Abstract
1483
Objectives: Traumatic brain injury (TBI) is a leading cause of sustained impairment in military and civilian populations. Mild TBI (mTBI) often exists without visible brain lesions, and its diagnosis lacks objective clinical biomarkers. Recent studies show that mTBI can increase the risk of developing dementia and Parkinson’s disease.[1] Repetitive mTBI can result in white matter inflammation which in turn can cause dysfunction and degeneration.[2] Imaging modalities such as positron emission tomography (PET) can provide additional information on metabolic abnormalities, which could aid in TBI diagnosis. In this study, we investigated metabolic changes in subjects with repetitive mTBI using the data acquired during the course of a 6 year observational imaging study which included FDG PET as well as multi-modal MRI.
Methods: Subjects were active duty service members, and the diagnosis of mTBI was determined using routine comprehensive clinical assessments by health care professionals at the WRNMMC based on VA/DoD clinical practice guidelines.[3] A total of 508 male subjects with mTBI (>6 month post-injury) were studied. Subjects were assigned to TBI≥3 group (n=298, mean and standard deviation of age = 37.8 ± 6.8) if the number of TBI events was greater than 2, and the rest of the subjects were assigned to the TBI≤2 group (n = 210, age = 34.2 ± 7.6). Each individual subject’s FDG PET uptakes were normalized to the whole brain. The PET images were analyzed using non-parametric permutation-based tests with age as a covariate. Clusters of voxels showing significant differences from the t-statistic maps at corrected p≤0.05 thresholds were extracted using FSL’s cluster tool. Locations and peak intensities of local maxima in the t-statistic maps were reported.
Results: Voxel-wise permutation tests found that subjects with 3 or more TBI events have a distinct spatial pattern of metabolic changes within the brain that are significantly different compared to subjects with 2 or less TBI events as shown in Figure 1. Subjects with 3 or more TBI events had higher metabolism in the cerebral cortices and deep gray matter including the thalamus and putamen. The cluster sizes and locations of voxels showing maximum intensity in each cluster in Table 1. Furthermore, subjects with 3 or more TBI events had lower metabolism than was seen in the white matter and midbrain. Conclusions: Subjects with TBI ≥ 3 have significantly higher glucose uptake in the dorsal thalamus and cingulate gyrus when compared to subjects with TBI ≤ 2. Cerebral glucose uptake is generally thought to be associated with physiologic metabolism within the gray matter structures of the brain associated with activation of neuronal tissue. An alternative source of increased glucose uptake could be from inflammatory processes. Prior studies with 11C-PK11195 PET, a marker of microglial activation, have demonstrated a link between traumatic brain injury and chronic inflammation which may predispose mTBI subjects to neurodegeneration.[4] These studies provide evidence of microglial activation leading to thalamic inflammation many years after TBI. Inflammation might produce the patterns observed in PET FDG uptake differences as seen in Figure 1. Moreover, a dynamic inflammatory process in response to mTBI might explain the evolution of symptoms and risk for potential neurodegenerative processes. A physiologic inflammatory response may result in slow healing and improvement in symptoms over time as microglial cells perform the function of removing injured neurons and neuronal cell components. This may be seen as a mild increase in the metabolic activity associated with the white matter tracts which are generally accepted to be the most vulnerable to mild head trauma. However, in the setting of recurrent mTBI, the inflammatory process may be chronically activated leading to involvement of the deep gray matter structures and result in greater risk for neurodegeneration.
