The surgical repair of an abdominal wall defect may be complicated by infection. We examined the in vitro and in vivo behavior of Staphylococcus aureus (Sa) and Staphylococcus epidermidis (Se) when placed in contact with three collagen bioprostheses. For the in vitro study, 1 cm2 fragments of the collagen meshes (Collamend®, Surgisis®, and Permacol®) and a control polytetrafluoroethylene mesh, Preclude®(ePTFE) were incubated on blood agar plates inoculated with Sa or Se. In the in vivo study, 2 partial 3 × 3 cm defects were created in the abdominal wall of 72 rabbits and infected with a suspension-containing 106 Colony-forming unit (CFU) of Sa or Se. The defects were then repaired using the above materials. At 14 and 30 days postimplant, mesh specimens were obtained for histological, morphometric, and biomechanical analysis. The incubated collagen meshes showed significantly greater bacterial loads than the ePTFE. In vivo, large abscesses comprised of bacteria (Sa/Se), detritus and white cells could be seen 14 days post-implant. At 30 days, the bacterial infiltrate was reduced in the Se group. In conclusion, in presence of bacterial contamination, no benefits were observed of the use of the collagen bioprostheses tested over the use of a non porous ePTFE mesh (Preclude®). © 2013 Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater, 102B: 366–375, 2014.