PURPOSE:: The purpose of this in vitro study was to evaluate the resistance and patterns of fracture of endodontically treated maxillary premolars (ETPs) restored with different methods. METHODS AND MATERIALS:: Mesio-occluso-distal cavities were prepared in 50 extracted caries-free human maxillary premolars after endodontic treatment. The teeth were divided into five groups (n=10), according to the restorative method. G1: intact teeth (control group); G2: conventional composite resin; G3: conventional composite resin with a horizontal glass fiber post inserted between buccal and palatal walls; G4: bulk-fill flowable and bulk-fill restorative composites; and G5: ceramic inlay. For direct restorations, Filtek Z350 XT, Filtek Bulk Fill Flowable Restorative, and Filtek Bulk Fill Posterior Restorative were used. Indirect restorations were fabricated from a pressable lithium disilicate glass-ceramic (IPS e-max Press) and adhesively cemented (RelyX Ultimate). All specimens were subjected to thermocycling (5°C to 55°C/5000 cycles) and additionally submitted to cyclic loading 50,000 times in an Electro-Mechanical Fatigue Machine. Next, the specimens were subjected to a compressive load at a crosshead speed of 1 mm/min until fracture. The fractured specimens were analyzed to determine the fracture pattern using a stereomicroscope, and then representative specimens were carbon coated to allow for the studying of the fracture surface under scanning electron microscopy. One-way analysis of variance (ANOVA) was used to compare fracture resistance of the groups. The results of fracture patterns were submitted to the Fisher exact test (α=0.05). RESULTS:: All specimens survived fatigue. Mean (standard deviation) failure loads (N) for groups were as follows: G1: 949.6 (331.5); G2: 999.6 (352.5); G3: 934.5 (233.6); G4: 771.0 (147.4); and G5: 856.7 (237.5). The lowest fracture resistance was recorded for G4, and the highest ones were recorded for G2, followed by that of G1 and G3. One-way ANOVA did not reveal significant differences between groups ( p>0.05). The highest repairable fracture rates were observed in G1 (100%) and G3 (80%). CONCLUSIONS:: ETPs restored with conventional composite resin with or without horizontal fiber post, bulk-fill composite, and ceramic inlay showed fracture resistance similar to that of sound teeth. Conventional composite resin restorations exhibited the highest prevalence of unrepairable fractures, and the insertion of a horizontal fiber post decreased this prevalence. Intact teeth showed 100% of repairable fractures. It is difficult to extrapolate the results directly to a clinical situation due to the limitations of this study.