Diagnostic Performance of GFAP, UCH-L1, and MAP-2 Within 30 and 60 Minutes of Traumatic Brain Injury

Importance Data on the performance of traumatic brain injury (TBI) biomarkers within minutes of injury are lacking. Objectives To examine the performance of glial fibrillary acidic protein (GFAP), ubiquitin carboxy-terminal hydrolase L1 (UCH-L1), and microtubule-associated protein 2 (MAP-2) within 30 and 60 minutes of TBI in identifying intracranial lesions on computed tomography (CT) scan, need for neurosurgical intervention (NSI), and clinically important early outcomes (CIEO). Design, Setting, and Participants This cohort study is a biomarker analysis of a multicenter prehospital TBI cohort from the Prehospital Tranexamic Acid Use for TBI clinical trial conducted across 20 centers and 39 emergency medical systems in North America from May 2015 to March 2017. Prehospital hemodynamically stable adult patients with traumatic injury and suspected moderate to severe TBI were included. Blood samples were measured for GFAP, UCH-L1, and MAP-2. Data were analyzed from December 1, 2023, to March 15, 2024. Main Outcomes and Measures The presence of CT lesions, diffuse injury severity on CT, NSI within 24 hours of injury, and CIEO (composite outcome including early death, neurosurgery, or prolonged mechanical ventilation ≥7 days) within 7 days of injury. Results Of 966 patients enrolled, 804 patients (mean SD age, 41 19 years; 418 74.2% male) had blood samples, including 563 within 60 minutes and 375 within 30 minutes of injury. Among patients with blood drawn within 30 minutes of injury, 212 patients (56.5%) had CT lesions, 61 patients (16.3%) had NSI, and 112 patients (30.0%) had CIEO. Among those with blood drawn within 60 minutes, 316 patients (56.1%) had CT lesions, 95 patients (16.9%) had NSI, and 172 patients (30.6%) had CIEO. All biomarkers showed significant elevations with worsening diffuse injury on CT within 30 and 60 minutes of injury. Among blood samples taken within 30 minutes, GFAP had the highest area under the receiver operating characteristic curve (AUC) to detect CT lesions, at 0.88 (95% CI, 0.85-0.92), followed by MAP-2 (AUC, 0.78; 95% CI, 0.73-0.83) and UCH-L1 (AUC, 0.75; 95% CI, 0.70-0.80). Among blood samples taken within 60 minutes, AUCs for CT lesions were 0.89 (95% CI, 0.86-0.92) for GFAP, 0.76 (95% CI, 0.72-0.80) for MAP-2, and 0.73 (95% CI, 0.69-0.77) for UCH-L1. Among blood samples taken within 30 minutes, AUCs for NSI were 0.78 (95% CI, 0.72-0.84) for GFAP, 0.75 (95% CI, 0.68-0.81) for MAP-2, and 0.69 (95% CI, 0.63-0.75) for UCH-L1; and for CIEO, AUCs were 0.89 (95% CI, 0.85-0.93) for GFAP, 0.83 (95% CI, 0.78-0.87) for MAP-2, and 0.77 (95% CI, 0.72-0.82) for UCH-L1. Combining the biomarkers was no better than GFAP alone for all outcomes. At GFAP of 30 pg/mL within 30 minutes, sensitivity for CT lesions was 98.1% (95% CI, 94.9%-99.4%) and specificity was 34.4% (95% CI, 27.2%-42.2%). GFAP levels greater than 6200 pg/mL were associated with high risk of NSI and CIEO. Conclusions and Relevance In this cohort study of prehospital patients with TBI, GFAP, UCH-L1, and MAP-2 measured within 30 and 60 minutes of injury were significantly associated with traumatic intracranial lesions and diffuse injury severity on CT scan, 24-hour NSI, and 7-day CIEO. GFAP was the strongest independent marker associated with all outcomes. This study sets a precedent for the early utility of GFAP in the first 30 minutes from injury in future clinical and research endeavors.