When an oral implant system is loaded in vivo by biting forces and moments, marginal bone loss can sometimes occur around the im plant Such bone loss can predispose the entire system to failure. Many researchers have identified the bone quality around the implant as one of the most important factors related to failure of an implant system. Nevertheless, to date, few experiments have tried to determine the properties of interfacial bone around an im plant One of the reasons for this is that the area of bone-implant interface is quite small, which makes it difficult to apply conventional mechanical testing methods to measure the properties of the interfacial bone. Consequently, it is necessary to develop new test methods to measure properties of interfacial bone at the microstructural level, especially to allow accurate bone properties in improved FE models.

The goals of this study are to identify important factors that affect the stability of bone-implant system during healing periods after surgery and to develop better input data on bone properties for use in FE models. To achieve these goals, the present study conducted a series of experiments w ith newly developed methods to estimate interfacial bone properties at the small region of bone-implant interface. Based on the values determined by the experiments, finite element analyses (FEA) were performed to simulate the mechanics of the implant system. Two healing periods of an implant system were considered:​ before and after healing. The before healing case represented the implant system immediately after implantation, and the after healing case represented the situation of the implant system at one year healing after implantation. The essential differences between the two conditions were (1) non-bonding and bonding at the bone-implant interface for the before healing case vs. the after healing case, respectively, and (2 ) different interfacial bone properties adjacent to the implant.

First, to help identify the most critical risk factors possibly affecting the micromechanical environment around the bone-implant interface, a FEA was conducted. The properties of interfacial bone were assumed to change with distance from the implant in the two different healing periods. The results showed that the strain magnitudes were affected by 1 ) the different interfacial bone properties, 2 ) microstructure of bone such as Haversian canals, and 3) non-bonding/bonding conditions at the bone-implant interface. The risk factors estimated by this FEA were then examined in more detail in a series of micromechanical experiments.

To collect more accurate mechanical properties of the interfacial bone, the newly developed nanoindentation technique was used to measure properties as a function of distance from the implant. For the before healing case, while no differences in the elastic moduli of interfacial bone were measured, in the after healing case, the modulus of the interfacial bone was found to be about half of the value for normal bone.

As another approach to investigate the interfacial bone properties, a new composite model was developed to compute the overall properties of interfacial bone. The new model assumed that bone has pores such as Haversian canals as well as a solid part. The present model was based on Mori-Tanaka scheme, which can approximate the overall properties of bone in a transversely isotropic manner. The analysis showed that the computed magnitudes of the overall bone properties were similar to values measured by traditional mechanical tests.

As previously suggested from both FE and composite models for interfacial bone, the inhomogenous microstructures of bone (with pores and Haversian systems) are expected to affect the local microstrain distributions in bone under loading. Therefore, to see how strain levels could vary at the microstructural level when bone is loaded, image analysis was conducted using a method based on displacement machine vision photogrammetry (DISMAP). A microtensile testing apparatus was built to handle a small-sized bone sample that was tested and imaged under the reflected light microscope. Results showed that at the microstructural level in bone under uniaxial tension, magnitudes of microstrain were 4 to 14 times higher than the overall strain as computed by using an average strain approach. Also, microstrains in bone were shown to be concentrated around the threads of implant in bone and at other strain concentrating features within the bone’s microstructure.

To illustrate the use of data collected from the previous parts of this thesis, a more comprehensive FE model was developed for the bone-implant interface. The interfacial bone properties were assigned with the results from nanoindentation tests and Mori-Tanaka model for bone. It was found that the distributions of microstrains in the interfacial bone was sim ilar to the results of a DISMAP image analysis. Moreover, as a plausible mechanism to describe the crestal bone loss sometimes seen around loaded implants, a FE model showed the possibility of a positive feedback mechanism involving high strains as a trigger for bone remodeling. These simulations involving interface - micromechanics demonstrated the usefulness of data collected in this thesis.