MRI Features Associated With Clinical Symptoms in Mild Traumatic Brain Injury

Investigators aimed to predict outcomes in patients with mild traumatic brain injury by identifying MRI features that are associated with clinical symptoms.

VANCOUVER, British Columbia — Select metrics of magnetic resonance imaging (MRI) in patients with mild traumatic brain injury (mTBI) significantly correlated with clinical measures, according to new research.

Presented at the 2016 annual meeting of the American Academy of Neurology (AAN), the longitudinal study aimed to predict outcomes by determining multimodal MRI data features that are associated with clinical symptoms in patients with mTBI.

“The pathology underlying neurocognitive change after mTBI is poorly understood. Multimodal MRI was used to study the impact of mTBI to the functional and structural organization of the brain,” the researchers wrote in the abstract. “In this study, global, local, and network-specific metrics were extracted from MRI data and correlated to clinical symptoms.”

The study included 86 patients with uncomplicated mTBI and 26 controls who received neurologic examination and MRI scans consisting of resting-state functional MRI (rs-fMRI), T1, T2 fluid-attenuated inversion recovery, diffusion, and perfusion imaging on MR750 3T scanners (GE Healthcare).

Patients with mTBI underwent evaluation and scanning at 4 time periods—3 days, 7 days, 1 month, and 3 months after injury—whereas participants in the control group were scanned twice, 1 week apart.

Researchers created pipelines for analysis of rs-fMRI, diffusion, and volumetry of brain structures; computed functional and structural networks; and extracted connectivity metrics. They noted that multishell diffusion data permitted computation of radial and orientational quantities, such as diffusional kurtosis, fiber tractography, and conventional diffusion tensor imaging metrics. They then correlated these features with clinical symptom severity score via the Sports Concussion Assessment Tool 2.

Results indicated that voxel and network-based fMRI metrics, such as fractional amplitude of low frequency fluctuations (P<.02), were associated with symptom scores.

“A subset of functional networks displayed broader spatial spread in mTBI subjects than in controls,” the researchers wrote.

In addition, there was no significant correlation between symptoms and whole white-matter diffusional kurtosis and fractional anisotropy histograms; voxel-based analysis is currently underway in 40 patients.

There was also a correlation between symptom score and decreased whole brain, hippocampus, thalamus, caudate, and amygdala volumes, but only at the fourth encounter (P<.05).

As a result of these findings, the researchers concluded that they successfully described a longitudinal, multimodal MRI study in an mTBI population, and reported statistically significant correlations between clinical measures and select MRI metrics.

Click here for more coverage from the 68th Annual Meeting of the American Academy of Neurology, April 15-21, 2016, in Vancouver, British Columbia, Canada.


Marinelli L, Chebrolu VV, Gallenberg AJ, et al. A multimodal MRI study in mild traumatic brain injury: correlation of imaging features with clinical symptoms. Presented at: The 68th Annual Meeting of the American Academy of Neurology; April 15-21, 2016; Vancouver, British Columbia, Canada. Abstract S11.002.