Clinical Scales Do Not Reliably Identify Acute Ischemic Stroke Patients With Large-Artery Occlusion

Abstract

Background and Purpose-It remains debated whether clinical scores can help identify acute ischemic stroke patients with large-artery occlusion and hence improve triage in the era of thrombectomy. We aimed to determine the accuracy of published clinical scores to predict large-artery occlusion. Methods-We assessed the performance of 13 clinical scores to predict large-artery occlusion in consecutive patients with acute ischemic stroke undergoing clinical examination and magnetic resonance or computed tomographic angiography ≤6 hours of symptom onset. When no cutoff was published, we used the cutoff maximizing the sum of sensitivity and specificity in our cohort. We also determined, for each score, the cutoff associated with a false-negative rate ≤10%. Results-Of 1004 patients (median National Institute of Health Stroke Scale score, 7; range, 0-40), 328 (32.7%) had an occlusion of the internal carotid artery, M1 segment of the middle cerebral artery, or basilar artery. The highest accuracy (79%; 95% confidence interval, 77-82) was observed for National Institute of Health Stroke Scale score ≥11 and Rapid Arterial Occlusion Evaluation Scale score ≥5. However, these cutoffs were associated with false-negative rates >25%. Cutoffs associated with an false-negative rate ≤10% were 5, 1, and 0 for National Institute of Health Stroke Scale, Rapid Arterial Occlusion Evaluation Scale, and Cincinnati Prehospital Stroke Severity Scale, respectively. Conclusions-Using published cutoffs for triage would result in a loss of opportunity for ≥20% of patients with large-artery occlusion who would be inappropriately sent to a center lacking neurointerventional facilities. Conversely, using cutoffs reducing the false-negative rate to 10% would result in sending almost every patient to a comprehensive stroke center. Our findings, therefore, suggest that intracranial arterial imaging should be performed in all patients with acute ischemic stroke presenting within 6 hours of symptom onset.

Predictive Value of Modifications of the Prehospital Rapid Arterial Occlusion Evaluation Scale for Large Vessel Occlusion in Patients with Acute Stroke

“…The RACE scale score was not detailed item-by-item in 16 cases and therefore we studied 341 patients (53.4% men; age (mean±SD) 70±13 years; NIHSS score at admission (median [interquartile range] 8 [3][4][5][6][7][8][9][10][11][12][13][14][15][16] had 85% sensitivity, 68% specificity and correctly classified 71% patients. The RACE and NIHSS scores were strongly correlated (r=0.78; p<0.001).…”

“…The RACE scale has been externally validated as one of the most accurate clinical scales to detect LVO (14).…”

Brazilian guidelines for endovascular treatment of patients with acute ischemic stroke

“…About one third of patients with AIS present with an occlusion of a large, proximal artery of the brain circulation 1 . Recently, endovascular treatment has been shown to improve functional outcome in five randomized clinical trials of selected patients with AIS associated with PAO of the anterior circulation 2, 3,4,5,6,7 .…”

“…In the Interventional Management of Stroke III trial (IMS-3), patients were mainly selected based on an NIHSS ≥10, and one third of them eventually did not have a target occlusion on catheter angiography. Cut-off values of NIHSS scores with the highest accuracy for detecting PAO (scores from 10 to 14) still present false-negative rates >20%, and scores with sufficient sensitivity to detect >90% of patients with PAO (NIHSS from 1 to 5) would result in sending almost every patient to catheter angiography 1,24 . Therefore, vessel imaging by CTA or MRA is highly recommended for endovascular treatment indication 23 .…”

Stroke and the emergency medical services – enhancing performance within the chain of survival

“…The highest accuracy has been reached with NIHSS and RACE, but the suggested cut-off values would still result in a loss of opportunity for over 20% of patients. (Turc et al 2016) Although HEMS paramedics have been able to conduct the NIHSS in the prehospital setting with a moderate to good agreement with stroke team physicians (Kesinger et al 2015), using this scale requires considerable training and skill. Currently, the use of prehospital stroke recognition tools is recommended by both the National Academy of Emergency Medical Service Physicians and the American Stroke Association.…”