Peri-implant bone reactions to immediate implants placed at different levels in relation to crestal bone. Part I: A pilot study in dogs

Purpose: The aim of the present study was to evaluate bone remodeling and bone-to-implant contact (BIC) after immediate placement at different levels in relation to the crestal bone of Beagle dogs. Materials and methods: The mandibular bilateral second, third and fourth premolars of six Beagle dogs were extracted and six implants were immediately placed in the hemi-arches of each dog. Randomly, three cylindrical and three tapered implants were inserted crestally (control group) and 2mm subcrestally (experimental group). Both groups were treated with a minimal mucoperiosteal flap elevation approach. A gap from the buccal cortical wall to the implant was always left. Three dogs were allowed a 4-week submerged healing period and the other three an 8-week submerged healing period. The animals were sacrificed and biopsies were obtained. Biopsies were processed for ground sectioning. Histomorphometric analysis was carried out in order to compare buccal and lingual bone height loss, and BIC between the two groups. Results: All implants osseointegrated clinically and histologically. Healing patterns examined microscopically at 4 and 8 weeks for both groups (crestal and subcrestal) yielded similar qualitative bone findings. The distance from the top of the implant collar to the first BIC in the lingual crest (A-Lc) showed a significant difference (P=0.0313): 1.91 ± 0.2mm in the control group and 1.08 ± 0.2mm in the experimental group. There was less bone resorption in subcrestal implants than crestal implants. The mean percentage of newly formed BIC was greater with the cylindrical implant design (46.06 ± 4.09%) than with the tapered design (32.64 ± 3.72%). Conclusion: These findings suggest that apical positioning of the top of the implant does not jeopardize bone crest and peri-implant tissue remodeling. However, less resorption of the Lc may be expected when implants are placed 2mm subcrestally.