Traumatic Brain Injury (TBI) is often identified as the signature injury of soldiers involved in the Iraq and Afghanistan wars and is commonly caused by improvised explosive devices or motor accidents [1,2]. Mild TBI (mTBI) is the most prevalent form of TBI and is often missed at time of initial injury. The range of mTBI post-injury symptoms varies widely from metabolic changes with quick recovery to long-term damage [3]. Early detection and treatment of TBI is vital for minimizing long term effects. However, some soldiers are suggested to be at risk of mild TBI (mTBI) due to the nature of their work [4, 5]. Previous studies suggest increases in metabolism at the site of injury [5] and ³¹P Magnetic Resonance Spectroscopy, in particular Pi/PCr ratio, is suggested to be a sensitive probe of the energetic status of tissue. In this study we use ³¹P MRS to investigate whether changes can be identified in soldiers at risk of mTBI due to blast overpressure.

14 healthy males were recruited from in-service breachers who train with low-level explosives. The study involved 5 MR scanning sessions at baseline, and at 1, 3, 7 and 28 days following blast training. MR scans acquired were as follows:

T₁-weighted TFE: 1mm³ isotropic voxels FOV = 256x256mm, 176 slices, TI/TR/TE = 900/1950/2.26ms, flip angle = 10⁰. Cortical reconstruction and volumetric segmentation was performed using Freesurfer software longitudinal stream [6]. Brain volumes were measured for whole brain, white matter, grey matter, as well as hippocampus, thalamus and corpus callosum based on previous hypothesis.

³¹P MRS: Data were acquired from a 216ml cubic voxel encompassing the visual cortex using a rigid double tuned 31P coil (Rapid Medical Systems GmBH) and ISIS localization (BW=6000Hz, 2048 points, TR=6000ms, 40 averages). Data were truncated to 1024 points and zerofilled to 2048 before 12Hz Lorentzian line-broadening was added. Phase corrected spectra were analyzed using AMARES in jMRUI, fitting to PCr, ATP (2x doublet and triplet), PME, PDE and Pi. Data are presented for Pi/PCr (related to bioenergetics status), pH, and PME/PDE (associated with membrane breakdown and proliferation).

Figure 1 shows a typical ³¹P MRS spectrum with peak fitting in AMARES. Brain volumes were not significantly altered from baseline at any timepoint. However, Pi/PCr ratio tended to show a gradual decrease from baseline following risk of mTBI, with decreases reaching significance 28 days after training with low level explosives (Figure 2). In contrast pH and PME/PDE showed no significant changes over time following incidence.

Evidence for decreases in Pi/PCr ratio indicates potential changes in metabolic status in the brain following injury which appears to increase in severity over a month following potential injury. However, although hypermetabolism has previously been suggested as a compensatory mechanism following injury, the direction of the changes seen in this study (decreases in Pi/PCr) are in the opposite direction to those expected during muscle exercise studies (decreases in PCr and increases in Pi during exercise). These changes could indicate long term hypometabolism, however energy consumption rates are not directly detected in this study, but are inferred from changes in steady state concentrations.Despite no reported brain injuries during training events, this study indicates changes in Pi/PCr ratios, measured using ³¹P MRS. However, due to the duration of the study it is not clear whether these changes return to baseline. Further understanding of the effects of the blast events on the brain may indicate the necessity for different training schedules to allow for additional recovery between sessions, or for changes in breacher safety equipment. ³¹P MRS may provide a useful tool for assessing whether any changes are necessary.