Purpose: The purpose of this meta-analysis was to evaluate the short- and long-term outcomes for marginal bone level changes around implants that benefit from one of three commercially available implant surface preparations: Astra Tech OsseoSpeed (ATO), Straumann SLA/SLActive (SLA), and Nobel Biocare TiUnite (NBT).
Materials and methods: A MEDLINE (full text) search was conducted in July 2019 using the search terms "dental implant," "prospective," "bone level," and either implant surface/brand "TiUnite," "Nobel," "Nobel Biocare" (Nobel Biocare); or "OsseoSpeed," "Astra Tech" (Astra Tech); "Dentsply" or "SLA"; "SLActive" or "Straumann," in different combinations. Additionally, searches in Google Scholar, Elsevier, and Wiley publisher homepages were made. The abstracts were screened for eligibility, according to predefined inclusion and exclusion criteria. Initial screening resulted in 588, 813, and 616 publications for ATO, SLA, and NBT, respectively. The principal outcome measure was change in marginal bone levels at the 1- and 5-year follow-ups. A weighted mean value for the marginal bone level changes (MBLCw) based on the number of implants analyzed in each of the studies and a standard error mean for the MBLCw were calculated (SEMw). A test of the null hypotheses that the MBLC was equal in each pair of the three groups was carried out for all data available at the implant level, using both the Student t test and Wilcoxon rank sum test. A P value below 5% was considered statistically significant.
Results: After a full-text analysis and screening, a total of 37 ATO, 23 SLA, and 53 NBT publications qualified for inclusion in the meta-analysis for both the 1- and 5-year data. At the 1-year follow-up, data were available for 2,586 implants for ATO, compared with 1,490 and 3,948 implants for SLA and NBT, respectively. A weighted analysis revealed that mean MBLCw was -0.29 mm (SEMw: 0.0005 mm, SD: ± 0.42 mm) for ATO, compared with -0.83 mm (SEMw: 0.0025 mm, SD: ± 0.36 mm) for SLA and -0.87 mm (SEMw: 0.0006 mm, SD: ± 0.56 mm) for NBT. There was a statistically significant difference between ATO vs SLA (P < .0001), ATO vs NBT (P = .0012), and SLA vs NBT (P = .0488). Data at the 5-year follow-up were available for 1,168, 202, and 1,683 implants from the ATO, SLA, and NBT groups, respectively. Analysis revealed that mean MBLCw was -0.35 mm (SEMw: 0.0038 mm, SD: ± 0.66 mm) for ATO compared with -0.74 mm (SEMw: 0.0154 mm, SD: ± 0.45 mm) for SLA and -1.19 mm (SEMw: 0.0107 mm, SD: ± 0.61 mm) for NBT. There was a statistically significant difference between ATO vs SLA (P = .0024) and ATO vs NBT (P = .0240). The difference between SLA and NBT did not reach statistical significance (P = .0769).
Conclusion: Based on the current meta-analysis, the null hypotheses could be rejected, and even if all groups demonstrated only a small mean amount of change in marginal bone levels at the 1- and 5-year follow-ups, there was a statistically significant difference between the three implant surface preparations, with the ATO surface showing superior marginal bone maintenance.