This controlled preclinical study analyzed the effect of implant surface characteristics on osseointegration and crestal bone formation in a grafted dehiscence defect minipig model. A standardized 3 mm × 3 mm acute-type buccal dehiscence minipig model grafted with deproteinized bovine bone mineral and covered with a porcine collagen membrane after 2 and 8 weeks of healing was utilized. Crestal bone formation was analyzed histologically and histomorphometrically to compare three implant groups: (1) a novel, commercially available, gradient anodized (NGA) implant, to two custom-made geometric replicas of implant "1," (2) a superhydrophilic micro-rough large-grit sandblasted and acid-etched surface, and (3) a relatively hydrophobic micro-rough large-grit sandblasted and acid-etched surface. At 2 and 8 weeks, there was no difference between the amount and height of newly formed bone (NBH, new bone height; BATA, bone area to total area) for any of the groups (p > 0.05). First bone-to-implant contact (fBIC) and vertical bone creep (VBC) at 2 and 8 weeks were significantly increased for Groups 2 and 3 compared to Group 1 (p < 0.05). At 8 weeks, osseointegration in the dehiscence (dehiscence bone-implant-contact; dBIC) was significantly higher for Groups 2 and 3 compared to Group 1 (p < 0.05). The amount of newly formed bone (BATA) and NBH was not influenced by surface type. However, moderately rough surfaces demonstrated significantly superior levels of osseointegration (dBIC) and coronal bone apposition (fBIC) in the dehiscence defect compared to the NGA surface at 2 and 8 weeks. For this type of study, clinical trial registration is not required. This study was conducted at the Biomedical Department of Lund University (Lund, Sweden) and approved by the local Ethics Committee of the University (M-192-14).