Is Diffusion Tensor Imaging A Potential Biomarker for Mild Traumatic Brain Injury?
ARTICLE IN BRIEF
In a small trial, diffusion tensor image showed that the damage from mild traumatic brain injury is detectable even in patients with minimal impairments on standard neuropsychological tests.
Diffusion tensor imaging (DTI) is emerging as an important tool for detecting injury in mild traumatic brain injury (mTBI). A new study published online Feb. 10 ahead of the print edition of Neurology shows that damage is detectable even in patients with minimal impairments on standard neuropsychological tests. The results confirm that signs of injury may persist months after clinical recovery, and suggest that DTI may have a place in longitudinal assessment in mTBI patients wider than currently practiced.
“One of the frustrating problems in patients with mild traumatic brain injury is that conventional neuroimaging doesn't show anything,” according to lead researcher Andrew R. Mayer, PhD, research scientist at the Mind Research Network in Albuquerque, NM. CT scans and T1- or T2-weighted MRI typically come back negative, “but almost all patients report some cognitive or neurological complaint in the first couple of weeks.” While most patients appear to have complete resolution in three to six months, about five to twenty percent have persistent complaints.
Changes in white matter have been seen with DTI in animal models, and it has been applied to small groups of patients as well. However, Dr. Mayer said, most of the previous work in patients has been on those with long-term complaints. “That's a big methodological concern in the field. When you are only seeing twenty percent of the population, you are not getting a good estimate of the population in total.”
STUDY METHODS, RESULTS
For his study, Dr. Mayer recruited 22 patients from the emergency room approximately two weeks after their injury, whether or not they had cognitive complaints, and excluding only patients with pre-existing neurologic or neuropsychiatric disease. Patients underwent DTI imaging upon enrollment, and then again after three to five months. They also underwent standard neuropsychological tests. He also imaged and tested a group of control subjects matched for educational level and statistically matched for reading ability, which, Dr. Mayer explained, is the cognitive skill most resistant to effects of mTBI, and therefore provides a measure of preinjury intellectual function. “It's not perfect, but it's the best thing we have,” he said.
Diffusion tensor imaging measures the Brownian motion of water molecules. Water moves less freely across axonal tracts than it does along them, a phenomenon known as “anisotropy.” Disruption of axons alters the random movement of water, and can lead to a decrease or increase in anisotropy. Six regions of the brain were chosen for imaging, to give a snapshot of the major white matter tracts across the brain.
“We were expecting fractional anisotropy [a measure of anistrophy] to be reduced in the mTBI group compared to controls, but that's the opposite of what we found,” Dr. Mayer said. “The anisotropic diffusion actually increased following injury,” meaning water tended to move less randomly, more directionally following injury. The explanation is not yet clear, but may be due to swelling of injured axons, which might constrain water movement even more than in the healthy brain. It is also possible that injury induces shifts in relative water content within the myelin sheath, which may complicate the picture further.
“There may be subtle changes that can be picked up by DTI that are not reflected in neuropsychological testing,” Dr. Mayer said. Encouragingly, DTI abnormalities in some of the white matter tracts had partially resolved in those patients who returned for a follow-up visit at three to six months.
If this result is confirmed in larger studies, it may suggest that DTI could serve as a biomarker for brain recovery. “This is one of the largest studies of its kind to date,” Dr. Mayer said, “but the sample is still relatively small.” He hopes to eventually expand the study by an order of magnitude, and increase follow-up to one year. “This is not the end of this story.”
“One of the problems in mild traumatic brain injury has always been that there is an enormous amount of controversy about the extent to which symptoms are psychogenic or neurogenic,” according to Elisabeth Wilde, PhD, a neuropsychologist who specializes in neuroimaging and is assistant professor in the Department of Physical Medicine and Rehabilitation at Baylor College of Medicine. This study, she said, which replicates findings in several previous ones, “gives more credence to the idea that this could have a neurogenic basis.”
The absence of neuropsychological test findings in patients with persistent white matter changes doesn't surprise Dr. Wilde. “As a clinician, absolutely not,” she said. “Speaking as a neuropsychologist, these tests are subjective to some degree, and they just may not be sensitive enough.”
The DTI findings, she suggested, might be useful for determining which patients might not be recovering as quickly, and therefore which may benefit from more in-depth intervention. “If we could predict who those patients are going to be, by using DTI as a biomarker, we might be able to help them earlier, by focusing treatments on that early period,” which might minimize their symptoms later. DTI, she said, “is potentially one of the better tools we have to look at this phenomenon.”
Dr. Wilde thinks that baseline DTI scans are in order for athletes at high risk for mild TBI, such as professional football players. “I think they absolutely should” receive such a scan, she said.
But before that becomes likely, the field will need to become a bit more standardized. In conventional imaging, a scan from one facility can be easily read by an expert from any other facility. “You can't really do that with DTI,” Dr. Wilde said. “Anything you do to change the acquisition or post-processing parameters is going to change the values you get. That's been one of the major obstacles in getting normative values for clinical use.” Because of this, the technology “is still not ready for widespread clinical use across all kinds of patients and conditions.”
Nonetheless, she said, “I think there is the potential for widespread application,” in brain injury as well as in other conditions. “I think these issues will get resolved, because it has become such a powerful tool.”