Objective: To study whether osseointegration once established following implant placement in a fresh extraction socket may be lost as a result of tissue modeling.
Material and methods: Seven beagle dogs were used. The third and fourth premolars in both quadrants of the mandible were used as experimental teeth. Buccal and lingual full-thickness flaps were elevated and distal roots were removed. Implants were installed in the fresh extraction socket. Semi-submerged healing of the implant sites was allowed. In five dogs, the experimental procedure was first performed in the right side of the mandible and 2 months later in the left mandible. These five animals were sacrificed 1 month after the final implant installation. In two dogs, the premolar sites on both sides of the mandible were treated in one surgical session and biopsies were obtained immediately after implant placement. All biopsies were processed for ground sectioning and stained.
Results: The void that existed between the implant and the socket walls at surgery was filled at 4 weeks with woven bone that made contact with the SLA surface. In this interval, (i) the buccal and lingual bone walls underwent marked surface resorption and (ii) the height of the thin buccal hard tissue wall was reduced. The process of healing continued, and the buccal bone crest shifted further in the apical direction. After 12 weeks, the buccal crest was located>2 mm apical of the marginal border of the SLA surface.
Conclusion: The bone-to-implant contact that was established during the early phase of socket healing following implant installation was in part lost when the buccal bone wall underwent continued resorption.