This study evaluates the load fatigue performance of different abutment-implant connection implant types—retaining-screw (RS) and taper integrated screwed-in (TIS) types under 3 applied torque levels based on the screw elastic limit. Three torque levels—the recommended torque (25 Ncm), 10% less, and 10% more than the ratio of recommended torque to screw elastic limits of different implants were applied to the implants to perform static and dynamic testing according to the ISO 14801 method. Removal torque loss was calculated for each group after the endurance limitation was reached (passed 5 × 106 cycles) in the fatigue test. The static fracture resistance results showed that the fracture resistance in the TIS-type implant significantly increased (P < .05) when the abutment screw was inserted tightly. The dynamic testing results showed that the endurance limitations for the RS-type implant were 229 N, 197 N, and 224 N and those for the TIS-type implant were 322 N, 364 N, and 376 N when the screw insertion torques were applied from low to high. The corresponding significant (P < .05) removal torque losses for the TIS-type implant were 13.2%, 5.3%, and 2.6% but no significant difference was found for the RS-type implant. This study concluded that the static fracture resistance and dynamic endurance limitation of the TIS-type implant (1-piece solid abutment) increased when torque was applied more tightly on the screw. Less torque loss was also found when increasing the screw insertion torque.