IMPACT BIOMECHANICS OF THE ZYGOMATIC COMPLEX Narayan Yoganandan, Ph.D., Frank Pintar, Ph.D., John Reinartz, M.S., Anthony Sances, Jr., Ph.D. Department of Neurosurgery, Medical College of Wisconsin, Milwaukee, WI 53226 Veterans Administration Medical Center, Milwaukee, WI 53295 and Department of Biomedical Engineering, Marquette University, Milwaukee, WI 53233 This study was undertaken to describe the impact biomechanics of the human zygomatic complex and the resulting pathologic changes to the facial skeleton. Fifteen fresh human cadavers were used. Specimens were prepared by rigidly fixing the isolated heads in a neurosurgicai halo ring. A specially designed vertical drop impact test system was used to suitably hold the specimens and dynamically load either zygoma by contacting the junction of the left lower spoke and rim of the steering wheel. The wheels were placed at 30 degrees with the horizontal. Energy absorbing (EA) and non energy absorbing (NEA) wheels were used in the study. A six-axis load cell placed under the hub of the steering wheel measured the generalized force histories. A system of lineary and rotary potentiometers placed below the impact site measured the deformations, and a triaxiai accelerometer on the skull measured the accelerations. Pre-test and post-test, x-rays, 2-D and 3-D CT’s, and defieshed skulls documented the pathology. Facial trauma was classified based on the tripianar strut classification. There was a good correlation between the 2-D and 3-D CT, and defleshed skulls. Fractures correlated well with peak transformed forces at the impact site, while a weak correlation was found with the contact areas and deformations. Further, the head injury criterion (HIC) had a weak correlation with facial fracture emphasizing that HIC is not a good predictor for facial injury. 715 THEORETICAL DETERMINATION OF THE APPROPRIATE FREQUENCY CONTENT OF STRAIN FOR THE OPTIMAL INFLUENCE ON BONE ADAPTATION Kenneth J. McLeod and Clinton T. Rubin Musculo-Skeletal Research Lab Department of Orthopaedics, SUNY at Stony Brook, NY 11754-8181 The maintenance of bone mass is known to be dependent on the dynamic mechanical environment of the bone. However, the specific parameters of the mechanical load which regulate bone remodeling remain poorly defined. in an effort to identify those components which represent osteoregulatory stimuli, we have compared the temporal characteristics of specific mechanical loading patterns to their efficacy in preventing the bone loss of disuse, or in initiating new bone formation. Fourier analysis techniques were used to obtain the induced strain spectrum arising from the loading of a bone preparation with a symmetric trapezoidal waveform. induced spectral energy was then compared to the efficacy of these same loading patterns to prevent bone loss in the isolated avian ulna model of disuse osteopenia. The best correlation between induced strain energy and efficacy is obtained within a relatively “high” frequency range (13-40 Hz). Induced energy within this range demonstrates appropriate dose dependency at both low and high strain levels, and as well, provides an indication of a source of the variability in empirical data seen at high peak strain levels. As only a small portion of the total induced energy is encompassed in the high frequency range, the analysis implies that extremely small strains at appropriately high frequencies may be sufficient to regulate the remodeling response. The apparent ability of the very small “high” frequency components of applied strain to regulate physiologic bone remodeling supports the common premise that the actual signaliing mechanism is tied to a derivative process of the strain such as electrokinetic or piezoelectric currents. COMPUTER AIDED SIMULATION OF GROWTH AND ITS APPLICATION TO BIOLOGICAL AND ENGINEERING STRUCTURES C. Mattheck, S. Burkhardt, D. Erb Kernforschungszencrum Karlsruhe, Institut fiir Materialund FestkErperforschung IV 7500 Karlsruhe, West Germany A new method of Computer Aided Shape Optimization is shown which copies the method of self-optimization of trees by growth. That means that the Finite Element structure of the mechanical structure to be optimized is allowed to swell (growth) at places of high local stresses and not to swell (or as an option even to shrink) at places of lower stresses. The method works well, as many examples (tree branch-stem joints, wound healing in trees, optimized engineering components) show. Furthermore it is shown that the design of some biological structures constitutes a mechanical optimum. The method is working well for twoas well as for threedimensional structures. It is based on a 'misuse' of the option of volumetric swelling in the FEM-code ABAQUS. The main secret is that only a thin surface layer of the FEM-structure is allowed to swell (growth) which is equivalent to the actual growth ring in trees which also alone adapts to the actual loading.