Abstract
The present review aimed to evaluate the difference of dental implant failure rates and marginal bone loss (MBL) between implants inserted in fresh extraction sockets or healed sites. Electronic search was undertaken in three databases, plus manual search of journals, including studies randomized or not. Meta-analyses were performed besides meta-regressions, in order to verify how the odds ratio (OR) and MBL were associated with follow-up time. The review included 163 publications. Altogether, there were 17,278 and 38,738 implants placed in fresh extraction sockets and healed sites, respectively. Pairwise meta-analyses showed that implants in sockets had a higher failure risk in comparison to healed sites: OR 1.349, all studies included; OR 2.070, only prospective non-RCTs; OR 2.487, only RCTs (all p < 0.001). The difference in implant failure between the groups was statistically significant in the maxilla (OR 1.616, p = 0.029), but not in the mandible (OR 2.192, p = 0.075). The MBL mean difference (MD) between the groups was -0.053 mm (p = 0.089). There was an estimated decrease of 0.003 in OR (p = 0.284) and an increase of 0.006 mm (p = 0.036) in the MBL MD between groups for every additional month of follow-up. In conclusion, implants placed in fresh extraction sockets present higher risk of failure than implants placed in healed sites.