The aim of this study was to determine the in vitro shear bond strength (SBS) and tensile bond strength (TBS) of 45 metal brackets bonded with Transbond XT to bovine enamel. The SBS was determined by loading the short and the long sides of the bracket base. Testing took place after storage of the specimens for 72 hours in water at 37°C. Fractures were analysed with the adhesive remnant index (ARI) and scanning electron microscope (SEM). The stresses in the system were analysed with finite element (FE) analysis models of the experimental set-up to identify the initial fracture point and the stress distribution at fracture. Statistical analysis of bond strengths was performed using analysis of variance (ANOVA) and the Tukey’s post hoc test (P < 0.05). The ARI scores were analysed using Kruskal–Wallis one-way ANOVA on ranks.
ANOVA showed significant differences between the three experiments. Loading the short side of the bracket resulted in the highest average bond strength. Tensile loading gave the lowest results. FE models supported the bond strength findings and SEM. FE analysis revealed peak stresses in the cement during loading, confirming that shear testing is sensitive to loading angles. The stress distribution over the bracket–cement–enamel system is not homogeneous during loading. Fractures are initiated at peak stress locations. As a consequence, the size of the bonding area is not predictive of bond strength. The bracket design and the mode of loading may be of greater relevance.