Effect of post-osseointegration loading magnitude on the dynamics of peri-implant bone: a finite element analysis and in vivo study.

PURPOSE: Much research has been invested in determining the effects of postoperative loading of implants and whether this loading contributes to implant failure, but the issue remains controversial. The present study aimed to elucidate whether cyclic lateral loading of an implant causes bone resorption or bone formation at various loading magnitudes, using a finite element method (FEM) and peri-implant morphologic and morphometric analyses. METHODS: An FEM model was created using Digital Imaging and Communications in Medicine (DICOM) data of rabbit tibia. For the animal study, implants were inserted into rabbit tibia and, after osseointegration, were subjected to lateral cyclic loading of 20N, 40N or 60N. RESULTS: Bone-implant contact was significantly higher in both 40N and 60N groups. Bone-abutment contact (BAC) was extraordinarily observed in all experimental groups. Bone height was higher than the implant platform level at higher levels of loading (60N). Among the three experimental groups, those receiving 40N loading had the highest bone height and BAC. Larger BAC values were observed on the compressive side than the tensile side. CONCLUSIONS: Peri-implant bone formation was enhanced with increased loading, with bone formation predominantly on the compressive side. BAC was highest in the 40N group, implying existence of a loading threshold for peri-implant bone formation and resorption.

A comparison of the peri-implant bone stress generated by the preload with screw tightening between 'bonded' and 'contact' model.

A number of finite element analyses (FEAs) for the dental implant were performed without regard for preload and with all interfaces 'fixed-bonded'. The purpose of this study was comparing the stress distributions between the conventional FEA model with all contacting interfaces 'fixed-bonded' (bonded model) and the model with the interfaces of the components in 'contact' with friction simulated as a preloaded implant (contact model). We further verified the accuracy of the result of the FEA using model experiment. In the contact model, the stress was more widely distributed than in the bonded model. From the model study, the preload induced by screw tightening generated strain at the peri-implant bone, even before the application of external force. As a result, the bonded model could not reproduce the mechanical phenomena, whereas the contact model is considered to be appropriate for analysing mechanical problems.