Purpose: To evaluate the fracture strength of endodontically treated molars filled with different composite resins, with or without fiber reinforcement.

Materials and methods: A total of 60 intact mandibular molars were selected and endodontically treated. A standardized mesio-occlusal-distal cavity was prepared with cervical margins 1 mm above the cementoenamel junction and oral and buccal walls with 1.5-mm remaining thickness. Universal adhesive was used in all specimens in etch-and-rinse mode. Specimens were divided into four groups (n = 15 each) according to restoration technique: CSM group, adhesive overlay with hybrid ceramic (CeraSmart, GC); EXP group, direct composite restoration (Essentia U, GC) and everX Posterior (GC) as core material; ESU group, direct composite restoration (Essentia U); and EST group, direct composite restoration (Essentia U) reinforced with horizontal bidirectional glass fibers placed over the pulpal chamber floor (everStick-Net, GC). After 7 days of water storage, samples were loaded until fracture using a universal testing machine. The maximum breaking loads were recorded in Newtons (N), and statistical analysis was then conducted with two-way analysis of variance and post hoc Tukey test. Fragments were analyzed using scanning electron microscopy.

Results: Mean fracture resistance was: CSM = 1,428.9 ± 316.90 N; EXP = 1,874.57 ± 299.47; ESU = 1,557.44 ± 355.65; and EST = 1,870.27 ± 145.11. CSM and EXP showed the highest strength values when compared to the other groups. The variable "fiber insertion" did not significantly alter the fracture resistance. Origins of the fractures were always located on the occlusal surface, mainly from the major contact loading area.

Conclusion: The use of everX Posterior showed great improvement in fracture resistance, similar to cuspal coverage restoration. However, only short fiber-reinforced composite resins showed a favorable fracture pattern.