PURPOSE: There are no entire maxillary finite element analysis models available, as a base of reference for the dimensions of conventional segment finite element analysis models. The objectives of this study were: (1) to construct a maxillary model derived from a human skull and to investigate the strain distribution around a posterior implant embedded in it; (2) to investigate the usability of conventional segment maxillary models. METHODS: CT DICOM data of a human dried skull maxilla was imported into the Mesh Generation Tools (ANSYS AI environment) and a computer-generated implant-abutment unit was bicortically embedded into it. In this Large model, Von Mises strains under axial and buccolingual loads were then calculated by a finite element program. Moreover, two simplified maxillary segments (Simplified models) were computer-generated and their Von Mises strains were similarly calculated. RESULTS: Although absolute values differed markedly, strain distribution patterns in the cortical bone were similar to those in the Simplified models: high Von Mises strains in the cortical bone concentrated in the sinus floor around the implant apex under axial load, and in the alveolar crest around the implant neck under buccolingual load. CONCLUSIONS: The simplified and segmented three-dimensional finite element models of the human maxilla showed the same locations of the highest equivalent strains as the full maxilla model created from CT DICOM data. If absolute strain values are not of interest, the Simplified models could be used in strain analyses of simulated posterior maxilla for diagnostic suggestions in implant placement.
PURPOSE: There are no entire maxillary finite element analysis models available, as a base of reference for the dimensions of conventional segment finite element analysis models. The objectives of this study were: (1) to construct a maxillary model derived from a human skull and to investigate the strain distribution around a posterior implant embedded in it; (2) to investigate the usability of conventional segment maxillary models. METHODS: CT DICOM data of a human dried skull maxilla was imported into the Mesh Generation Tools (ANSYS AI environment) and a computer-generated implant-abutment unit was bicortically embedded into it. In this Large model, Von Mises strains under axial and buccolingual loads were then calculated by a finite element program. Moreover, two simplified maxillary segments (Simplified models) were computer-generated and their Von Mises strains were similarly calculated. RESULTS: Although absolute values differed markedly, strain distribution patterns in the cortical bone were similar to those in the Simplified models: high Von Mises strains in the cortical bone concentrated in the sinus floor around the implant apex under axial load, and in the alveolar crest around the implant neck under buccolingual load. CONCLUSIONS: The simplified and segmented three-dimensional finite element models of the human maxilla showed the same locations of the highest equivalent strains as the full maxilla model created from CT DICOM data. If absolute strain values are not of interest, the Simplified models could be used in strain analyses of simulated posterior maxilla for diagnostic suggestions in implant placement.