Comparing National Institutes of Health Stroke Scale Among a Stroke Team and Helicopter Emergency Medical Service Providers

Background and Purpose-The use of tissue-type plasminogen activator is limited to a maximum of 4.5 hours after symptom-onset. Endovascular recanalization may improve outcomes for large-vessel occlusions (LVO), but efficacy decreases with time from symptom-onset. A National Institutes of Health Stroke Scale (NIHSS) score >= 12 is predictive of LVOs and could be used to triage patients if appropriately used by prehospital providers. The NIHSS has been considered too complex and has not been validated in the prehospital setting. Methods-We reviewed all patients with ischemic stroke transported by helicopter emergency medical services (HEMS) to a single comprehensive stroke center in 2010. HEMS NIHSS were compared with in-hospital stroke team physician scores. NIHSS was categorized based on 3 clinically relevant groupings and ability to predict LVO was investigated. Results-Three-hundred five patients met inclusion criteria, 68.9% having LVO. Moderate agreement existed between HEMS and physicians (72.1%; kappa=0.571). Interclass correlation was 0.879 (95% confidence interval, 0.849-0.904). Excluding patients with tissue-type plasminogen activator before HEMS transport, there were 216 patients and good agreement (82.7%; kappa=0.619). Among patients presenting within 8 hours postonset and NIHSS >= 12, HEMS had a sensitivity of 55.9% and positive predictive value of 83.7% in predicting LVO. Conclusions-HEMS providers can administer NIHSS with moderate to good agreement with the receiving stroke team. The use of the NIHSS in HEMS may identify patients with LVO and inform triage decisions for patients ineligible for tissue-type plasminogen activator.