Blood levels of glial fibrillary acidic protein (GFAP) identified mild traumatic brain injury (TBI) in patients even when a CT scan did not detect it, the TRACK-TBI study showed. In people with negative CT findings, GFAP levels distinguished patients who were MRI-positive for intracranial lesions from patients who were MRI-negative, reported Geoffrey Manley, MD, of the University of California San Francisco, and co-authors, in Lancet Neurology. The best performance -- when the area under the receiver operating curve (AUC) was 0.852 -- was in samples taken 9 to 16 hours after injury. GFAP is expressed almost uniquely in astrocytes and could help identify patients with concussion who might otherwise go undiagnosed, Manley and colleagues noted. "Having these sensitive tools could provide physicians more real-time, objective information and improve the accuracy of detecting TBI," Manley said in a statement. "This research shows that blood tests have the potential to help physicians triage patients suspected of brain injury quickly and accurately." In 2018, the FDA approved a test using GFAP and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1) to help guide whether CT scans were needed in patients with suspected TBI. Previous studies had also demonstrated that GFAP alone can identify, with reasonable accuracy, patients with mild TBI. Another option for detecting the condition in CT-negative individuals is MRI scans, as a 2012 study suggested, but a blood test would clearly be preferable. TRACK-TBI is a prospective cohort study of individuals with head injuries who received CT scans within 24 hours of injury at level 1 trauma centers in the U.S. In this analysis, TRACK-TBI researchers studied 450 patients who had normal CT findings, were fully conscious or had minimal impairment (Glasgow Coma Scale scores of 13 to 15), and had MRI scans 7 to 18 days post-injury. MRI scans were negative in 330 patients and positive in 120; positive MRI results served as the study's gold standard for detecting mild TBI. The investigators compared data from these 450 patients (mean age 36; 63% men) against 122 orthopedic trauma controls and 209 healthy controls. The average time from injury to venipuncture for GFAP measurement was 12.1 hours, and the researchers measured GFAP concentrations with a prototype assay that generated results within 15 minutes. GFAP discriminated CT-/MRI+ from CT-/MRI- findings with an AUC of 0.777 (95% CI 0.726–0.829) over 24 hours. This increased to 0.852 (95% CI 0.781-0.923) when GFAP was measured 9 to 16 hours after injury. Median GFAP was highest in CT-/MRI+ patients (mean 414.4 pg/mL). This was followed by CT-/MRI- patients (74.0 pg/mL), orthopedic trauma controls (13.1 pg/mL), and healthy controls (8.0 pg/mL; all P<0.0001). Since the study was not intended to qualify the test for marketing, however, the researchers didn't identify an optimal GFAP cutoff value. Among the 120 head injury patients who had positive MRI findings, those with diffuse axonal injury had significantly higher plasma GFAP concentrations than those with traumatic axonal injury. "These results are promising but, as the investigators acknowledge, are some distance from suggesting that GFAP assays should become part of current emergency department management of patients with TBI and negative CT scans," noted Fiona Lecky, PhD, of the University of Sheffield in England, in an accompanying editorial. The clinical significance of positive MRI findings is unknown, and data about patient symptom severity at 2 weeks or later were not reported in the study, Lecky pointed out. "If the frequencies of disabling concussion symptoms do not differ in the MRI-positive and MRI-negative cohorts during follow-up, it is hard to argue that either GFAP or the MRI scans are providing key information to guide further management," she observed. But these shortcomings are not necessarily fatal flaws, she added: the EU-funded Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI) study has used a similar approach in European patients. "If these findings from the TRACK-TBI study can be replicated -- and their clinical significance demonstrated -- then TBI characterization with GFAP could lead to improved patient outcomes," she wrote. One potential downside cited by Manley and colleagues was that the test's best performance was with samples taken 9 to 16 hours after injury, a less than optimal delay. The study also did not compare GFAP differences in isolated mild TBI versus polytrauma and did not follow patients beyond 2 weeks. Source
