Understanding the failure modes and the fracture resistance is critical in evaluating the performance of an adhesive for sternal fixation. In this paper, a fracture mechanics testing methodology was used to assess the adhesion of a bioactive glass-based adhesive to bovine bone in terms of a measured mode I critical strain energy release rate (GIC). Reinforced double cantilever beam (DCB) samples were observed to produce repeatable values of GIC. The measured GIC was found to increase significantly from 5.41 to 12.60 J/m2 with an increase in adhesive thickness from 390 to 990 μm because of the constraint from the two adherends regulating the plastic zone size ahead of the crack. The specimens failed cohesively in all cases demonstrating that there was good adhesion to bone, a condition necessary to restrict micromotion and thus provide rigid sternal fixation when used along with sternal wires. It was also found that when the bone was flooded with liquid during adhesive application a much lower GIC of between 0.69 and 1.15 J/m2 was measured. Overall, the results demonstrate that the fracture mechanics approach can be used to provide a quantitative measure of the adhesion of the bioactive glass-based adhesive to the bone and that the adhesive should only be applied to clean bone in a dry environment.