Which is the best antibiotic prophylaxis protocol to prevent early implant failures?

Selection criteria The inclusion criteria of this systematic review were patients undergoing dental implant placement. Only randomised clinical trials (RCTs) that compared placebo, no antibiotic and/or any type of antibiotics, administered pre-operatively, intra-operatively, post-operatively or combinations of these, at any dose and for any duration were considered eligible. Included RCTs were required to have a follow up period of at least three months with at least 20 patients per treatment arm. No restrictions on date of publication or language were applied.Key study factor Four electronic databases (MEDLINE, SCOPUS, CENTRAL and Web of Knowledge) in addition to six related journals (Journal of Clinical Periodontology, Clinical Oral Implants Research, Clinical Implant Dentistry and Related Research, Journal of Periodontology, European Journal of Oral Implantology, International Journal of Oral & Maxillofacial Implants) were searched in duplicate for RCTs up to July 2017. Additional relevant literature was identified through hand-searching of reference lists, and through grey literature databases. Two independent reviewers screened the titles and abstracts . Data extraction and risk of bias assessment was performed simultaneously by two reviewers independently and in duplicate using the Cochrane tool for risk of bias assessment. A Network Meta-analysis (NMA) was conducted by integrating direct and indirect comparisons and the probability that each protocol was optimal was estimated. Subgroup and sensitivity analyses were planned to test the effect of risk of bias and of different variables on the results, but were not conducted due to the limited number of included studies.Main outcome measure Outcomes analysed were adverse events and early implant failures, defined as removal of mobile or stable implants with progressive marginal bone loss or infection in the first year after placement.Main results A total of 2248 RCTs were identified after removing duplicates, nine of which were finally included. Different protocols of antibiotic prophylaxis were compared with a total number of 1,693 participants. Seven of the included trials compared the use of one or more protocols of antibiotic prophylaxis with no prophylaxis or prophylaxis with a placebo, and two trials compared the use of different protocols, without the use of a no prophylaxis/placebo group. Amoxicillin was the only type used in all studies. Doses and timing varied among studies, although most of them used a single dose taken just before the implant placement. For the investigated outcomes, two trials were considered at low risk of bias and seven at high risk of bias.All protocols were more effective in reducing implant failures compared to placebo/no antibiotic (mean OR 0.08 to 0.45). Meta-analysis of direct comparison was only possible for the four trials comparing 2 g amoxicillin one hour preoperatively (B) to no antibiotic or placebo (A), indicating B as more effective (pulled OR = 0.40; 95% CI: 0.19-0.88; heterogeneity chi-squared 1.40, P = 0.706). These results were consistent with NMA effect estimates (mean OR = 0.45; 95% CI: 0.0210.93). A single dose of 3 g of amoxicillin administered one hour pre-operatively (C) was statistically more effective in reducing implant failures if compared to no prophylaxis/placebo (OR = 0.41, 95% CI = 0.180.91) and was considered as the most effective protocol. The single dose of 2 g of amoxicillin administered one hour pre-operatively was less effective than protocol C. Adverse events could not be studied in a meta-analysis due to an insufficient number of trials reporting it.Conlusions Implant patients are likely to benefit from antibiotis being administered one hour preoperatively in a dose of 3 g orally. The use of post-operative antibiotics does not seem, however, to be justified.