Pre-hospital Diagnosis of Large Vessel Occlusion in Patients Being Transported by Helicopter.

Background: During the helicopter transportation of patients suspected of large vessel occlusion (LVO), an accurate and rapid decision-making process is required.
Aims: We attempted to create an algorithm for the pre-hospital diagnosis of the presence of LVO in patients suspected of stroke using data from patients transported urgently by helicopter.
Methods: One hundred and sixty-five patients transported by helicopter were divided into two subgroups: a training dataset and a validation dataset. We extracted clinical information obtained on site, the unadjusted score of the National Institutes of Health Stroke Scale, and previously reported pre-hospital scales as an LVO screen. On the basis of the analyses of these factors, an algorithm was devised to predict the presence of LVO and its predictive accuracy was evaluated using the validation dataset.
Results: Ischemic stroke with LVO was diagnosed in 36 out of 121 cases (29.8%) in the training dataset and in 10 out of 44 cases (22.7%) in the validation dataset. Combining five factors (conjugate deviation, upper limb paresis, atrial fibrillation, Japan Coma Scale ≥ 200, and systolic blood pressure ≥ 180), an algorithm was created to classify cases into six groups with different likelihoods of LVO presence. The algorithm predicted correctly 6 out of 10 cases in the validation dataset. Furthermore, it definitively ruled out 17 out of 34 cases in the validation dataset.
Conclusions: Using the newly created algorithm, emergency staff could easily and accurately distinguish patients suitable for urgent endovascular thrombectomy from patients with non-LVO or stroke mimics.
(Copyright © 2020 Elsevier Inc. All rights reserved.)