THE JOURNAL OF PROSTHETIC DENTISTRY VOLUME 87 NUMBER 5have proven their clinical reliability in combination with the metal-ceramic technique. 1,2 The advent of metal-free restorative materials, however, has led to the use of indirect composite or ceramic fixed partial dentures (FPDs) as an alternative to conventional metal-ceramic AFPDs. Better bonding properties to composite cements, more appropriate biomechanical behavior, and enhanced esthetics are expected with the use of composite or ceramic compared to metal alloys. alternative to conventional metal-ceramic adhesive fixed partial dentures (AFPDs). Little information about the adequate restorative material and tooth preparation design for inlay-anchored AFPDs is available to the clinician.Purpose. The purposes of this simulation study were: (1) to use 2-dimensional finite element modeling to simulate stresses at the surface and interface of 3-unit posterior AFPDs made with 6 different restorative materials, and (2) to investigate the influence of 3 different abutment preparation configurations on the stress distribution within the tooth/restoration complex. Material and methods.A mesio-distal cross-section of a 3-unit AFPD was digitized and used to create 2-dimensional models of the periodontal membrane, supporting bone, different restorative materials (gold, alumina, zirconia, glass-ceramic, composite, and fiber-reinforced composite), and different abutment preparation configurations (interproximal slots vs. 2-surface [MO, DO] vs. 3-surface [MOD]). A simulated 50-N vertical occlusal load was applied to the standardized pontic element. The principal stress within the restorative materials, stresses at the tooth/restoration interface, and surface tangential stresses at the level of the pontic were calculated in MPa from the postprocessing files and compared to each other.Results. All materials and tooth preparation design exhibited a similar stress pattern, with a definite compressive area at the occlusal side of the pontic, a tensile zone at the gingival portion of the pontic, and tensile stress peaks in the abutment/pontic connection areas. Among isotropic materials, standard non-reinforced composites exhibited better stress transfer and reduced tensile stresses at the adhesive interface than ceramics and gold. Optimized placement of the glass fibers within the composite resulted in similar stress distribution when tested in 2-surface abutment preparation configuration. There was no detectable influence of preparation design on the behavior of the pontic area. Among all 3 preparation designs, only the DO design exhibited almost pure compression at the interface. Conclusion.