Neuroimaging May Predict Recovery After Mild TBI

Pauline Anderson

July 21, 2016

Researchers believe it may be possible to use neuroimaging to predict which patients with mild traumatic brain injury (mTBI) have a better chance of recovery.

Their study showed that patients with abnormally high fractional anisotropy (FA), as seen on diffuse tensor imaging (DTI) at the time of their injury, are more likely to have a good outcome.

The results could affect the development and targeting of treatments for patients after concussion, said author Michael L. Lipton, MD, PhD, associate director, Gruss Magnetic Resonance Research Center, Albert Einstein College of Medicine, Bronx, New York.

"Developing an effective intervention requires first identifying the people who need it," said Dr Lipton, adding that most patients recover within a few months of a concussion with no intervention.

The study was published online June 9 in the American Journal of Neuroradiology.

The study included 39 adult patients (mean age, 38.5 years), who were evaluated in an emergency department and diagnosed with a concussion within 48 hours of their injury. Those with a previous concussion were excluded from the study.

The analysis also included 40 healthy volunteers. Patients and controls had similar preinjury intellectual functioning.

Study participants underwent DTI within 16 days of their injury. The healthy controls provided comparison imaging to allow the identification of abnormal FA.

FA is a measure of microstructural organization, Dr Lipton notes. "It's a feature of what is sometimes called structural connectivity, but it per se does not really speak to the functionality of the connection but speaks to the presence of normally structured white matter."

White Matter Regions

The researchers selected nine white matter regions known to be susceptible to mTBI and/or linked to functions associated with mTBI morbidity. These included the left and right frontal lobe, left and right temporal lobe, left and right thalamus, left and right cerebellum, and corpus callosum.

For each patient with mTBI, researchers developed an "EZ-MAP" to detect microstructural abnormalities. They separated these maps into images showing abnormally high FA (hFA) and abnormally low FA.

Cognitive function was assessed at 1 year by using the IntegNeuro, a computerized battery of tasks, with a summary z-score computed for each of the three domains (executive function, episodic memory, and attention).

Also at 1 year, researchers assessed postconcussion symptoms using the Rivermead Post-Concussive Symptom Questionnaire. On this instrument, patients rate the severity of 16 symptoms, with symptoms being categorized into cognitive, somatic and emotional factors.

As well, investigators measured health-related quality of life using the Sickness Impact Profile. On this scale, patients grade various aspects of daily functioning. The tool includes a physical dimension, psychological dimension, and social dimension.

The analysis included 26 of the patients with mTBI who returned for the 1-year follow-up. Six of the 26 had moved and were unavailable for onsite computerized assessments but completed the Rivermead and Sickness Impact assessments by phone.

The study showed that brain-wide imaging measures were significantly associated with memory performance at 1 year but not with attention or executive function. Mean FA, radial diffusivity, and mean diffusivity were significantly associated with health-related quality of life.

Researchers found several significant differences in 1-year assessments between participants with and without hFA in various white matter regions. For tasks of attention, those with hFA in the left frontal lobe performed significantly better (P = .008), as did those with hFA in the left temporal lobe (P = .004).

Participants with hFA in the right thalamus had significantly fewer emotional postconcussion symptoms (P = .003). Those with hFA in the left cerebellar white matter and those with hFA in the right cerebellar white matter had significantly fewer somatic postconcussion symptoms (P = .003 for both).

"We found that the presence of abnormally high FA was a predictor of better outcome," said Dr Lipton.

He noted that the better outcomes were not just for cognitive function but also for "emotional function and what we call health-related quality of life."

Patient characteristics did not seem to explain which participants had hFA and fared better.

"We looked at a whole range of characteristics, including age, sex, education, IQ, premorbid medical problems, and none accounted in any systematic way for the differences," said Dr Lipton.

Brain Compensation

The researchers believe that areas of hFA may represent evidence of the brain compensating for the injury.

"The way the brain recovers from, or responds to, injury is through mechanisms like neuroplasticity," said Dr Lipton. "What we are proposing here is that the areas of high FA that we are seeing may actually represent areas of compensatory or neuroplastic responses, not the actual injury itself."

He stressed that the study group was small and the number of factors examined "limited."

He also pointed out that there were also some areas of abnormally low FA.

To better understand the relationship between what's going on in the brain at the time of the injury and long-term outcomes, researchers could use a similar approach to examine other groups of patients, said Dr Lipton.

If that supports the idea that hFA at the time of injury is a predictor of better outcome, it might change the focus of treatments for TBI, he said.

To date, most of that focus has been on trying to "fix the injury," but it might prove more fruitful to enhance the compensatory mechanisms that are involved.

Dr Lipton used the example of stimulants, which are sometimes used to treat TBI. These drugs, he said, are believed to "ramp up" areas of the brain to compensate for dysfunctional networks elsewhere (as in attention-deficit/hyperactivity disorder).

Perhaps this and other treatments could be "revisited" in patients, such as those in the study with abnormal FA and less positive outcomes, who could potentially benefit.

But Dr Lipton stressed that researchers first have to better understand the mechanism of dysfunctional networks involved in TBI.

Another Notch

Asked to comment, Francis X. Conidi, MD, Florida Center for Headache and Sports Neurology, Port St Lucie, said this was "another good study supporting the use of DTI" in traumatic brain injury, or "another notch in the belt for DTI."

While other experts feel there's not yet enough research on DTI, he said, "I'm a huge supporter of it; I think it's the best biomarker for TBI."

However, he noted some things he would have liked to have seen in the study. "I would have used more formal neuropsychological testing; I'm not a big fan of computerized testing."

And he said he would have used "more of a symptoms checklist" to look for correlations between symptoms and FA values. "I would have liked to have known if headache, for example, is the more common problem."

A genetic component might help explain why some people with mild head trauma go on to have permanent sequelae while others recover with no problems, according to Dr Conidi.

"Maybe that's what they're seeing here," he said. "That higher FA value may be a genetic response to the injury and that's maybe why they have a better outcome."

Dr Lipton reports receiving a grant from the National institutes of Health. Dr Conidi has disclosed no relevant financial relationships.

Am J Neuroradiol. Published online June 9, 2016. Abstract

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.