Can repair increase the survival of failed restorations when compared with replacement? Systematic-Review and Meta-Analysis

Abstract Background: The aim of this systematic review was to collect and compare survival data of repair and replacement techniques for restoration failure in both dentitions. Methods: To conduct a systematic review to gather and compare survival data between repair and replacement techniques in failed restorations of primary and permanent teeth. The systematic searches were performed in PubMed/MEDLINE, Scopus, Web of Science, Embase, ProQuest and OpenSigle databases. Two independent calibrated reviewers (kappa=0.87) assessed as inclusion criteria: (1) Repair or replacement studies, (2) success, longevity, or survival data, (3) randomized clinical controlled trials; and for the exclusion criteria (1) Loss to follow-up over 30%, (2) less than 12 months follow-up, (3) anterior teeth. The risk of bias was assessed by the RoB 2 tool, and the certainty of the evidence was assessed using the GRADE tool. Meta-analysis was conducted considering the treatment’s survival rate. Results: We identified 4,070 potentially relevant publications, after removing duplicates and assessing title and abstracts, 2,115 studies were ineligible because they were not about repair and replacement, did not have percentage survival data, or were not RCTs. Seven remaining articles were assessed for the exclusion criteria, and we found all papers for full-text analysis. Three studies were included in the quantitative analysis, all evaluating permanent dentition. A standard follow-up time between studies was collected, and the pooled survival rate between studies was 99% after three years. There was no statistically significant difference between approaches and no heterogeneity between studies. All included studies had a high risk of bias. The certainty of evidence for the outcome measure of success between repair and replacement interventions in failed restorations is very low.Conclusion: The longevity for both techniques appers to be similar. However, the certainty of evidence is very low, and the trials have a high rick of bias.