Patient demographics

 In total, 196 patients from 17 centers met the inclusion criteria for this study and were included in the per-protocol analysis. In total, 285 implants were placed (Table 1). At the 5-year follow-up, data were available for the 137 patients who completed the study (Fig. 1). Patient demographic data is presented in Tables 2 and 3.

Implant success

Implant success was reported according to the criteria for implant success laid down by Albrektsson et al. [30], as well as that by Buser et al. [29]. According to Albrektsson et al., there were three implant failures post-loading and three implants which did not meet the success criteria due to bone loss (n = 2) and peri-implantitis (n = 1). According to Buser et al., there were three implant failures post-loading and one implant which did not meet the success criteria due to peri-implantitis (n = 1) (Table 4). The three implants which were late failures were lost at 2 years post-loading due to important bone loss and at 3.6 years and at 4.6 years post-loading (all platform switching). Additional five implants were lost before loading as a result of no osseointegration (early failures) and therefore were not considered for the analysis.

The cumulative success rates did not differ according to both criteria at 1-year follow-up or at 3-year follow-up, being 100% and 99.6%, respectively. However, at 5-year follow-up, the success rate according to Buser et al. was higher at 98.0% than that according to Albrektsson et al. at 97.1%. The sub-group analysis revealed that the success rate for platform-matching implants was 100% at 1-year and at 3-year and 96.2% at 5-year follow-up according to Albrektsson et al. and 100% at each follow-up according to Buser et al. Conversely, for platform-switching implants the success rate was 100% at 1-year follow-up, 99.4% at 3-year follow-up, and 97.4% at 5-year follow-up, according to both criteria.