Previous attempts by researchers to predict the fatigue behavior of bone cement have been capable of predicting the location of final failure in complex geometries but incapable of predicting cement fatigue life to the right order of magnitude of loading cycles. This has been attributed to a failure to model the internal defects present in bone cement and their associated stress singularities. In this study, dog-bone-shaped specimens of bone cement were micro-computed-tomography scanned to generate computational finite element (FE) models before uniaxial tensile fatigue testing. Acoustic emission (AE) monitoring was used to locate damage events in real time during tensile fatigue tests and to facilitate a comparison with the damage predicted in FE simulations of the same tests. By tracking both acoustic emissions and predicted damage back to scans, barium sulfate agglomerates were found not to be significant in determining fatigue life of specimens. Both the experimental and numerical studies showed that diffuse damage occurred throughout the gauge length. A good linear correlation (, ) was found between the experimental and the predicted tensile fatigue life. Although the FE models were not always able to predict the correct failure location, damage was predicted in simulations at areas identified as experiencing damage using AE monitoring.