Monday, February 26, 2018

Diagnosing concussion with a blood test?

- Jennifer Middleton, MD, MPH

The Food and Drug Administration (FDA) reported last week that it had approved a new blood test to help diagnose mild traumatic brain injury (mTBI or "concussion") called the Banyan Brain Trauma Indicator. The test measures two proteins, UCH-L1 and GFAP, that are released by the brain into the bloodstream within 12 hours of injury. The FDA press release includes discussion regarding this test's potential to reduce the number of CT scans patients with suspected mTBI receive. The study cited by the FDA to approve this test, however, has not yet been published for physicians to review.

A 2012 AFP article reviewing "Current Concepts in Concussion: Evaluation and Management" describes the difficulties in making a concussion diagnosis:
Concussion can be difficult to recognize, complicated by the lack of a universal definition. Additionally, there are no direct objective measures for diagnosis or recovery, no treatments with well-documented effectiveness, and limited empiric prospective data to guide return-to-play decisions.
The article reviews the high incidence of mTBI in the United States (estimated to be between 1.6-3.8 million injuries annually) and cites that over 1 million of these patients are evaluated in US emergency departments every year. Many of these patients get CT scans of their brain to rule out serious intracranial processes despite the existence of guidelines like the New Orleans criteria that discourage CT scans in patients without headache, vomiting, intoxication, amnesia, seizure, or visible trauma. Using serum biomakers to reduce the use of CT scans in suspect mTBI patients would save patients from unnecessary radiation exposure; this radiation exposure may increase risk for future cancers, especially in patients who receive multiple scans over their lifetimes. An AFP Journal Club article author from 2010 is quoted as saying, "After I have told a patient or parent that a head CT is the equivalent of 100 to 200 chest radiographs, you'd be surprised how many say they really don't need or want the test."

Prior studies regarding the utility of UCH-L1 and GFAP to diagnose concussion have had mixed findings. A previous study examining the use of these 2 proteins along with a 3rd not mentioned in the Banyan Brian Trauma Indicator found decent sensitivity for detecting mTBI but poor specificity; a positive test result was suggestive of mTBI, but a negative test result did not reliably rule it out. A systematic review of these biomarkers' utility from last year was less enthusiastic, finding "insufficient evidence" that these tests are "ready for clinical application."

While the FDA's desire to make this test available as quickly as possible is admirable, not sharing the data that led to its approval makes it challenging for physicians to use it. As with most medical diagnostic tests, the results of the Banyan Brain Trauma Indicator will not merely be "yes" or "no," and understanding the chances of a false positive (or negative) along with the positive (or negative) predictive value will be important for physicians to apply it thoughtfully. Hopefully more information will be forthcoming from the FDA soon, and hopefully, too, the study they cite will be available for our scrutiny as well. In the meantime, here are some mTBI resources from AFP for your review.