Biomechanics of shaken baby syndrome
In the first part of this work, an Anthropometric Test Dummy (ATD) was used to obtain torso acceleration data for Shaken Baby Syndrome (SBS). These data were used to drive computational simulations of SBS, in studies of the effect of neck stiffness and head-torso impact on injury risk. Finally, physical models were used to investigate the strain induced in brain tissue during shaking. Clinical literature describes victims of Shaken Baby Syndrome (SBS) as young infants with life-threatening brain injuries, and poor long-term outcome. However, biomechanical studies using ATI)s to study head motion during shaking have been inconclusive about the capacity for shaking alone to cause these injuries 11,21. This work comprises a series of investigations into these conflicting findings. Torso acceleration data for SBS, obtained using a specially constructed ATD, were found to be consistent with previous findings. The data were used to simulate shaking in computational studies of SBS, using Rigid Body Models (RBM) of the infant head and neck. Parametric studies were used to investigate the importance of neck stiffness in assessing the injury capacity of SBS, and showed that in order to exceed current injury criteria for SBS, impact was required. Head torso impact was then simulated, and although this resulted in higher injury risk than shaking alone, criteria for injuries associated with SBS were not reached. Since these investigations did not predict brain injury in cases of SBS without impact, the origins of injury criteria were reviewed. It was found that they are derived from single high energy events, which is distinct from the type of motion in SBS. In order to establish if cyclic, low-energy motion contributes to brain injury in SBS, Physical Continuum Modelling was used to study strain in brain tissue during shaking. A test rig was constructed to shake silicone gel models, and high-speed video used to capture the motion of optical markers with in the gel. Their movement was tracked using optical flow methods, and Green-Lagrangian strain derived by tensor algebra. No evidence was found to indicate a build up in strain between cycles, but published critical strains for damage to neural tissue were exceeded. Although shaking alone was not found not induce head motion in excess of brain injury criteria, tissue damage criteria were exceeded. The application of current brain injury criteria to SBS maybe therefore be inappropriate.