• Should Endovascular Therapy Be Centralized Regionally?

    Theoretical study says yes, but practical concerns loom large

    Action Points

    • Delivery of endovascular therapy (EVT) for stroke patients with suspected large vessel occlusion (LVO) should be centralized regionally, according to a modeling study.
    • Note that if treatment times are slow at the thrombolysis center, patient transport systems should consider bypassing it if the two centers are 60 minutes apart or less whereas if travel time is more than 60 minutes between centers, bypassing it is not always the best option.

    Delivery of endovascular therapy (EVT) for stroke patients with suspected large vessel occlusion (LVO) should be centralized regionally, a theoretical, conditional probability model suggested.

    In a decision-making system derived from published clinical trial data, the choice to bypass a nearby thrombolysis center and go directly to a thrombectomy-capable hospital for EVT depended on the patient's distance to each center, treatment speed, and the accuracy of LVO screening tools, wrote Jessalyn Holodinsky, of the University of Calgary in Canada, and co-authors in JAMA Neurology.

    If treatment times are slow at the thrombolysis center, patient transport systems should consider bypassing it if the two centers are 60 minutes apart or less. If travel time is more than 60 minutes between centers, bypassing it is not always the best option.

    "Pre-hospital decision making for ischemic stroke patients with suspected large vessel occlusion is complex and context-specific," Holodinsky told MedPage Today. "There is not one bypass rule that will be appropriate for every hospital system."

    Treating ischemic stroke with LVO can be done with intravenous alteplase (Activase) and/or endovascular therapy (EVT), but each treatment is time sensitive. EVT is more effective for stroke patients with LVO, but often, only urban tertiary centers offer it.

    Prior studies have shown that patients who received EVT with long transfer delays between hospitals have worse outcomes than patients without transfers. And while neurovascular imaging is the standard to determine EVT eligibility, high-quality imaging in the field -- such as a mobile stroke unit with CT angiography -- is not available in many areas.

    Holodinsky's group sought to model the best transport strategies for acute stroke with suspected LVO, balancing the benefit of early alteplase treatment, the greater efficacy of EVT, and the declining benefits of both treatments over time.

    The researchers used data from existing clinical trials to generate the model. They assumed stroke onset time was known and transport decisions would be made after an emergency medical services (EMS) evaluation with an LVO screening tool. They visualized scenarios in California and looked at two transport strategies: direct to the endovascular center (mothership) or immediate alteplase treatment followed by transfer to the endovascular center (drip and ship).

    The patient's travel time, speed of treatment, and positive predictive value of the screening tool in each scenario affected which strategy had the best outcomes. The transport time threshold for bypassing the thrombolysis center was especially pronounced when door-to-needle time at thrombolysis centers was 60 minutes (door-in, door-out time of 120 minutes), the authors reported.

    "This is the current reality in many stroke systems," they noted. Hospitals in the Get With The Guidelines Target-Stroke program had a post-intervention median door-to-needle time of 67 minutes, they added. "Our results imply that, based on current treatment times, delivery should be regionally centralized to realize the full benefit of EVT on a population basis."

    Although time from stroke onset to reperfusion is the goal, "little progress is possible without reform of the EMS system and the introduction of standardized approaches to pre-hospital acute stroke care," observed Lee Schwamm, MD, of Massachusetts General Hospital in Boston, in an accompanying editorial.

    "While we may be able to shorten some of these time intervals, we must acknowledge that any decision algorithm will require a high degree of training for EMS paraprofessionals, use of adjunctive technology to execute complex algorithms, and ultimately must balance prioritizing the approximately 20% of patients who will have a stroke because of LVO versus the approximately 80% of stroke patients who do not."

    Geographic disparities in resources and population density make it unlikely any one approach will work universally, and part of the challenge is that acute stroke is rare in pre-hospital care.

    "Acute stroke makes up less than 5% of all EMS transports, less than 5% of all emergency department visits, and less than 5% of all hospital admissions," Schwamm pointed out. "When you factor in that strokes arising from LVO are only approximately 20% of all strokes, these numbers drop to close to 1%. This low prevalence means that the a priori positive predictive power of any human-based screening tool will be low, and we will need to decide the minimum acceptable threshold for this value."

    This theoretical modeling study has limitations, the authors noted: it assumed treatment outcomes were time invariant, but patients with intracerebral hemorrhage may require the higher level of care available at EVT centers like neuro-intensive care units and neurosurgical teams, and that care might not be time invariant. The results of this study apply to the system design level only and are not meant to predict outcomes on an individual patient basis, Holodinsky added.

    Holodinsky disclosed funding from Alberta Innovates and Quality Improvement and Clinical Research (QuICR) Alberta Stroke Program.

    Co-authors disclosed relevant relationships with Medtronic, Stryker, Merck, Hoffmann-La Roche Canada, Boehringer-Ingelheim, NoNO, Alberta Innovates Health Solutions, CIHR, and Calgary Scientific.

    Schwamm disclosed relevant relationships with Genentech, LifeImage, Lundbeck, Penumbra, and Medtronic.

    Source:

    JAMA Neurology

    Source Reference: Holodinsky J, et al "Modeling stroke patient transport for all patients with suspected large-vessel occlusion" JAMA Neurology 2018; DOI: 10.1001/jamaneurol.2018.2424.

    JAMA Neurology

    Source Reference: Schwamm L "Optimizing prehospital triage for patients with stroke involving large vessel occlusion: The road less traveled" JAMA Neurology 2018.

     

    Read the original article on Medpage Today: Should Endovascular Therapy Be Centralized Regionally?

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Review our Privacy Policy for more details