Objectives: To assess the influence of restorative materials and load cusps on the fatigue resistance of endodontically treated molars. Method and Materials: Thirty extracted molars received root canal treatment followed by a standardized tooth preparation (3-mm cuspal reduction and immediate dentin sealing). Twenty Cerec 3 overlays (Sirona Dental Systems) were milled in the ceramic Vita MKII block (Vident; groups MKIIGL and MKIIGL-Z, oven-glazed), and 10 restorations were duplicated with a composite resin (Miris 2, Coltène/Whaledent; group M2). The fitting surfaces of the restorations were hydrofluoric acid etched (porcelain only) and silanated. Preparations were airborne-particle abraded and etched. All restorations were luted with preheated Filtek Z100 (3M ESPE) and subjected to cyclic isometric chewing (5 Hz) starting at 200 N (5,000 cycles), followed by stages of 400, 600, 800, 1,000, 1,200, and 1,400 N at a maximum of 30,000 cycles each. A stainless steel load sphere was used for groups MKIIGL and M2, while a composite resin load sphere was used in group MKIIGL-Z. All samples were loaded until fracture or to a maximum of 185,000 cycles. Groups were compared using the Kaplan-Meier survival curves (P = .05). Results: None of the molars restored with porcelain withstood all 185,000 loading cycles (survival = 0%). The mean fracture load for MKIIGL was 1,060 N and for MKIIGL-Z, 1,280 N. In group M2, the survival rate was 50%. The rate of fracture below the CEJ was 40%, 30%, and 20% for MKIIGL, MKIIGL-Z, and M2, respectively. Conclusion: Miris 2 overlays showed higher fatigue resistance than MKII porcelain (P = .01) when loaded with a stainless steel antagonist. (Quintessence Int 2009;40:729–737)