Survival of direct resin restorations in posterior teeth within a 19-year period (1996–2015): A meta-analysis of prospective studies

•88 prospective, clinical trials (1996–2015) were analyzed in terms of survival of posterior composite restorations.•Significant findings were only observed for short-term studies and by including all studies (short- and long-term studies).•The observation period, the recall rate, the ratio of Class I/II fillings and the number of restorations/patients had a significant influence on the overall failure rate when including all studies (short- and long-term).•No influence of operator status, isolation method or bonding systems on the overall failure rate was found.•In this study a trend of excluding patients at-risk (e.g. high caries activity) and including patients with good oral hygiene was identified, which produces more favorable results.

ObjectivesThe aim of this study is to analyze the survival of posterior composite restorations published within the last 19 years (1996–2015).MethodsIn this study only prospective, clinical trials with specification of the failure rate according to Class I/II composite fillings were included. Studies were analyzed according to the observation period (all studies vs. short-term vs. long-term studies). Retrospective studies and/or open laminate studies, tunnel restorations and Class V restorations were excluded. The following variables possibly influencing the failure rate were extracted from the studies: observation period, recall rate, average age of patients, number of patients, ratio of Class I/II fillings, number of restorations, ratio of premolars/molars, operator, method of isolation, bonding generation and filler size.ResultsA total of 88 studies were included for statistical analysis. The observation period of the studies varied between 1 and 17 years, while most of the studies did not last longer than 5 years. Fracture of the restorations, secondary caries and marginal gap are the main causes for failure in the first 5 years (in descending order), while fracture and secondary caries are similarly distributed in long-term studies. Variables of investigation differed greatly in significance according to the respective observation period. The observation period, the recall rate, the ratio of Class I/II fillings and the number of restorations and patients had a significant influence on the overall failure rate when including all studies (short- and long-term). A linear correlation between the observation period and the failure rate was observed. In long-term studies these variables were not significant any longer. No significant difference in the failure rates between the materials per study was observed. The most common commercial composites investigated were: Tetric Ceram, Surefil, Filtek Supreme (incl. XT), Filtek Z250.The mean annual failure rate was 1.46% (±1.74%) for short-term studies and 1.97% (±1.53) for long-term studies.There is still a big need for clinical studies lasting longer than 5 years, as failure rates of composite restorations in posterior teeth increases with longer observation periods.SignificanceA decreasing failure rate with an increasing recall rate as observed in our study suggests a patient selection in regard to availability and dental awareness. Internationally standardized evaluation criteria are mandatory in order to allow comparisons of the outcomes of clinical studies.