Abstract
BACKGROUND: Abutment screw loosening (ASL) is the most frequent mechanical complication in dentistry, leading to prosthetic instability and biological risks. Preload, generated during screw tightening, is critical for maintaining stability but is influenced by torque application, screw geometry, and cyclic loading. METHODS: This in vitro study compared torque loss between two implant systems (Osstem TSIII and KSIII) with different abutment screw designs. Fifty implant-abutment assemblies (n = 5 per torque group) were tested under tightening torques of 20, 25, 30, 35, and 40 Ncm. Initial removal torque (T1) was measured 5 min after tightening, followed by cyclic loading (150 N, 14 Hz, 100,000 cycles). Post-fatigue removal torque (T2) was then recorded, and torque loss rate (%) was calculated. Independent t-tests and a one-way ANOVA were used for statistical analysis. RESULTS: KSIII consistently exhibited higher T1 and T2 values than TSIII across all torque levels (p < 0.05). The torque loss rate for TSIII ranged from 36.5% (35 Ncm) to 51.8% (20 Ncm), showing a torque-dependent trend (p < 0.05). In contrast, KSIII maintained torque loss rates below 25% at all levels, with no significant differences between torque groups (p > 0.05). On average, torque loss in TSIII was approximately 2.5-3.0 times higher than in KSIII. CONCLUSIONS: The KSIII system demonstrated superior biomechanical stability, with significantly lower torque loss compared with TSIII, independent of torque level. Clinically, these findings suggest that the KSIII system may reduce the incidence of screw loosening and associated complications. A tightening torque of approximately 35 Ncm appeared to provide the most stable preload. Long-term in vivo studies are warranted to confirm these results under clinical conditions.