Sagittal anterior-posterior rotational acceleration induced head injuries in the rat

Rapport, 2007

A new test device to produce sagittal plane rearward rotational acceleration induced diffuse brain injury to the rat has been developed. During trauma, the heads, which were fixed to a rotating bar by means of curved plates that were glued to the skull bones, were exposed to rotational acceleration between 0.3 and 2 Mrad/s2. The animals were sacrificed 3, 24 and 72 hours after trauma. Blood samples were collected for serum analyses prior to sacrifice and subsequent dissection of the brains. Transverse cryostat brain tissue sections were cut at three locations in the brains. The sections were stained with β-Amyliod Precursor Protein (β-APP) and Neurofilament (NF) antibodies and probes for Cyclooxygenase 2 (COX2) mRNA in-situ hybridisation to detect decaying axons, cytoskeleton changes and affected nerve cells. Bands of β-APP positive axons, i.e. axons with reduced plasma flow and hence probably dying axons, were seen in the corpus callosum, thalamus and hippocampus and in the border of these regions in most animals exposed to rotational trauma at 1.1 Mrad/s2 or higher. Similarly for the COX2 marker; above 0.9 Mrad/s2 the numbers of stained cells were large for a number of locations in the cortex and hippocampus region. Only negligible β-APP and COX2 mRNA upregulation were observed in the sham exposed controls and normal animals. NF changes were not observed in controls or exposed animals euthanized prior to 3 hours after trauma, but visible 24 hours after trauma. The S100 serum analyses indicate that blood vessel and glia cell injuries occur at rotational accelerations above 1.1 Mrad/s2. The data clearly indicate that the rat brain is injured at a rearward rotational acceleration of 1.0 Mrad/s2 when the rotational acceleration pulse have a duration of 0.4 ms. The scaled representative global head rotational acceleration and duration for humans is estimated to be 9 krad/s2 and 4 ms, respectively. However, the study also highlight that these threshold values are highly influenced by the choice of brain dimensions used in the scaling equations. In addition, differences such as position of the cerebellum, lack of gyruses in the rat brain, and location and shape of the ventricles, etc were not taken into account. In order to improve the understanding the effect on scaling laws of these differences between the human and the rat, it is recommended to carry out a parametric study on these issues using detailed FE-models of the rat and humans.