, according to authors of a recent study. In particular, they said, using neurite orientation dispersion and density imaging (NODDI) to monitor long-term mTBI impact on brain regions related to cognitive and emotional processing can help clinicians assess recovery, predict progression, and optimize treatment.
“Currently,” said co-senior study author Ping-Hong Yeh, PhD, “there is a lack of minimally invasive, quantitative diagnostic biomarkers for monitoring progression or recovery after mild TBI. However, mild TBI can be quite disabling, with many patients reporting symptoms months or even years after injury. This is the most difficult part to diagnose.” Dr. Yeh is a researcher at the National Intrepid Center of Excellence (NICoE) at Walter Reed National Military Medical Center, Bethesda, Maryland.
The NICoE, a Department of Defense organization and the senior member of Defense Intrepid Network for Traumatic Brain Injury and Brain Health, is among several centers charged with improving support for injured service members’ recovery, rehabilitation, and reintegration into their communities. The overarching goal, said Dr. Yeh, is to enable community neurologists to refer service members and veterans to these centers for treatment and advanced imaging when needed.
Invisible Wounds
Limitations of conventional MRI and CT make it tough to discern which patients with mTBI will return to baseline functioning, and which will develop long-term complications. Addressing the silent or invisible wounds of mTBI will require improved diagnostic, prognostic, and therapeutic tools, he said.
For their study, published in JAMA Network Open, Dr. Yeh and colleagues compared diffusion tensor imaging (DTI) and NODDI data from 65 male service members with remote (more than 2 years old) mTBI against scans of 33 noninjured controls matched for age, sex, and active-duty status.
“Although DTI is very sensitive in detecting microstructural changes in mild TBI,” he said, “it is not specific to the underlying pathophysiological changes.”
Conversely, NODDI uses biophysical modeling of intracellular diffusion, extracellular diffusion, and free water to help physicians to understand subtle pathophysiological changes with greater sensitivity and specificity than does DTI. “This will allow us to correlate symptoms with brain structural changes, making the invisible wound visible.”
In the study, the greatest differences between injured and control patients appeared in the following NODDI metrics (P <.001 in all analyses):
- Intracellular volume fraction (ICVF) of the right corticospinal tract (CST)
- Orientation dispersion index (ODI) of the left posterior thalamic radiation (PTR)
- ODI of the left uncinate fasciculus (UNC)
Regarding patient-reported neurobehavioral symptoms, Neurobehavioral Symptom Inventory cognitive subscores were associated with fractional anisotropy of the left UNC. In addition, PTSD Checklist–Civilian version total scores and avoidance subscores corresponded, respectively, with isotropic volume fraction (ISOVF) of the genu of corpus callosum and with ODI of the left fornix and stria terminalis.
Next Steps
Presently, Dr. Yeh said, conventional MRI and CT usually cannot differentiate between axonal injury, axonal inflammation (which develops during the chronic phase of mTBI), and demyelination. “But newer biophysical modeling, such as NODDI, will allow us to tell the difference.” Along with providing prognostic information, he said, such technology can guide appropriate treatment, such as anti-inflammatory agents for chronic inflammation.
Most community neurologists refer patients with persistent mTBI symptoms in the absence of red flags using CT and conventional MRI for advanced neuroimaging, said Dr. Yeh. But because few community neurologists are familiar with NODDI, he said, broadening its reach will require educating these providers. Additional steps that Dr. Yeh said could occur over the next decade or more include boosting advanced dMRI sensitivity levels through improved hardware, software, and diagnostic tools.
“We need to make these techniques clinically feasible,” he added. Currently, protocols that allow advanced dMRI scans in about 10 minutes can be achievable.
The investments required to implement advanced dMRI techniques will be substantial. A state-of-the-art 3T MRI scanner that can support NODDI and DTI can easily cost $1 million, said Dr. Yeh. Factor in additional equipment options and construction costs, he added, and the total price tag can easily exceed $2 million. But rather than replacing all existing MRI systems, said Dr. Yeh, AI one day may help translate high-gradient capability even to widely used lower-field MRI scanners operating at 0.5T.
Streamlining systems that incorporate disparate scanners with different acquisition parameters will require standardized data acquisition and sharing parameters. Along with helping to evaluate new techniques as they become available, data harmonization and sharing can facilitate a shift from research comparisons between large groups to comparing a single patient against many others — a move that Dr. Yeh said must occur for advanced dMRI techniques to achieve clinical relevance.
In addition, experts will need to revise clinical guidelines for use of new technologies as their availability grows. “Improper use of these techniques will not only increase health costs, but also probably result in adverse health results.” Such guidelines could be very useful in evaluating the suitability and quality of referrals for diagnostic images, Dr. Yeh said.
Dr. Yeh reports no relevant financial interests. The project was partially funded by the US Army Medical Research and Materiel Command.