The fixation of osteosynthesis screws remains a severe problem for fracture repair among osteoporotic patients. Polymethyl-methacrylate (PMMA) is routinely used to improve screw fixation, but this material has well-known drawbacks such as monomer toxicity, exothermic polymerization, and nonresorbability. Calcium phosphate cements have been developed for several years. Among these new bone substitution materials, brushite cements have the advantage of being injectable and resorbable. The aim of this study is to assess the reinforcement of osteosynthesis screws with brushite cement. Polyurethane foams, whose density is close to that of cancellous bone, were used as bone model. A hole was tapped in a foam sample, then brushite cement was injected. Trabecular osteosynthesis screws were inserted. After 24 h of aging in water, the stripping force was measured by a pull-out test. Screws (4.0 and 6.5 mm diameter) and two foam densities (0.14 and 0.28 g/cm3) were compared. Cements with varying solid/liquid ratios and xanthan contents were used in order to obtain the best screw reinforcement. During the pull-out test, the stripping force first increases to a maximum, then drops to a steady-state value until complete screw extraction. Both maximum force and plateau value increase drastically in the presence of cement. The highest stripping force is observed for 6.5-mm screws reinforced with cement in low-density foams. In this case, the stripping force is multiplied by 3.3 in the presence of cement. In a second experiment, cements with solid/liquid ratio ranging from 2.0 to 3.5 g/mL were used with 6.5-mm diameter screws. In some compositions, xanthan was added to improve injectability. The best results were obtained with 2.5 g/mL cement containing xanthan and with 3.0 g/mL cements without xanthan. A 0.9-kN maximal stripping force was observed with nonreinforced screws, while 1.9 kN was reached with reinforced screws. These first results are very promising regarding screw reinforcement with brushite cement. However, the polyurethane foam model presents noninterconnected porosity and physiological liquid was not modelized.