Blood Test May Accurately Diagnose Traumatic Brain Injury, Need for CT Scan

Batya Swift Yasgur, MA, LSW

July 26, 2018

A high-sensitivity blood test appears to accurately identify the need for computed tomography (CT) scan in patients with suspected traumatic brain injury (TBI), new research suggests.

Investigators studied close to 2000 adult patients who presented to the emergency department (ED) with suspected TBI and who had undergone both head CT and blood collection within 12 hours of the injury.

They measured ubiquitin C-terminal hydrolase-L1 (UCH-L1) and glial fibrillary acidic protein (GFAP), 2 biomarker proteins that are released into the bloodstream after TBI.

They found that in more than 96% of patients, a blood test combining both of these biomarkers accurately predicted the absence of acute intracranial injury, as confirmed by CT scan.

"We see this test not as a substitute for head CT scan, but as an accurate and objective way to help emergency care providers decide which patients really need one," Jeffrey Bazarian, MD, MPH, professor, Department of Emergency Medicine, University of Rochester Medical Center, New York, told Medscape Medical News.

"Our results suggest that this blood test holds potential for ruling out the need for head CT scan among head-injured patients presenting to EDs, in whom a head CT scan is felt to be clinically indicated; in fact, up to a third of head CT scans done in the acute setting of TBI could be avoided, with very low false-negative rates," he said.

The study was published online July 24 in Lancet Neurology.

"Promising" Biomarker

In the United States, TBI results in more than 2.5 million ED visits annually, with most being mild TBI (concussion), the authors write.

The "diagnostic modality of choice" in evaluating patients for traumatic intracranial injuries is a head CT scan, which accounts for approximately 20 million CT scans performed annually in the United States.

"[T]he widespread use of head CT scanning has been questioned due to potential adverse effects of radiation exposure, unnecessary [ED] resource use, and cost," the authors state.

Moreover, the prevalence of CT-detected intracranial injury in patients with mild TBI is typically less than 10%.

Previous research has suggested the potential for blood-based brain injury biomarkers to predict the absence of intracranial injury after TBI and aid in reducing unnecessary head CT scans.

"A blow to the head can result in injury to the cells in the brain, resulting in the release of the proteins that make up these cells," Bazarian explained.

"These proteins leave the brain and enter the bloodstream through the blood–brain barrier, which transiently opens after a head injury."

S100B, a well-established biomarker for TBI, is already in clinical use in Europe; however, to date, no blood-based brain biomarker tests have been approved by the US Food and Drug Administration for clinical use in the United States.

Two novel brain proteins that have been advanced are UCH-L1 and GFAP, but preliminary investigations into these "promising" biomarkers have been limited by small cohort sizes, variability in blood sample acquisition timing, and retrospective determination of cutoff values.

The researchers therefore conducted the ALERT-TBI trial (A Prospective Clinical Evaluation of Biomarkers of Traumatic Brain Injury), conducted at 22 sites globally, to evaluate the ability of a biomarker test combining the two proteins at predetermined cutoff values to predict traumatic intracranial injuries on head CT scan within 12 hours of TBI.

Patients were required to be at least 18 years old, present to participating EDs with suspected nonpenetrating TBI resulting from external force, and have a Glasgow Coma Scale (GCS) score of 9 to 15.

Neither loss of consciousness nor amnesia was required for inclusion.

Patients had to have undergone noncontrast head CT scanning and blood sampling within 12 hours of injury.

There were 2 primary study outcomes: the negative predictive value and sensitivity of the UCH-L1 and GFAP test results for intracranial injury on head CT.

The secondary outcomes were to determine the UCH-L1 and GFAP test specificity, positive predictive value, and likelihood ratio negative values for the presence or absence of intracranial injury detected on CT and neurologically manageable injury.

Expectations Exceeded

Of the patients with TBI who had a valid head CT scan and test results available for analysis (n = 1959), 98% (n = 1920) had a GCS score of 14 to 15 and 6% (n = 125) had a traumatic intracranial injury on head CT, whereas less than 1% (n = 8) had neurosurgically manageable lesions.

Subarachnoid hemorrhage was the most common finding on head CT, followed by acute subdural hematoma and parenchymal hematoma. This distribution was similar in the subset of patients with a GCS score of 14 to 15.

Acute subdural hematoma was the most common finding among those with neurosurgically manageable lesions, followed by ventricular compression and subarachnoid hemorrhage.

Patients underwent blood draw a median of 3.2 hours after injury.

Both GFAP and UCH-L1 concentrations were significantly higher among patients who were CT positive compared with those who were CT negative.

The median GFAP in CT positive vs negative patients was 135.0 vs 22.2 pg/mL (P < .0001), respectively.

The median UCH-L1 in CT positive vs negative patients was 604.8 vs 261.0 pg/mL (P < .0001), respectively.

Among all patients with TBI, 66% (n = 1288) had a positive test and 34% (n = 671) had a negative test.

For acute intracranial injury, the overall test sensitivity was 0.976 (95% confidence interval, 0.931 - 0.995), and negative predictive value was 0.996 (95% confidence interval, 0.987 - 0.999).

The test performed similarly among patients with GCS scores of 14 to 15 and had slightly better performance among the patients with a GCS score of 9 to 13.

Notably, the test was 100% sensitive with a 100% negative predictive value for detecting neurosurgically manageable lesions.

When the researchers conducted a sensitivity analysis comparing the diagnostic accuracy of the combined biomarker test with each protein individually, they found that the combination of both proteins outperformed each marker separately on the primary outcomes.

The authors note several limitations.

For example, although patients with extracranial injuries were included, the researchers did not evaluate their effect on test performance.

"We knew the test would perform well, but the results exceeded our expectations," Bazarian reported.

"We were pleasantly surprised to see that 97.6% of patients with a CT-detected intracranial injury, like hemorrhage, tested positive, and 99.7% of patients with a negative test had a normal head CT scan," he said.

Unanswered Questions

Commenting on the study for Medscape Medical News, Andrew Maas, MD, PhD, emeritus professor of neurosurgery, Antwerp University Hospital and University of Antwerp, Belgium, who was not involved with the study, called it "an important step forward in the question for objective blood-based biomarkers to support a diagnosis of TBI and to inform clinical decision-making."

However, he cautioned, "ALERT-TBI does not provide the final answer, [as] specifically, the results are relevant to triaging the need for CT scanning in patients with mild TBI."

Moreover, "from a scientific perspective, the study has left important questions unanswered: Does the test indeed offer 'added value' over current practice that includes clinical decision rules for triaging the need for CT scanning?"

A second question, said Maas, who is the coauthor of an accompanying editorial, is whether two biomarkers are better than one, and if better, would this be cost-efficient?

"The scientific community would do well to critically assess the results of ALERT-TBI and their limitations before widely adopting the test into clinical practice," he emphasized.

Bazarian disagreed. "I think this test is ready for clinical use, although it is not yet commercially available."

He noted that the test was cleared for use in the United States by the US Food and Drug Administration in February 2018.

The study was funded by Banyan Biomarkers and US Army Medical Research and Materiel Command. Bazarian received research funding from Banyan Biomarkers and research funding from BrainScope Company while the study was conducted. The other authors' disclosures are listed on the original paper. Maas declares grants from the European Union Framework Programme 7, and personal fees for speaking and consultancy from Integra Life Sciences. His coauthor's disclosures are listed on the original editorial.

Lancet Neurol. Published online July 24, 2018. Article abstract, Editorial extract

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.