Abstract WP330: Refining Prehospital Stroke Severity Measures: Is There Added Benefit to Combining Field Glasgow Coma Scale and Los Angeles Motor Score?

• Published in: Stroke (1970) Vol. 50; no. Suppl_1
• Main Authors: Cheung, Shauna, Shkirkova, Kristina, Liebeskind, David, Sabra, Mark, Starkman, Sidney, Hamilton, Scott, Saver, Jeffery L, Sanossian, Nerses
• Format: Journal Article
• Language: English
• Published: 01.02.2019
• ISSN: 0039-2499; 1524-4628
• DOI: 10.1161/str.50.suppl_1.WP330

Abstract only Background: Prehospital research requires severity screening instruments that can be reliably performed in the field and that are validated for predicting 3-month outcome. The Los Angeles Motor Scale and Glasgow coma scale are regularly performed in the field. The LAMS, a pure motor scale typically used to screen for LVO, has been validated as a stroke severity measure which correlates well with 90-day outcome. Objective: to determine if the combination of GCS+LAMS better predicts stroke severity at presentation and long-term outcomes than either instrument alone. Methods: All subjects were participating in the FAST-MAG clinical trial. LAMS and GCS measures were obtained by paramedics in the field. NIHSS was obtained by study nurses after hospital arrival, mRS was obtained at 90 days. The composite LAMS+GCS was calculated by adding one point for every point lost on the GCS (1-12 possible) to the LAMS score (0-10 possible). Receiver operating characteristics were calculated for 90-day outcome LAMS, GCS and LAMS+GCS with determination of best fit. Correlation was calculated for ED NIHSS. Results: Among 1700 patients (age 69±13 years, female 42.6%) 73.3% of patients had final diagnosis of cerebral ischemia, 22.8% intracranial hemorrhage, and 3.9% stroke mimic. Time from onset to prehospital LAMS was median 30 minutes (IQR 20-50). Predictive accuracy (C statistics) for poor 3-month outcome (mRS 3-6 at day 90) in all patients was 0.69 for LAMS+GCS, 0.68 for LAMS and 0.59 for GCS. Best fit for determining poor outcome was LAMS+GCS >=5 (Sensitivity 64%, Specificity 67%), LAMS >= 4 (Sensitivity 73%, specificity 56%), and GCS<=14 (92%, 45%). Correlation to hospital arrival NIHSS was strongest for the combined measure (R= 0.529) followed by LAMS (R=0.431) and GCS (R=-0.401). Among ICH cases (N=386) LAMS+GCS performed better than LAMS or GCS alone (C-statistic 0.693 vs. 0.642 vs. 0.600). Conclusions: Adding GCS to LAMS in the prehospital assessment does not improve its ability to identify patients who are likely to have poor outcomes. LAMS+GCS may be a better measure in ICH cases or in predicting initial prehospital stroke severity, capturing non-motor deficits. LAMS alone can be utilized as a stroke severity instrument in prehospital clinical trials.