The most often used graft material in our evaluation was Geistlich Bio-Oss (53.0%) followed by autogenous bone (32.5%). When compared to no graft, the use of both grafts resulted in significantly higher implant survival rates. In various studies, the bone substitute was found to promote bone regeneration and to allow for long-term stability of the augmented volume. A recent meta-analysis compared Geistlich Bio-Oss and autogenous bone: For maxillary sinus floor augmentation, a mean implant survival rate of 98.6 ± 2.6% was found for bone substitute, 88.6 ± 4.1% for autogenous bone + bone substitute and 97.4 ± 2.2% for autogenous bone alone. While there was a trend in favour of Geistlich Bio-Oss, the differences were not statistically significant. When the authors evaluated the studies with vertical and/or lateral alveolar ridge augmentation, they found similar mean implant survival rates for the three treatment modalities (97.4 ± 2.5% for Geistlich Bio-Oss, 100 ± 0.0% for autogenous bone + Geistlich Bio-Oss, 98.6 ± 2.9% for autogenous bone alone). Data is based on 4687 implants in 1816 patients. This finding is in line with the results from our study. In contrast to Geistlich Bio-Oss, the use of a synthetic material (Cerasorb) resulted in a significantly lower implant survival than no graft in the pairwise comparison, although in the log-rank test, the synthetic material demonstrated better results than no graft material used. However, the heterogeneity of the data does not allow drawing statistical conclusions on the superiority between those bone substitutes that were used rather rarely.

The retrospective analysis presented here included a large number of implants followed up to 20.2 years. While it allows conclusions on the efficacy of augmentation procedures in daily practice, there are some limitations. It was not possible to evaluate the initial defect size and morphology. Therefore, it is not clear whether the different graft materials and membranes were used in comparable clinical situations or whether differences in original defect size may have accounted for some of the differences in survival rates. In addition, the data were only evaluated for implant survival and not for implant success. This is due to the fact that in clinical studies, the success of dental implants is commonly defined by implant survival. Although different criteria for implant success were suggested in the 1980s and early 1990s, the success rate addressing prosthetic, biological and aesthetic complications was largely absent from the literature in the 1990s and was newly established only 5 to 10 years ago. Despite of the international proposed criteria, a common consensus could not be reached so far.