Background: Elevations of glial fibrillary acidic protein (GFAP) and ubiquitin carboxyl-terminal hydrolase L1 (UCH-L1) have been associated with traumatic brain injury (TBI) and recently, the panel (both biomarkers) has been approved for determining the need for head computed tomography (HCT) among patients with mild to moderate TBI. The Canadian CT head rule (CCTHR) is a clinical decision instrument that aids clinicians in determining the need for HCT for patients with mild TBI. The goal of this study was to compare the biomarker panel’s performance to the CCTHR.
Methods: This is a secondary analysis of a prospective observational study that included adult patients with a suspected TBI (GCS 9-15) who underwent HCT and blood draw within 12 hours of injury. Patients selected for this study were those with a GCS 14-15 and met criteria for CCTHR determination. A similar analysis was done for patients with GCS 15 only. The primary outcome was a HCT showing intracranial injury. Test characteristics (exact 95% confidence interval [CI]) were determined.
Results: Of 1959 patients in the original study, 919 patients with GCS 14-15 met this study’s criteria of which 94.0% (864) had a GCS 15. The mean age was 43 (standard deviation 18), 61.3% were male, and 64.4% white. Falls accounted for 47.6% of injuries, 33.6% were motor vehicle crashes, and 11.1% were assault. A positive HCT was found in 67 (7.3%) patients. For the biomarker panel and CCTHR respectively; sensitivities were 95.5% (87.5% - 99.1%) vs. 70.1% (57.5% - 80.7%), specificity 38.8% (35.6% - 42.2%) vs. 55.5% (52.1% - 58.9%), and negative predictive values (npv) were 99.1% (97.4% - 99.8%) vs. 95.9% (93.8% - 97.5%) for patients with GCS 14-15. For patients with GCS 15 only, the sensitivity of the panel was 94.5% (84.9% - 98.9%) but the CCHR dropped to 63.6% (49.6% - 76.2%). In all instances, the biomarker panel was more sensitive than the CHCTR and it retained acceptable specificity and a high npv.
Conclusion: The GFAP/UCH-L1 biomarker panel was highly sensitive, had a high npv for a positive HCT finding, and outperformed the CCTHR indicating potential clinical utility of the panel. The lower sensitivity of the CCTHR may be, in part, due to this study not explicitly excluding small hemorrhages. Future work will include patients on anti-coagulants and those without amnesia or loss of consciousness.