Objective To evaluate the effect of different subgingival root exposure methods, namely, crown lengthening or forced eruption, and different ferrule lengths on fracture resistance of a residual root restored with a carbon fiber post-and-core system. Methods Fifty-sixextracted endodontically-treated mandibular first premolars were sectioned 1.0 mm coronal to the buccal cementoenamel junction. All the models were divided randomly into seven groups that each consist of eight roots. Group A was given non-ferrule as control. Simulated crown lengthening was performed for the dentin ferrule design in the cervical tooth structure for Groups B, C, and D with a ferrule length of 1.0, 2.0, and 3.0 mm, respectively. Simulated forced eruption was performed with a ferrule length of 1.0, 2.0, and 3.0 mm in the cervical tooth structure for Groups E, F, and G, respectively. After restoration with prefabricated carbon fiber post-and-core system, each specimen was embedded in a self-cured acrylic resin block from 2.0mm apical to the margins of a cast Ni-Cr alloy crown, then loaded at 150° from the long axis in a universal testing machine at a crosshead speed of 1.0mm•min-1 until fracture. Data of failure loads and fracture mode were recorded and analyzed. Results Mean failure loads for Groups A to G were (1.13±0.15), (1.27±0.18),(1.02±0.11), (1.05±0.12), (1.63±0.14), (1.92±0.19), and (1.93±0.15) kN, respectively. The effects of root exposure method and ferrule design differed significantly,and two factors exhibited significant interaction(F=33.396,P<0.0001). When the ferrule lengths
were the same, fracture loads in the simulated forced eruption groups were significantly higher than those of the simulated crown lengthening groups (P<0.001). No statistical differences in fracture mode were observed among all groups (P=1.00). Conclusion More than 1.0 mm ferrule placement on the apical crown margin by using forced eruption significantly increases fracture resistance.