Restorative materials for the direct coronal restoration of permanent posterior dentition: an overview of systematic reviews

Authors: Lewis SR, Walsh T, Glenny AM, Banerjee A, Boyers D, Hatton PV, Mackie C, Martin N, Palin WM, Pritchard MW, Quinn BM, Ramsay CR, Ricketts D, Riley P, Clarkson JE

Here, we provide an Executive Summary of an overview conducted in 2025. The full report is undergoing peer review and will be published in the Cochrane Library in 2026.  

Objective

•            To summarise the evidence from 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.

Policy Relevance 

Following the global agreement under the Minamata Convention on Mercury, and its subsequent ratification by governments, there is now widespread commitment to phasing down, and ultimately eliminating, the use of dental amalgam. This shift underscores the urgent need to evaluate the clinical efficacy and broader impact of mercury-free restorative alternatives. This Cochrane overview is relevant to all countries, including those that have commenced their own phase-down/phase-out schedule.

While several published systematic reviews compare one type of restorative material with another, an evidence gap remains: there is currently no single, comprehensive synthesis that evaluates the clinical effectiveness of direct-placement materials for coronal restorations in posterior permanent teeth, alongside an appraisal of the certainty of the evidence. This overview addresses the existing evidence gap by identifying, appraising, and synthesising relevant data from existing reviews to inform key stakeholders - including patients, dental professionals, and policymakers.

Population/Setting

The overview includes evidence from systematic reviews of children and/or adults undergoing direct coronal restoration with any direct-placement restorative material in the permanent posterior dentition, regardless of the size of cavity requiring restoration. There was no restriction on the clinical setting within which the restorations were placed (e.g. general practice or hospital).

Methodology

Review type: Overview of systematic reviews

Search strategy: In April 2025, the Cochrane Library, MEDLINE, Embase, Epistemonikos, and PROSPERO were searched for systematic reviews comparing the clinical effectiveness of restorative materials for direct-placement coronal restorations in the permanent posterior teeth of children and adults. Eligible reviews comprised quantitative syntheses comparing at least two direct-placement restorative materials: resin-based composite (RBC), resin-modified glass ionomer cement (RMGIC), glass ionomer cement (GIC), compomer, dental amalgam, or other material. For RBC, bulk-fill was also compared with incremental-layered (conventional) RBC. Key outcomes of interest were tooth loss due to restoration failure, restoration failure/survival, time to failure, and adverse effects.

Reviews published since 2015 were included to reflect current practice, acknowledging that these reviews will include primary studies conducted prior to 2015. Backward citation searches were performed, PROSPERO was checked for ongoing reviews, and screening for retractions of eligible reviews and included studies was undertaken.

Risk of bias assessment: Cochrane’s recommended tool for assessment of systematic reviews, ROBIS, was used to assess the risk of bias in included reviews.

Data synthesis: Pooled level outcome data from each systematic review were extracted and reported, including effect estimates and 95% confidence intervals (CIs). Narrative summaries were presented using the Synthesis Without Meta-analysis (SWiM) guidelines recommended by Cochrane. The overview planned to explore the following prespecified groups: people with needs that require special dental consideration; size of cavity or defect to be restored; treatment for a new restoration or replacement restoration; experience of dental health professional; setting (general practice, hospital, or rural/field setting); country (high income vs low to middle income). Insufficient evidence was found to allow for this exploration of data.

Findings were organized by restorative material type and review publication date. To improve clarity, pooled effect estimates from the original reviews were re-expressed as risk differences (RDs) with consistent directionality across reviews. These adjustments followed standard meta-analysis principles to enhance interpretability for end-users of the overview.

Data from the longest follow-up was prioritised in each review. The degree of overlap of primary studies in similar reviews was calculated for each clinical outcome of interest. Where overlap was high or very high, one review was selected, giving preference to Cochrane reviews. If no Cochrane reviews were available, the most relevant review based on methodological quality, recency, and comprehensiveness was chosen. Sensitivity analyses were undertaken to address potential data loss from overlap management.

Assessment of the certainty of evidence: The certainty of evidence was assessed for key outcomes: tooth loss due to restoration failure, restoration failure/survival, time to failure, and adverse effects. If primary studies overlapped significantly, certainty was assessed only using data from a prioritised review. Where available, existing GRADE assessments from reviews were reported; otherwise, the overview authors conducted their own GRADE assessments. Certainty of the evidence was rated as high, moderate, low, or very low.

Brief economic commentary: The review included a brief economic commentary, the purpose of which was to place an economic lens on the findings.

Results

Number of systematic reviews included: Fourteen reviews (evaluating 57 primary studies) with outcome data relevant to this overview were included. Very few primary studies were conducted in general practice (about 10%, where reported in primary studies in prioritised evidence). No studies included people with needs requiring special consideration. Reviews compared six combinations of restorative materials:

•            RBC and dental amalgam (three reviews)

•            RBC and GIC (four reviews) 

•            Bulk-fill RBC and incremental-layered (conventional) RBC (eight reviews)

•            RMGIC and GIC (two reviews)

•            GIC and dental amalgam (one review)

•            GIC with compomer (one review)

Risk of bias in included reviews: Only two reviews were judged to have an overall low risk of bias using ROBIS. The most frequent concern was related to data analysis methods, particularly the analysis of split-mouth studies. Unit of analysis issues in meta-analyses were also frequently observed, with overcounting of restorations in multi-arm studies or studies with multiple publications over different time periods. 

Summary of the evidence: 

RBC compared with dental amalgam. One Cochrane review reported low-certainty evidence that the difference in risk of restoration failure may be 7 percentage points less with dental amalgam than RBC (RD 0.07, 95% CI 0.05 to 0.09; 2 primary studies, 3010 restorations; 5 to 7 years follow-up). Studies in this review began recruitment in the late 1990s which may affect the applicability of this evidence to contemporary practice. The failure rate for RBC in these older studies was higher than that for more contemporary evidence in other reviews (almost 15% failure rate versus approximately 5% failure rate). Although there was similar evidence of restoration failure from two other reviews, the certainty of this evidence was rated as very low. Only one review reported similar postoperative pain and discomfort (about 5%) for both materials which was judged to be very low-certainty evidence. Evidence included both Class I and Class II restorations.

RBC compared with GIC. 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; 1 primary study, 60 restorations; 10 years follow-up), or the risk of postoperative sensitivity (RD 0.03, 95% CI -0.03 to 0.10; 2 primary studies, 118 restorations); low-certainty evidence from one review. 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. Most reviews for this comparison (n = 8), reported similar risk differences, and the evidence was judged to be of moderate certainty. There is probably no difference in risk of restoration failure between bulk-fill and incremental-layered RBC (RD 0.00, 95% CI -0.03 to 0.03; 7 primary studies, 511 restorations; 1 to 10 years follow-up), with failure rates for both materials at less than 5%. In one review, there was almost no postoperative sensitivity reported for either type of RBC (RD 0.00, 95% CI -0.01 to 0.02; 5 primary studies; 510 restorations; 2 to 3 years follow-up). Evidence included restorations in both Class I and Class II cavities, with overall more Class II restorations (multiple-surface load bearing).

RMGIC compared with GIC. At 2-year follow-up, RMGIC may reduce the risk of restoration failure compared to GIC in both Class I restorations (RD -0.19, 95% CI -0.37 to -0.02; 1 primary study, 50 restorations), and Class II restorations (RD -0.71, 95% CI -0.93 to -0.48; 1 primary study, 38 restorations); low-certainty evidence. In this single study, most restorations were Class I (occlusal and non-load bearing).

GIC compared with dental amalgam, and GIC compared with compomer. No eligible systematic reviews reported on restoration failure or other critical outcomes of interest to this overview review for these materials.

Limitations: To ensure relevance, the overview was limited to reviews that closely aligned with predefined criteria, resulting in the exclusion of 26 reviews. For transparency, excluded reviews, including brief summaries of their conclusions, have been documented in the Cochrane overview. 

It was not always feasible to apply the prespecified prioritisation hierarchy. Instead, study overlap at an outcome level was assessed, selecting the review that was judged to provide the most relevant data to this overview rather than basing decisions on factors such as the date of the search. Again, the findings across most reviews were broadly comparable, reinforcing the robustness of the selected evidence base.

Brief economic commentary findings: Six economic reports identified no strong conclusions regarding the most cost-effective mercury-free direct-placement materials; future long-term modelling studies are required to inform decision-making. 

One Health Technology Assessment (HTA) report conducted in Canada comparing dental amalgam with a mercury-free alternative (RBC) found that amalgam restorations lasted longer and were less costly overall to healthcare payers; however, time to failure models in this HTA report were based on a study with recruitment starting in the late 1990s. 

Policy and practice implications

This Cochrane overview addresses the clinical effectiveness of direct-placement restorative materials used to restore the tooth's shape and function in the event of permanent damage to the tooth structure mostly caused by caries. However, in most cases caries is a preventable disease, and promotion of optimal oral hygiene behaviours/practices should be seen as the most impactful route to avoiding caries and subsequent dental restorations.

Evidence indicates that dental amalgam may lead to fewer restoration failures compared to RBC. However, this evidence is based on older studies and, given the change in composite material properties over time as well as experience of practitioners in its use, may not be comparable to contemporary practice.

There is probably no difference in restoration failure between bulk-fill or a conventional incremental-layered RBC, with overall failure rates for both RBC materials at less than 5% and low risks of postoperative sensitivity. 

There is limited evidence for other comparisons. 

The results of clinical effectiveness in this overview should be considered alongside the health and environmental considerations of these materials. Factors such as cost, acceptability, and clinical presentation should also be considered. Most studies in this evidence-base were conducted in university hospital settings. Some restorative materials may be more technique sensitive than others, and/or require more time for restoration placement (e.g. RBC). The use of  mercury-free restorative materials in general practice should consider these additional resource implications, as well as education and training for high-quality restoration and maintenance. Above all, oral healthcare delivery should focus on strategies and funding for caries prevention. 

Equity considerations: Research is needed to address the different situations in which different types of restorative materials may be more effective, for example in people with special dental considerations, different cavity depths, or different settings in which the restoration is conducted.

This Executive Summary has been written by the authors and the authors are solely responsible for its content. This Executive Summary and the overview which it describes has not been through Cochrane's editorial or peer review process. and is not endorsed or approved by Cochrane in any way.