Purpose: To study the force used for tightening tapered one-piece prosthetic abutments and their influence on the removal torque value and stress level of the prosthetic abutment after cyclic loading. Materials and Methods: Fourteen implants and prosthetic abutments were divided into two groups (n = 7): G1, 20 Ncm; and G2, 32 Ncm (manufacturer recommended). A 20-mm T-shaped horizontal bar was adapted to the abutments. A 12-Hz cyclic loading was applied to the specimens in an electrodynamic testing system with the maximum number of cycles set to 106. Specimens were inclined by 15 degrees from the vertical axis, and a 5-mm off-center vertical load was applied to generate a combination of bending and torquing moments on the tapered connections. Progressive loads (from 164.85 to 362.85 N) were applied when the previous sample survived 106 cycles. The paired t test compared the screw removal torque with the initial tightening torque for each group (α = .05). A finite element analysis (FEA) of the mechanical testing analyzed the regions of stress concentration. Results: No specimens failed after 106 cyclic loadings. The mean screw removal torque for both groups was similar to the initial abutment torque value applied for each group (G1, 20.36 ± 8.73 Ncm; and G2, 35.61 ± 6.99 Ncm) (P > .05). FEA showed similar stress behavior for both groups in the study despite the different simulated screw preloads (G1: 200 N; G2: 320 N). The coronal region of the implant body presented the highest strain values in both groups. Conclusion: Tightening tapered one-piece prosthetic abutments at 20 and 32 Ncm maintains a stable connection after cyclic loading. The stresses generated by the different tightening forces during cyclic loading are highest at the coronal level of the connection. Keywords: dental implants, fatigue testing, finite element analysis, torque