Summary: Blood levels of two protein biomarkers—GFAP and UCH-L1—measured within 24 hours of a traumatic brain injury (TBI) can help predict which patients are at highest risk of death or severe disability six months later.
Source: UCSF
Blood tests taken within 24 hours of a traumatic brain injury (TBI) can indicate which patients are likely to die and which are likely to survive with severe disabilities, according to a large multicenter study led by researchers at UC San Francisco, the University of Pennsylvania and the University of Michigan.
These rapid tests, which return results within minutes, may help clinicians decide on urgent surgical interventions, improve triage in mass-casualty events, and provide clearer, earlier guidance for conversations with families facing devastating injuries.
The assays measure two brain-specific proteins: glial fibrillary acidic protein (GFAP), found in glial cells, and ubiquitin C-terminal hydrolase L1 (UCH-L1), found in neurons. Both biomarkers were previously cleared by the U.S. Food and Drug Administration for identifying mild TBI patients who should receive CT scans. This study, published in The Lancet Neurology, evaluates the prognostic performance of these proteins when measured on the day of injury.
Co-senior author Geoffrey Manley, MD, PhD, professor and vice chair of neurosurgery at UCSF, emphasized that the tests are diagnostic and prognostic, and that they are simple, rapid, and cost-effective to administer at the bedside.
Researchers analyzed day-of-injury blood samples from 1,696 TBI patients enrolled in TRACK-TBI, a multisite observational cohort led by UCSF. Samples were tested using the Abbott i-STAT Alinity portable analyzer and ARCHITECT laboratory assays. Outcomes were assessed at six months using the Glasgow Outcome Scale–Extended (GOSE), which ranges from death (level 1) to full recovery (level 8).
One in five patients died or experienced severe disability
Participants were evaluated at 18 Level 1 trauma centers after injuries primarily caused by traffic accidents and falls. The cohort was two-thirds male with an average age of 39. At six months, 7% of patients had died and 14% had an unfavorable outcome (GOSE levels 2–4), including vegetative state and severe disability requiring daily support. Another 67% showed incomplete recovery, including moderate disabilities or persistent neurological or psychological deficits. One-third of patients achieved full recovery (GOSE level 8).
Day-of-injury GFAP and UCH-L1 measurements had strong predictive accuracy for death and severe disability at six months. The area under the receiver operating characteristic curve (AUC) for predicting death was 0.87 for GFAP and 0.89 for UCH-L1; for predicting unfavorable outcome the AUCs were 0.86 for both biomarkers. However, both markers were less accurate at distinguishing incomplete recovery from full recovery (AUCs ~0.61–0.62).
“These biomarkers can encourage clinicians to act more decisively when life-saving interventions are warranted,” said Manley. “Contemporary trauma care can produce good outcomes even in injuries once thought unsurvivable.”
Beyond informing individual patient care, rapid biomarker testing could aid triage and resource allocation in large-scale incidents, when quick, objective measures of prognosis are especially valuable.
Frederick Korley, MD, PhD, the study’s first author and an associate professor of emergency medicine at the University of Michigan, noted this is the first large-scale report assessing how well day-of-injury GFAP and UCH-L1 predict neurological recovery at six months. Earlier studies were smaller and produced less precise estimates.
Combining biomarkers with clinical models improves accuracy
The investigators also tested whether adding biomarkers to established clinical prognostic models would improve prediction. In patients with moderate to severe TBI (Glasgow Coma Scale scores 3–12), combining GFAP and UCH-L1 with variables such as age, motor response, pupil reactivity and CT findings raised predictive accuracy substantially—AUCs increased to roughly 0.90–0.94 for death and 0.83–0.89 for unfavorable outcome. In patients with mild TBI (GCS 13–15), the combined approach modestly improved prediction of incomplete versus full recovery (AUC ~0.69), suggesting other mechanisms beyond structural injury may drive long-term deficits in mild cases.
Ramon Diaz-Arrastia, MD, PhD, co-senior author and professor of neurology at the University of Pennsylvania, emphasized that while GFAP, UCH-L1 and imaging reflect structural brain injury, further research is needed to understand the mechanisms behind poor outcomes after mild TBI.
Nsini Umoh, PhD, program director at the National Institute of Neurologic Disorders and Stroke, which funded the work, highlighted the value of a large, diverse dataset like TRACK-TBI for developing better prognostic tools to guide patient care and therapeutic decisions.
Co-authors: See the full study for the complete author list.
Disclosures: Frederick Korley, MD, PhD, has consulted for and received research funding from Abbott Laboratories.
About this TBI research news
Author: Suzanne Leigh (UCSF)
Source: UCSF
Contact: Suzanne Leigh – UCSF
Image: The image is in the public domain
Original Research: “Prognostic value of day-of-injury plasma GFAP and UCH-L1 concentrations for predicting functional recovery after traumatic brain injury in patients from the US TRACK-TBI cohort: an observational cohort study” by Geoffrey Manley et al., published in Lancet Neurology. Clinical trial registration: ClinicalTrials.gov NCT02119182.
Abstract
Prognostic value of day-of-injury plasma GFAP and UCH-L1 concentrations for predicting functional recovery after traumatic brain injury in patients from the US TRACK-TBI cohort: an observational cohort study
Background
GFAP and UCH-L1 are brain-specific proteins that can be measured in blood soon after injury. Although they are used diagnostically to guide CT scanning after TBI, their ability to predict long-term functional outcomes when measured on the day of injury has been unclear. This study quantified their prognostic accuracy and assessed whether they add meaningful information to established clinical prediction models.
Methods
TRACK-TBI enrolled patients aged 17–90 evaluated for TBI at 18 U.S. level 1 trauma centers. All participants received head CT at evaluation and completed six-month functional assessments using the Glasgow Outcome Scale–Extended (GOSE). Day-of-injury plasma samples were analyzed for GFAP and UCH-L1 and examined as continuous measures and by quintile.
Findings
Of 2,552 enrolled patients, 1,696 had complete baseline and six-month outcome data and were included in the analysis. At six months, 7.1% had died, 13.9% had an unfavorable outcome (GOSE ≤4), 66.9% had incomplete recovery (GOSE <8), and 33.1% had full recovery (GOSE = 8). GFAP and UCH-L1 showed AUCs of approximately 0.87–0.89 for predicting death and 0.86 for predicting unfavorable outcome, while AUCs for predicting incomplete recovery were lower (~0.61–0.62). Adding biomarkers to clinical models most improved prediction in patients with GCS scores of 3–12.
Interpretation
Day-of-injury plasma GFAP and UCH-L1 offer good to excellent prognostic value for predicting death and unfavorable outcomes at six months, particularly for moderate and severe TBI. However, these biomarkers are less effective at distinguishing incomplete from full recovery in milder injuries. They provide the greatest additional prognostic information for patients presenting with lower GCS scores.
Funding
This research was supported by the U.S. National Institutes of Health, National Institute of Neurologic Disorders and Stroke, the U.S. Department of Defense, One Mind, and the U.S. Army Medical Research and Development Command.