Probabilistic analysis of preload developed in a dental implant system.

摘要

The preload or clamping force between dental implant components is an important factor in the success of the mechanical joint. Although parametric studies of 3-D implant models have been reported concerning developed preload, a complete probabilistic analysis providing the preload probability distribution and sensitivities is absent. This study aimed to examine variation of material properties, surface interactions and input torque in a dental implant system and determine preload developed before the implant was put in service. The goal was to determine the probability of obtaining desired preload values and to identify and quantify significant variables affecting the preload.; Three-dimensional finite element models of a Branemark System 3.75 x 10-mm titanium Mark III implant, a CeraOne titanium abutment, and a Unigrip gold alloy abutment screw were generated using Hyperworks software. Modeling the actual and complete helix design of the abutment screw and mating implant screw threads allowed simulation of screw tightening using the ABAQUS finite element analysis (FEA) software. The Nessus software carried out the Mean Value, Advanced Mean Value, and Monte Carlo probabilistic methods based on random variable distributions and the predicted preload response from FEA. Random variables included material elastic moduli, material Poisson's Ratio, abutment-implant coefficient of friction, thread-implant coefficient of friction, and applied torque. Distributions for random variables were determined from values published in literature and analyzed using the BestFit software package. The probabilistic analysis was carried out for two environments, well-lubricated and dry, thus, producing two probability distributions for preload.; The Advanced Mean Value method results for preload developed in a dry environment produced a mean of 347 N with a standard deviation of 62 N. The Monte Carlo method results for preload in a well-lubricated environment produced a mean of 615.5 N with a standard deviation of 92 N. The probability of obtaining a satisfactory stress in the screw of 60-75% of material yield strength was approximately 0.02% for the dry environment and 56.1% for the well-lubricated environment. Preload is most affected by the coefficient of friction between the abutment screw and the implant threaded surfaces and applied torque for low- to mid-range values, up to 718 N. After 718 N, preload becomes less sensitive to the coefficient of friction between the abutment screw and the implant and more sensitive to material properties, specifically the elastic modulus of the abutment screw. Results show that the degree of lubrication during installation of the abutment screw has a significant effect on the amount of preload developed. The 56.1% success range for well-lubricated environments suggests further experiments are needed to accurately predict the variability in material properties and the coefficient of friction between abutment screw and implant threaded surfaces.