pragmatic trials to determine optimal prehospital LVO triage practices.

Because our study was limited to states with uniformly adopted guidelines, we may have undercounted protocols in the remaining states in which guidelines are adopted heterogeneously at the county or local level. Additionally, regional differences in health care infrastructure and resource availability may have impacted specific protocols and their implementation. Finally, we were unable to determine adherence to published protocols and real-world prehospital triage practices, which should be addressed in future studies using available EMS tracking databases.

Accepted for Publication: July 9, 2021.

Published Online: September 20, 2021. doi:10.1001/jamaneurol.2021.3227

Corresponding Author: Michael E. Reznik, MD, Division of Neurocritical Care, Department of Neurology, Rhode Island Hospital, 593 Eddy St, APC 712, Providence, RI 02903 (michael_reznik@brown.edu).

Author Contributions: Mr Chuck had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Chuck, Martin, Kalagara, Yaghi, Reznik.

Acquisition, analysis, or interpretation of data: Chuck, Martin, Kalagara, Madsen, Furie, Reznik.

Drafting of the manuscript: Chuck, Kalagara, Reznik.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Chuck, Martin, Kalagara, Reznik.

Administrative, technical, or material support: Martin.

Study supervision: Martin, Reznik.

Conflict of Interest Disclosures: Dr Madsen has received grants from the National Heart, Lung, and Blood Institute. No other disclosures were reported.

Additional Contributions: We thank Daniel Kim, BS, and Owen Leary, BS (Alpert Medical School, Brown University, Providence, Rhode Island), for their work in contributing to the drafting and preparation of the manuscript. We also thank Matthew Anderson, MD, and Joshua Feler, MD (Alpert Medical School, Brown University), for their help providing technical feedback. None of the contributors were compensated for their work.