Fatigue resistance of repaired restorative composites

Objectives: Repairing restorative composite fillings with small defects is a popular treatment option mainly because of its tissue-saving nature. However, there is limited information on the ability of repaired restoration to serve as a load-bearing restoration. The aim of this study was to evaluate the fatigue performance of repaired restorative resins.
Methods: A four-point bending configuration with a specimen size of 2x2x12 mm was used. Filtek Z250 (3M ESPE) was used as a substrate in the study. The specimens were polymerized according to the manufacturer’s recommendations and aged in distilled water for one month at room temperature before repairing. The test groups were: Control (C, unrepaired), Hydrofluoric acid application (9%, 90s) (HF) and Scotchbond Multipurpose (3M ESPE), Sandblasting/silane application (Cojet and ESPE-Sil, 3M ESPE) and Scotchbond Multipurpose (S), only Scotchbond Multipurpose (SM), only Scotchbond Universal (3M ESPE) (SU). The specimens were subjected to either quasi-static (N=10) or cyclic (N=20) four-point flexure testing using a stress ratio of 0.1 and a frequency of 4 hertz to failure in fully hydrated conditions. The specimens were loaded until 1.2 million cycles and the test was stopped if the materials did not fail. Data was analyzed using one-way ANOVA for quasi-static and Wilcoxon Rank Sum test for fatigue data (α=0.05).
Results: Under quasi-static loading, the flexural strength of the materials were 172,75 MPa for C, 58,45 MPa for HF, 81,14 MPa for S, 50,1 MPa for SM and 85,12 MPa for SU. Control, SU and S were statistically higher than others (p=0,01). For fatigue data, unrepaired specimens were significantly higher than repaired ones and among repaired ones, sandblasting-silane group showed significantly higher performance than the other groups (p<0.05).
Conclusions: Repaired composites have significantly lower fatigue resistance and sandblasting with silane application is the most successful repair strategy.