Purpose: A clinical evaluation of the long-term survival rate of three subtypes of a dental tube-in-tube implant with special emphasis on peri-implant bone resorption was conducted and is presented here. Materials and Methods: A retrospective follow-up examination of patients treated with tube-in-tube implants between August 2001 and August 2005 was conducted. For patients available for clinical recall, success criteria according to Albrektsson and Buser were evaluated. Differences in peri-implant bone resorption with regard to implant subtype and depth of implant placement (supracrestal or crestal) were calculated via the Student t test. Results: One hundred twenty-seven patients were provided with a total of 382 Camlog implants. Of the implants placed, 239 were root-line (RL), 129 were screw-line (SL), and 14 were cylindric. One hundred four patients with 318 implants could be contacted. In 59 cases, bony augmentation was performed. After an average time in situ of 32 months (range, 8 to 56 months), 307 of 318 surveyed implants (96.5%) remained in function. Six implants were lost as a result of absence of osseointegration, four following peri-implantitis, and one owing to hypesthesia. Two patients with eight implants have died since 2005, and 21 patients with 47 implants could not be contacted. In 75 patients with 240 implants, clinical success rates were 94% (Albrektsson) and 97% (Buser). A significant difference in bone resorption was found between RL and SL implants (P .0001). Implants that had been placed supracrestally showed significantly less bone resorption than crestally inserted implants, both overall and in subgroup analyses (RL, P = .015; SL, P .0001). Conclusion: The internal tube-in-tube implant system showed promising long-term results. Significantly less bone resorption was seen for RL implants as well as for implants that were placed supracrestally. Parole chiave: bone resorption, dental implants, implant-abutment interface, implant design, long-term study