The aim of the present clinical study was to test whether or not the addition of recombinant human bone morphogenetic protein-2 (rhBMP-2) to a xenogenic bone substitute mineral (Bio-Oss) will improve guided bone regeneration therapy regarding bone volume, density and maturation. In 11 partially edentulous patients, 34 Brånemark implants were placed at two different sites in the same jaw (five maxillae, six mandibles) requiring lateral ridge augmentation. The bone defects were randomly assigned to test and control treatments: the test and the control defects were both augmented with the xenogenic bone substitute and a resorbable collagen membrane (Bio-Gide). At the test sites, the xenogenic bone substitute mineral was coated with rhBMP-2 in a lyophilization process. Following implant insertion (baseline), the peri-implant bone defect height was measured from the implant shoulder to the first implant-bone contact. After an average healing period of 6 months (SD 0.17, range 5.7-6.2), the residual defects were again measured and trephine burs were used to take 22 bone biopsies from the augmented regions. The healing period was uneventful except for one implant site that showed a wound dehiscence, which spontaneously closed after 4 weeks. Later at reentry, all implants were stable. At baseline, the mean defect height was 7.0 mm (SD 2.67, range 3-12 mm) at test and 5.8 mm (SD 1.81, range 3-8 mm) at control sites. At reentry, the mean defect height decreased to 0.2 mm (SD 0.35, range 0-1 mm) at test sites (corresponding to 96% vertical defect fill) and to 0.4 mm (SD 0.66, range 0-2 mm) at the control site (vertical defect fill of 91%). Reduction in defect height from baseline to reentry for both test and control sites was statistically significant (Wilcoxon P<0.01). Histomorphometric analysis showed an average area density of 37% (SD 11.2, range 23-51%) newly formed bone at test sites and 30% (SD 8.9, range 18-43%) at control sites. The fraction of mineralized bone identified as mature lamellar bone amounted to 76% (SD 14.4, range 47.8-94%) at test compared to 56% (SD 18.3, range 31.6-91.4%) at control sites (paired t-test P<0.05). At BMP-treated sites 57% (SD 16.2, range 29-81%) and at control sites 30% (SD 22.6, range 0-66%) of the surface of the bone substitute particles were in direct contact with newly formed bone (paired t-test P<0.05). It is concluded that the combination of the xenogenic bone substitute mineral with rhBMP-2 can enhance the maturation process of bone regeneration and can increase the graft to bone contact in humans. rhBMP-2 has the potential to predictably improve and accelerate guided bone regeneration therapy.