In Vitro mechanical Analysis of Full-Arch Mandibular Implant-Supported, Complete Fixed Prosthesis Retainer Screws After Cyclic Loading

Abstract

The use of implant-retained and supported prostheses has become a very successful treatment for completely edentulous patients. One of the most common fixed solutions involving implants consists of 5 to 7 implants supporting a framework upon which either porcelain or prefabricated acrylic resin denture teeth are added. A screw is utilized to attach the framework/prosthesis to the implants. Screw loosening is the second most common clinical complication in the implant-prosthesis system. If clinicians fail to detect worn or loose retaining screws, prosthetic fracture could occur, leading to more complicated, time consuming, and expensive repairs. Unfortunately, there is no established parameter that indicates when to expect these complications, and there is no proven recall-maintenance protocol to prevent them. The aim of this study is to examine and compare differences among de-torque values and prosthetic retention screws, using a simulated 5 implant-supported, mandibular complete fixed prosthesis. Material and Methods: Nine groups, each with its respective control, using five Nobel Biocare implants and a milled titanium framework were fabricated, assembled and tested. Dynamic loading was p on the performed tested groups through a custom made loading device for anterior, posterior, and distal cantilevered segments of the prosthesis, calculated to simulate clinical usage time. Removal of screws after 2 years of simulated oral function was performed. Before and after testing, screws were evaluated with a Scanning Electronic Microscope (SEM), for presence of debris, thread striations and homogeneity. Control groups remained unloaded for the same time the loaded groups were tested. Results: Comparisons of the difference between initial tightening torque and de-torque screw values were performed between loaded/unloaded groups and with respect to implant position. The interaction between loaded and position was significant (p=0.002). The comparison between loaded/unloaded groups was not statistically significant (p=0.518). Loaded and unloaded groups were compared separately at each of the 5 implants position, which showed a significant difference (p=0.0002, α=0.001). The sequencing effect was only seen in the control groups and thus would only be relative to framework insertion. The sequence effect was found to be overcome by from loading and resulted in a totally different position related to screw tightness. Within the limitations of this in vitro study, it was concluded that sequence of torque application could play a role in the preload of screws even with a passive fit, regardless the load applied.