Response by Hastrup and Andersen to Letter Regarding Article, “Prehospital Acute Stroke Severity Scale to Predict Large Artery Occlusion: Design and Comparison With Other Scales”
We agree that in-hospital angiographies cannot be replaced by neither National Institutes of Health Stroke Scale nor any simpler clinical scales. As documented by the authors, National Institutes of Health Stroke Scale is, despite a significant association with LVO invariably, less accurate than angiography,3 and therefore, it is unlikely that a simpler clinical scale derived from National Institutes of Health Stroke Scale could equal angiography.
But the question is how to identify patients with suspected LVO in the prehospital phase because this is a prerequisite of the mothership model, which involves direct referral of all patients suspected of LVO to a Comprehensive Stroke Center (CSC) with endovascular treatment (EVT). The mothership as an alternative to the drip-and-ship model, which involves referral to the nearest Primary Stroke Center with thrombolysis and subsequent transfer to a CSC, may shorten time to EVT in many cases. We agree that choosing one model over the other is not an easy choice, and it may vary depending on distances between stroke centers, door-in-door-out delays, and the ability to identify LVO prehospital.
In metropolis, a CSC is never far away, but the drip-and-ship model may still impose significant delays because of the door-in-door-out, but also transfer time could be long albeit physical distances are short. In a metropolis, such as London, originally built for horses, the speed of an emergency vehicle is slow, and there is an inherent risk for healthcare providers, patients, and other road users too.4
Initial evaluation of all suspected strokes in a CSC could be an option in metropolis, but the limited capacity has to be considered. In mixed areas, it is simply impractical to transport everyone over long distances to a CSC. Also, this strategy would impose undesirable delays in thrombolysis in the non-LVO patient.
Authors concern about LVO patients not identified by the scales, but these patients, who may risk being misclassified in the prehospital setting, will still be evaluated with angiography in the nearest Primary Stroke Center and transferred if indicated. Also, these patients are missed because of a benign clinical presentation,2 whereas EVT is best proved in severe strokes.
It is a limitation that these scales have not yet been tested prehospital. A recent cohort study indicated that prehospital evaluation of LVO patients is feasible and results in reduction of time to EVT without delaying thrombolysis for non-LVO patients.5
We think our study like the authors’ calculations confirms that the simpler Prehospital Acute Stroke Severity scale equals more complex clinical scales in prediction of LVO, and although not ideal, it has promising characteristics in triaging prehospital. At the moment, we have no alternatives for prehospital selection apart from mobile stroke units, which is only an option for urban areas with good infrastructure. We could potentially reduce delays in EVT in the majority of LVO patients by implementing the Prehospital Acute Stroke Severity scale to select patients for direct referral to a CSC, and it is a rational choice to continue testing this in the prehospital phase. In future, we also hope to have more accurate prediction tools in the field like biomarkers, but for the present-day patients with LVO, we need to try to improve the logistics now.
Sidsel Hastrup, MD
Grethe Andersen, MD, DMSc
Department of Neurology
Aarhus University Hospital
- © 2016 American Heart Association, Inc.
- Heldner MR,
- Heinrich HP,
- Fischer U
- Hastrup S,
- Damgaard D,
- Johnsen SP,
- Andersen G
- Heldner MR,
- Zubler C,
- Mattle HP,
- Schroth G,
- Weck A,
- Mono ML,
- et al