Background
Direct‐placement dental restorative materials are required to replace the loss of tooth substance and restore the functional structural integrity of damaged posterior teeth. Whilst dental amalgam was traditionally used for these restorations, there are concerns about toxicity to human health and the environment. The Minamata Convention on Mercury recommends a phase‐down of amalgam use in dentistry. Alternative mercury‐free direct‐placement restorative materials are available and have been evaluated in systematic reviews.
Objectives
To summarise the evidence from Cochrane and other systematic reviews evaluating the clinical effectiveness and longevity of restorative materials for direct‐placement coronal restoration in the permanent posterior dentition.
To summarise the evidence from economic studies for the cost‐effectiveness of restorative materials for direct‐placement coronal restoration in the permanent posterior dentition.
Methods
In April 2025, we searched the Cochrane Library, MEDLINE, Embase, Epistemonikos and PROSPERO for systematic reviews that compared restorative materials for direct‐placement coronal restoration in the permanent posterior dentition in children and adults. We included reviews reporting quantitative syntheses and comparing at least two restorative materials, from: resin‐based composite (RBC), resin‐modified glass ionomer cement (RMGIC), glass ionomer cement (GIC), compomer, dental amalgam, or other material. For RBC, we also compared bulk‐fill with incremental‐layered (conventional) RBC.
We used Cochrane methodology to conduct an overview of the evidence from eligible reviews, and assessed the methodological quality of reviews using ROBIS. We prioritised data from selected reviews when we found a high degree of overlap of primary studies between reviews.
Critical outcomes were: tooth loss (owing to restoration failure), restoration failure, time to failure, and adverse effects.
Additionally, we searched for relevant economic evaluations of direct‐placement restorative materials and developed a brief economic commentary.
Main results
Overall, we found 14 reviews including 57 primary studies; only one was a Cochrane review. Very few primary studies (about 10%) were conducted in general practice. We prioritised data at the longest time point from six reviews, including 23 primary studies. Two reviews were at low risk of bias, and the others were at high risk. However, results across all reviews were largely comparable.
RBC compared with dental amalgam. One Cochrane review reported low‐certainty evidence that the risk of restoration failure may be 7% less with dental amalgam than RBC (RD 0.07, 95% CI 0.05 to 0.09; 2 studies, 3010 restorations; 5 to 7 years follow‐up; class I and II restorations). Studies in this review began recruitment in the late 1990s, which may affect the generalisability of this evidence to contemporary practice, and the failure rate for RBC in these studies was higher than in contemporary evidence of RBC in other reviews (almost 15% compared with approximately 5%). Although there was similar evidence of restoration failure from two other reviews, the certainty of this evidence was very low and, therefore, we had little confidence in the risk reductions reported in these reviews. Only one review reported postoperative pain and discomfort (about 5% in both groups), which reviewers judged to be very low‐certainty evidence.
RBC compared with GIC. In one review, there may be little or no difference in the risk of restoration failure between RBC or GIC (RD ‐0.07, 95% CI ‐0.17 to 0.04, favours RBC; 1 study, 60 restorations; 10 years follow‐up), or the risk of postoperative sensitivity (RD 0.03, 95% CI ‐0.03 to 0.10, favours GIC; 2 studies, 118 restorations); low‐certainty evidence in a small sample size. Evidence for postoperative sensitivity in another review was very low certainty (overall events ranging from zero to 10%) and therefore we are uncertain of any benefit for either restorative material. Whilst this evidence included class I and II restorations, most reported were class I restorations (occlusal non‐load bearing).
Bulk‐fill compared with incremental‐layered RBC. We found the most reviews for this comparison (n = 8), all reporting similar risk differences, and therefore this evidence was judged to be of moderate certainty. The risk of restoration failure is likely to be low, and no different between groups, at less than 5% (RD 0.00, 95% CI ‐0.03 to 0.03; 7 studies, 511 restorations; 1 to 10 years follow‐up). In one review, there was almost no postoperative sensitivity for either type of RBC (RD 0.00, 95% CI ‐0.01 to 0.02; 5 studies; 510 restorations; 2 to 3 years follow‐up). Overall, more restorations were in class II restorations (multiple‐surface load bearing).
RMGIC compared with GIC. In one review, RMGIC may be more likely to reduce the risk of restoration failure than GIC in class I restorations (RD ‐0.19, 95% CI ‐0.37 to ‐0.02; 1 study, 50 restorations), and class II restorations (RD ‐0.71, 95% CI ‐0.93 to ‐0.48; 1 study, 38 restorations), both at two years follow‐up from low‐certainty evidence in a very small sample.
GIC compared with dental amalgam, and GIC compared with compomer. No reviews reported critical outcome data for these comparisons.
Brief economic commentary: Six economic reports identified no strong conclusions regarding the cost‐effectiveness of mercury‐free restorative materials. One report, using data from older studies, found that amalgam lasted longer and was less costly than RBC; however, we identified no economic evaluations based on contemporary clinical evidence for amalgam.
Authors' conclusions
Most evidence compared bulk‐fill with incremental‐layered RBC; there is probably no difference between these materials in restoration failure. Although we found evidence of fewer restoration failures with dental amalgam than RBC, this may not be comparable to contemporary practice owing to changes in properties of RBC materials and practitioner experience.
We identified few economic evaluations to provide strong conclusions to support the clinical effectiveness findings for mercury‐free restorative materials.
Regarding implications for policy and practice, little evidence in this overview is from general practice. The results of clinical effectiveness should be considered alongside cost, acceptability, clinical presentation, time required for restoration placement (which may be technique‐sensitive), and the health and environmental considerations of the materials. These conclusions emphasise that caries prevention is critical to effective and sustainable oral health.
Funding
Cochrane Oral Health (COH) is supported by a collaborative research agreement between The University of Manchester and the University of Pennsylvania.
Registration
Protocol (2025): https://www.crd.york.ac.uk/PROSPERO/view/CRD420251004182
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)