Biomechanical characterisation of human cranial bone: an experimental and numerical approach

Skull fractures are the result of an intense acceleration or an impact to the head, which can be caused by falls, vehicle accidents, accidental hits in sport and recreation, and assaults. They represent one of the clinical risk factors for the development of intracranial injuries or Traumatic Brain Injuries (TBIs). In Europe, head injuries are the leading cause of death in the age group of 15-44 caused mainly by blunt force and constitute a considerable portion of the global injury burden, being a major cause of death in developed countries and are increasingly being considered as one of the global health priorities. Many research groups have dedicated their efforts to creating better head protective gears for which developing in silico models are key to simulate different accident conditions. The reliability of such models lies in the use of material models which closely reproduce the mechanical behaviour, among others, of the human skull. However, despite the existing data (like tensile and flexural properties), there are still mechanical properties that remain to be determined for the human skull. The present thesis aims to characterise the biomechanical properties of cranial bone, with particular emphasis on the flexural and fracture mechanics properties, using both experimental and numerical techniques for their incorporation in a finite element head model (FEHM). It is widely known that subject-specific mechanical properties should be assigned for a biofidelic finite element (FE) model of bone. Here, an elasticity-density- strain rate particularized for equine cranial bone was sought, using horse as human skull surrogate. Subjects ranging from 4 to 27 years of age were used to extract samples that were tested in a 3-point bend configuration, at three different speeds in quasi-static conditions to assess changes in the Young’s modulus in that regime. These tests were performed in two different orientations of the transversal plane. Results showed that the equine cranial bone is transversely anisotropic, which differs from what has been found for human cranial bone, making equine not a suitable surrogate for human skull studies, since human samples cannot be obtained in Ireland. The results also showed that the equation for calculating the elastic modulus of equine cranial bone in quasistatic conditions, depends on the orientation of the bone and the strain rate of the tests -which reflects the viscoelastic behaviour of the bone-; but unlike for other bones, for the equine parietal bones tested and the range of BMD measured, the bone mineral density (BMD) does not influence the elastic modulus, neither does age. 70% of the equine cranial bone samples contained Wormian bone (intrasutural bones), which are irregular isolated bones that can appear near the skull sutures and are often considered to be a simple anatomical variant. For these samples, no correlation between Wormian bone and age nor sex was found; nonetheless it seemed to reduce the stiffness of the cranial bone samples, making them more deformable. Then, the human skull’s resistance to fracture was investigated using the concepts of Linear Elastic Fracture Mechanics (LEFM). Human cortical cranial bone samples were harvested from unembalmed frontal, left and right parietal bones and three groups of tests were performed, namely Mode I, Mode II and Mixed-Mode I-II. For Mode I, samples were tested in the Single Edge Notched Beams (SENB) configuration under symmetric 3-point bend tests, while for Mixed-Mode I-II the samples were tested in an asymmetric 3-point bend. For Mode II, 4-point bend tests were carried out. All samples fractured in a brittle fashion. From these tests, reference values of stress intensity factor (KI and KII) and the strain energy release rate (JI, GI, GII, GI-II) for the frontal, left and right and left parietal bone were calculated and it was determined that they are not statistically different, but they do change according to sex and age, and are symmetrical with respect to the sagittal plane. It was also demonstrated that like for other human bones, human cranial bone’s values of fracture toughness are lower for females and vary with age. J-curves and R-curves were determined for those modes, reaffirming that human cortical cranial bone fractures in an unstable way when tested in a quasi-static conditions (17.24 ± 9.74 N for Mode I, 53.53 ± 26.53 N for Mode II, and 94.24 ± 36.98 N for Mixed-Mode I-II). These fracture mechanics properties were introduced in a subject-specific FEHM developed by KU Leuven, producing the first FEHM that includes the fracture toughness of cranial bone. The Extended Final Element Method (XFEM), together with an energy-type damage evolution approach, was used to predict experimental impact tests performed on human heads. This analysis reproduced 74% of the impact load measured during the experiments and showed that the initiation and propagation of the fracture occurred in accordance with the fracture mechanisms reported in the literature. This model was largely influenced by the use of static fracture mechanics properties and neglection of the viscoelastic behaviour of the bone that may affect use of the model to reproduce impact tests. The results of this work can be employed as a new fracture criterion for human cranial bone in FEHMs to predict skull fractures under impact scenarios which will help in the development of improved head protective gears.