Advanced Diffusion Imaging in The Hippocampus of Rats with Mild Traumatic Brain Injury

With this diffusion imaging protocol, it is possible to investigate microstructural changes in the hippocampus of a rat with a mild traumatic brain injury that are otherwise not visible on anatomical MRI. This technique can detect alterations in the brain following a mild and diffuse trauma which cannot be detected with CT or anatomical MRI. This technique makes it easier to monitor the recovery process after sustaining a mild traumatic brain injury in an objective and quantitative manner.

This diffusion imaging and analysis technique can also be applied in other disorders affecting the brain, such as dementia and multiple sclerosis, not only in preclinical studies but also in humans. In this protocol, it’s important that the quality of the diffusion scans and correction steps is high, therefore guidance from experienced technicians and analysts is suggested. Place the animal on a 37-degrees Celsius heating pad after confirming a lack of response to toe pinch in a 12-week old, female, Wistar H rat and insert a catheter into a lateral tail vein.

Inject 100 microliters of 2%lidocaine locally into the shaved and disinfected scalp and make a midline incision to expose the skull. Use a small scissors to remove any excess membranes and rub a cotton bud across the skull until the periosteum is no longer present, then use a drop of tissue glue to attach a 10-millimeter diameter, three-millimeter thick, metallic disk approximately 1/3 in front of and 2/3 behind the bregma. For traumatic brain injury induction, place the rat on a custom made bed with a foam mattress of a specific spring constant and place the rat directly under a transparent plastic tube with a 450-gram brass weight with the helmet as horizontal as possible.

Pull the weight up to one meter. With a second experimenter present, release the weight and have the second experimenter move the rat away from the tube immediately after the impact to prevent a second injury. Gently pull the helmet from the skull and use a gauze to stem any bleeding.

Close the skin with a suture and apply local analgesia gel to the incision. Place the rat on the bed of a CT scanner and administer a general-purpose, low-dose CT scan to rule out skull fractures, then place the rat in a clean cage on a 37-degrees Celsius heating pad with monitoring until full recumbency before returning the animal to its cage. Before and one day following the trauma induction, confirm a lack of response to toe pinch in the experimental animal and place the animal on the MR scanner bed in a headfirst, prone position.

Slide the quadrature volume coil over the head and advance the scanner bed into the scanner bore. To ensure a correct positioning, obtain a default 3-plane scout scan. When the scan is finished, load the scan in the image display and ensure that the head is lying straight and that the brain is positioned in the center of the magnet and coil.

Acquire T2-weighted images using the default settings, except for the field, the view, and matrix size which should be adjusted to a higher in-plane resolution of 109 by 109 micrometers. Open the Geometry Editor and place the slice package in the correct position, including the bulbus of the brain and the cerebellum and load three new echo-planar, diffusion-weighted, spin-echo sequences from the B_diffusion folder into the scan control protocol. Acquire diffusion-weighted images using the default settings and open the Edit Scan tab.

Set the slice orientation to axial and the number of slices to 25 to achieve a slice thickness of 500 micrometers and an inner slice distance of 600 micrometers and amend the readout direction into left-right. Under the Geometry tab, adjust the geometrical parameters and adjust the field of view and matrix size to 105 by 105 to ensure a resolution of 333 by 333 micrometers. Click the Diffusion tab within the Research tab for each of the three diffusion shells and adjust the number of diffusion directions to 32 for the first shell, 46 for the second shell, and 64 for the third shell.

Change the number of B0 images to five for the first shell, five for the second shell, and seven for the third shell, and adjust the gradient directions with custom gradient direction files. Adjust the B value per direction to 800 seconds per millimeter squared for the first shell, 1, 500 seconds per millimeter squared for the second shell, and 2, 000 seconds per millimeter squared for the third shell, then open the Geometry Editor and place the field of view between the bulbus and cerebellum containing only the cerebrum to reduce the artifact and scan time. At the completion of the scanning protocol, transfer the animal from the scanner bed to a clean cage with a 37-degrees Celsius heating pad with monitoring until full recumbency.

For diffusion MRI image processing, load the images in MRtrix3 and perform noise correction and Gibbs ringing correction on the diffusion-weighted images in the software program. Convert the corrected, diffusion-weighted images in T2 image to the NIFTI format as indicated. To perform correction for echo-planar imaging, motion and eddy current distortions, in the Plug-ins menu of ExploreDTI select Correction for subject motion EC/EPI distortions and select the preprocessed diffusion data file.

To calculate the diffusion tensor imaging metrics for each rat, click Plug-ins and Export stuff to NIFTI, then select the parametric maps of the diffusion tensor imaging model and export the parametric maps for the kurtosis and white matter tract integrity models. To create a mask file for the hippocampus of each rat, load the fractional anisotropy image of the rat in the MRtrix viewer and click the plus button to create a new region of interest. To extract the diffusion metrics of the hippocampus of the rat, import the created mask file in the AMIDE software and open the parametric maps and mask image of the rat.

To add the region of interest of the mask file into AMIDE, select the mask file image, click Edit, Add Region Of Interest and 3D isocontour and give the region of interest a meaningful name. Click on the region of interest displayed in the mask image and confirm that this volume should only contain voxels with a value of one. To calculate the mean values of the diffusion metrics in the hippocampus, click Tools and Calculate Region Of Interest Statistics and indicate the images and the region of interest to be included.

After clicking Execute, a pop-up window with the computed values that can be used for further statistical analysis will appear. In this representative experiment, there was no evidence of skull fracture as assessed by CT imaging and the T2 images did not show any abnormalities at the contusion site one day after the trauma. To examine the quality of the non-rigid co-registration step between the T2 image and the diffusion data set, an overlay of the T2 image was added to the color-encoded fractional anisotropy map.

The parametric maps for fractional anisotropy, mean diffusivity, axial diffusivity, and radial diffusivity parametric maps could then be calculated. Within the region of interest, calculation of the mean values for the axial, mean, and radial kurtosis values as well as values for the axonal water fraction, axial and radial extra-axonal diffusivity, and tortuosity of the white matter tract integrity could also be performed. In this representative experiment, analysis of the diffusion tensor imaging metrics revealed a significant increase in the fractional anisotropy values and a decrease in the diffusivity values following impact in the mild traumatic brain injury group.

Diffusion kurtosis metrics also showed a significant decrease in radial kurtosis following impact while no changes in the axial or mean kurtosis were observed. Using the white matter tract integrity model, the radial extra-axonal diffusivity displayed a significant decrease and the tortuosity demonstrated a significant increase in the mild traumatic brain injury group one day after the impact. During the image analysis, it’s important to check whether the data format from MRtrix was properly converted and imported into ExploreDTI and that each correction step was performed correctly.

Instead of ROI-based analysis, a voxel-by-voxel analysis can be applied to investigate whole-brain alterations. This technique is very valuable within the neuroimaging field and can be applied to other brain disorders as well, for example dementia and multiple sclerosis.