The need for a blood biomarker for traumatic brain injury is greater than ever. Now, results from a new study published in JAMA Neurology suggest that glial fibrillary acidic protein (GFAP) can detect mild to moderate TBI up to a week after injury.
Ubiquitin C-terminal hydrolase L1 (UCH-L1) and GFAP are known to be helpful in identifying patients with mild TBI. However, there is limited information on the temporal relationship of mild TBI and serum levels of UCH-L1 and GFAP after presentation.
In order to assess the diagnostic accuracy of UCH-L1 and GFAP for detecting TBI, neurosurgical intervention, and intracranial lesions on head CT over time, Linda Papa, MD, Director of Academic Clinical Research at Orlando Regional Medical Center, and colleagues conducted a prospective cohort study of 3025 adult trauma patients screened for mild to moderate traumatic brain injury (MMTBI). MMTBI was defined as Glasgow Coma Scale (GCS) of 9 to 15 and disorientation, loss of consciousness, or amnesia after blunt head trauma. Of those assessed, 1030 met eligibility criteria, and ultimately 584 (mean age: 40 years; 62% male) consented to participation. Most patients with TBI were considered mild with a GCS of 13 to 15 (97.8%), with most (96.9%) receiving a head CT. Intracranial lesions were found in 10.8% and neurosurgical intervention was required in 2.2% of participants with MMTBI.
Blood samples were obtained from the 584 participants within 4 hours of injury, with repeated blood sampling conducted at 4, 8, 12, 16, 20, 24, 36, 48, 60, 72, 84, 96, 108, 120, 132, 144, 156, 168, and 180 hours after injury.
GFAP levels were detectable within an hour of trauma, peaking at 20 hours, and decreasing steadily during the following 72 hours, however it was still detectable at lower levels from 72 to 180 hours. UCH-L1 on the other hand had a rapid elevation, peaking at 8 hours and steadily decreasing over 48 hours.
Over 7 days, UCH-L1 had a diagnostic range of 0.30 to 0.67 area under the curve (AUC) for MMTBI, 0.31 to 0.77 AUC for intracranial lesions on head CT, and 0.50 to 0.92 AUC for identifying patients needing neurosurgical intervention. In contrast, GFAP had a diagnostic range of 0.73 to 0.94 AUC for MMTBI, 0.80 to 0.97 AUC for intracranial lesions on head CT, and 0.91 to 1.00 for identifying patients needing neurosurgical intervention.
“GFAP and UCH-L1 had very distinct temporal profiles,” Dr Papa told Neurology Advisor. “It was a very pleasant surprise that GFAP performed so well and performed so consistently throughout the 7 days after injury.”
“Getting blood samples from these patients over time is very challenging but is extremely important because individuals with mild TBI can present at different times after injury,” she continued, pointing out that “some patients do not seek medical attention for several hours to days after their injury,” because symptom onset is delayed. “GFAP appears to be useful both early and late after injury,” she said.
In an accompanying editorial, Tanya Bogoslovsky, MD, PhD, and Ramon Diaz-Arrastia, MD, PhD point out that UCH-L1 may be helpful in a point-of-care or hyperacute setting, whereas GFAP may be helpful in subacute settings.
While the findings show great promise for the detection of TBI, the authors note that sensitivity of testing will need to be improved upon in order for the test to be used regularly in a clinical setting.
“The authors used assays developed for GFAP and UCH-L1 with lower limits of quantification of 0.030 ng/mL for GFAP and 0.100 ng/mL for UCH-L1. This level of sensitivity represents an improvement from prior assays but may still be insufficient for mTBI diagnosis,” they wrote. In the study, “superior sensitivity and specificity for diagnosing TBI was obtained when GFAP was combined with UCH-L1 (AUC, 0.94), thus supporting that a combination of biomarkers maybe superior compared with each alone for the diagnosis and prognosis of TBI.”
- Papa L, Brophy GM, Welch RD, et al. Time Course and Diagnostic Accuracy of Glial and Neuronal Blood Biomarkers GFAP and UCH-L1 in a Large Cohort of Trauma Patients With and Without Mild Traumatic Brain Injury. JAMA Neurol. 2016. doi:10.1001/jamaneurol.2016.0039.
- Bogoslovsky T, Diaz-Arrastia R. Dissecting Temporal Profiles of Neuronal and Axonal Damage After Mild Traumatic Brain Injury. JAMA Neurol. 2016. doi:10.1001/jamaneurol.2016.0290.