Purpose: This study sought to determine an optimal postinsertion retorque protocol to maintain implantsuprastructure joint stability after being subjected to a 1-year in vitro masticatory simulation. Materials and Methods: Ten mandibular implant-supported dentures were manufactured and 20 epoxy resin models were obtained for two fit levels: passive fit and misfit. Eight groups (n = 20) were created on the basis of vertical fit (passive or misfit) and prosthetic screw material (titanium or gold). The single-screw test was performed and the vertical misfit was quantified using an optical microscope. Loosening torque was measured after simulations of two types of clinical use: (1) 6 months of use, torque loosening, retightening, another 6 months of use, and loosening torque; and (2) 1 year of use followed by loosening torque. Data were analyzed by means of two-way analysis of variance and the Tukey test. Results: The factors of cycling time and condition use (fit level and screw type), as well the interaction between these factors, significantly influenced the loosening torque (P .05). After 6 months and another 6 months of clinical use simulation, titanium screws showed higher loosening torque values than did gold screws for the same fit level (P .05). After 1 year of clinical use simulation, titanium and gold screws in passively fit dentures showed higher loosening torque values than they did in misfit dentures (P .05). The titanium screws presented a decrease in the loosening torque after 1 year in misfit dentures. Conclusions: The stability of titanium screws was higher than that of gold screws after 6 months of simulation because of their lower plastic deformation. When the cycling time was analyzed, titanium screws were less stable after 1 year of simulation because of loss of torque in the presence of misfit. Schlagwörter: cyclic loading, implant prosthesis, loosening torque, prosthetic screw