The overall study flow was presented in Fig.
1. A buccal dehiscence defect model
50 (link) was constructed in order to simulate a damaged extraction socket (the experimental site). After inducing the dehiscence defect, the mesial roots of the mandibular second, third, and fourth premolars (P2, P3, and P4, respectively) were extracted 4 weeks later. Each distal root was subjected to root canal treatment after hemisection and was kept as a reference pristine site for the corresponding mesial root. The following two extraction sockets on the unilateral alveolar ridge were rotated into the following groups to produce an equal distribution of three other premolar sites:
DBBM/membrane group: ARP with DBBM (
Bio-Oss®; Geistlich Pharma AG, Wolhusen, Switzerland) and a membrane (
Bio-Gide®, Geistlich Pharma AG, Wolhusen, Switzerland)
DBBM/matrix group: ARP with DBBM (
Bio-Oss®; Geistlich Pharma AG, Wolhusen, Switzerland) and a matrix (Collagen Graft 2
®, Genoss, Suwon, Korea)