RESUMO
This study investigated the influence of mechanical cycling on screwed-in and tappedin implants restored with screw-retained metallic crowns. Three implant-abutmentcrown systems were evaluated: T1 (multi abutment) and T2 (standard abutment) received tapped-in abutments and S received a screwed-in abutment. The specimens were subjected to two million cycles of 0-150 N load, at 2 Hz, 30° inclination in a dry medium, and torque evaluation. Survival and removal torque were analyzed using chisquare, ANOVA, followed by Tukey's test. Differences between installation and removal torque were determined using a T-test for dependent samples. Analyses were performed in SPSS, considering α = 0.05. All specimens survived mechanical cycling in S, 40% in T1, 80% in T2 (p=0.008). Failures occurred due to loosening of the crown screw. A significant decrease in torque (p=0.000) was found. Group T1 had the lowest removal torque (1.6 ± 0.84 N.Cm²), followed by T2 (3 ± 1.49 N.Cm²) and S (6.3 ± 1.16 N. Cm²), and a statistical difference was found between Groups T1 and S. Both types of implantabutment connections were stable and can be considered for rehabilitative treatment, but failure and removal torque were influenced by the design of prosthetic abutment. Crowns were more susceptible to becoming loose in tapped-in systems.