Effects of the Age-Dependent Properties of the Braincase on the Response of the Infant Brain to Impact

INTRODUCTION  
Finite element models of the adult human head have been used to determine intracranial deformation of the brain under traumatic loading conditions [1].  Development of similar models for the infant head has been hindered by the lack of material property data for the pediatric brain and skull.  Recent determination of these properties [6] has facilitated the construction of an idealized infant head model which can be used to demonstrate the effect of age-dependent properties of the braincase on the response of the infant brain to impact. The longterm goal of this research is to quantify the intracranial deformation of the neural and vascular tissue within the brain during traumatic loading and to apply these findings to the development of specialized head injury tolerances for the pediatric population.  
MATERIALS AND METHODS  
A finite element model of the one-month-old infant head was constructed using ANSYS 5.3 (ANSYS inc., Houston, PA). The model consisted of a rigid impactor plate, and a skull/brain with five regions: the cranial bone of the skull. the sutures, the anterior fontanel, the foramen magnum and the brain (Fig. 1). The contact problem was solved using LS-DYNA3D's automatic surface-to-surface contact formulation (LSTC, Livermore, CA).  The model consisted of 12,772 elements and 11,823 nodes, and a zoning study was performed to verify convergence. The cranial bone, suture, fontanel and foramen magnum were modeled using four-node linear shell elements. The brain and impactor were modeled using eight--node linear brick elements.  
An "infant" model and an "adult" model were constructed which were similar in geometry, boundary conditions and loading conditions, but differed in skull material properties. The "infant model" was assigned pediatric material properties for the cranial bone, suture, and brain. In the "adult model," the sutures were assumed completely fused, having properties similar to adult human cranial bone [3].  
Skull material properties are listed in Table 1.  For the infant model, the elastic modulus of cranial bone and suture wre obtained from experiment [6].  remaining infant values are from the literature [4,5].  Fore the adult model, the cranial bone, sutures and anterior fontanel were assigned the properties of adult cranial cortical bone [5].  
 Brain material properties were based on the experimentally determined mechanical response of infant brain tissue [6]. The brain was modeled as a linear viscoelastic solid  

                      G(t) =  Ginf + (G0-Ginf)exp(-bt)  

where Ginf=2.32le-3 MPa, G0=5.99e-3 MPa and b=0.09428 sec.  The brain was assumed incompressible with a bulk modulus of 2110 MPa. The elastic modulus of the foramen magnum was chosen as 100 MPa to simulate the mechanical impedance of the spinal cord to brainstem herniation.  
 The base of the cranial vault was fixed to simulate pure impact loading with no rotation or translation of the head following impact. The brain and the skull were assumed to be displacement compatible with no slip at the interface between the two materials.  
 The impact loading conditions were based on accelerations measured by Duhaime et al. [2] in a biomechanical study of shaken baby syndrome. Peak loading magnitudes of 1000 N and 5000 N were selected to simulate minor and severe impact loads, respectively. Loads were applied to the parietal region of the skull at 450 relative to the vertical axis. The load was applied in a half-sinusoidal, 10 ms pulse.  

RESULTS  
The deformation of the braincase and the resulting intracranial strains were highly sensitive to the structural properties of the cranial vault.  The peak intrusion of the indentor was 150% greater in the infant model than in the adult model for the 5000 N load case (Fig. 2). Impact loading resulted in a diffuse distribution of maximum principal strains in the infant model but a focal/localized distribution in the adult model (Fig.3).  

DISCUSSION  
A finite element model of the one-month-old infant head has been developed which incorporates experimentally determined mechanical properties of infant cranial bone, suture, and brain tissue.  Maximum principal intracranial strains in excess of 0.15 have been associated with functional failure of neural and vascular tissues [7].  The model developed in this study supports the hypothesis that impact loading, commonly associated with focal brain injury in the rigid braincase, may result in diffuse brain injury in the compliant braincase of the infant.   Further research is necessary to more accurately represent the complex three-dimensional geometry of the infant head.  However, the idealized model presented in this study is a convenient tool by which to explore the age-dependent mechanical response of the braincase to impact loading.  

REFERENCES  
1. Bandak et al., SAE Paper No. 942215, 1994.  
2. Duhaime et al. J Neurosurgery, 66: 409-415, 1987  
3. Hubbard,J. Biomech., 4:251-263, 1971.  
4. Kriewall, Am J Obstet Gynecol., 143:707-714, 1982.  
5. McElhaney et al., J Biomech., 3: 495-511,1970.  
6. Thibault, Ph.D. Dissertation, Univ. of Penn., 1997.  
7. Thihault et al., IRCOBI Proc., 191-202,1990.  

ACKNOWLEDGMENTS  
Special Thanks to Tara Khatua of Failure Analysis Assoc., Inc., CDC Grant R49/CCR312712 and the University of Pennsylvania.