Purpose: To compare the strain development of conventionally cast three-unit fixed dental prostheses (FDPs) and FDPs fabricated through computer-aided design/computer-assisted manufacture (CAD/CAM) using the strain gauge technique. Materials and Methods: A two-implant situation was transferred to an acrylic resin replica, and strain gauges were attached mesially and distally adjacent to the implants. Seven groups (n = 10) of cement-retained FDPs were manufactured from all restorative materials available for the Etkon CAD/CAM system (Straumann CADCAM). Three groups (n = 10) of conventionally cast screw- and cement-retained superstructures served as controls. Strain development during FDP fixation was recorded, and the logarithm of the absolute strain values was used for statistical analysis (multivariate analysis of variance with Pillai's trace; α = .0033). Results: Significantly higher strain values for screw-retained FDPs were found than for all other restorations (P .0000). Conventionally cast cement-retained restorations showed significantly higher strain levels as compared to CAD/CAM restorations fabricated from titanium (P = .0032) and green machined zirconia (P = .0026). Screw-retained superstructures bonded to gold cylinders showed significantly lower strain levels than FDPs made from titanium (P = .0001) and green machined zirconia (P = .0012). No significant differences between the various groups of CAD/CAM restorations could be detected, except that polyamide-resin restorations showed significantly higher strain levels than InCeram Zirconia restorations (P .0000). Conclusions: CAD/CAM-fabricated restorations show fit that is at least as good as that seen for conventionally fabricated superstructures. The choice of restorative material seems to have only a minor effect on the passivity of fit. Schlagwörter: cement retention, computer-aided design, computer-assisted manufacture, passive fit, screw retention, strain development