Performance of the RACE Prehospital Triage Score During Working and Nonworking Hours

Abstract

Background Prehospital triage scores aim to identify large vessel occlusions (LVOs) in the field; however, their real‐world performance and accuracy across a 24‐hour period remains unknown. In this study, we compare the positive predictive value of the prehospital Rapid Arterial occlusion Evaluation (RACE) score for the detection of LVO during working hours and nonworking hours. Methods We performed a retrospective review of all patients presenting with a RACE score of ≥5 at one comprehensive and one thrombectomy‐capable hospital between July 2015 and December 2019. Patients were dichotomized to those presenting during “working hours” (7:00 am to 6:00 pm on weekdays) or “nonworking hours” (6:01 pm to 6:59 am on weekdays or anytime during weekends). The primary outcome was diagnosis of LVO. Secondary outcomes included diagnosis of acute neurovascular syndromes, door‐to‐treatment time metrics, and a modified Rankin Scale of ≤2 in those undergoing mechanical thrombectomy. Results Of the 701 patients with RACE score ≥5, 687 patients were included (355 nonworking hours and 332 working‐hours cohorts). Mean age was 71 and 72 years in the nonworking hours and working hours groups, respectively ( P =0.13). Median National Institutes of Health Stroke Scale and baseline demographics were comparable between the 2 groups. There was no significant difference in the diagnosis of LVO (36.3% versus 34.6%; P =0.69) or final discharge diagnosis. The positive predictive value of the RACE score for the detection of an acute neurovascular syndrome (transient ischemic attack and stroke) was improved during nonworking hours (76.3% versus 67.8%; P =0.01). In patients undergoing mechanical thrombectomy, rates of good clinical outcome were similar (44.7% versus 48%; P =0.76), despite the shorter door‐to‐groin‐puncture and revascularization times during working hours. Conclusions The RACE score shows a consistent positive predictive value in determining LVO during working and nonworking hours. However, it is more accurate in determining acute neurovascular syndromes during nonworking hours, which is driven by a decrease in stroke mimics during this time.