on the evidence

New types of fillings are often introduced into practice with limited evidence about their long-term clinical performance, say Rachel Richardson and colleagues

Tooth decay (dental caries) is one of the most common diseases and accounts for almost half of all tooth extractions in England and Wales. A variety of interventions can be used to prevent the development of caries: for example, the use of fluoride. But if decay is not prevented, cavities develop and can cause considerable pain and, ultimately, tooth loss.

In order to prevent pain and tooth loss, it may be necessary to remove the diseased tissues and restore the cavity (with a filling).

Several restorative materials are available at different costs, requiring varying amounts of expertise to prepare and complete a filling. Restorations have a limited life-span, and once a tooth is restored the filling is likely to be replaced several times in the patient's lifetime - the restorative cycle. Studies in the UK suggest that much of restorative dentistry is replacement of existing restorations, accounting for around 60 per cent of all restorative work. The treatment of carious teeth by the placement of simple, direct restorations alone costs the NHS in England and Wales£173m a year.

Restorations do not last forever; once inserted, they may fail at variable rates due to objective factors, including the characteristics of the filling material. The decision to replace a restoration is also influenced by more subjective factors, including the dentist's interpretation of the restoration's condition and the health of the tooth, the criteria used to define failure, and patient demand. These decisions are subject to a great deal of variation. There is a lack of standardisation, and no generally agreed criteria are used to decide when a restoration requires replacement.

Intra-coronal fillings (placed inside a cavity prepared in the crown of a tooth) are made from one of a range of materials: dental amalgam, composite resins, glass ionomer cements, resin-modified glass ionomer cements, componers and cermets, cast gold and other alloys, and porcelain.

Studies carried out under optimal conditions show fillings lasting a long time, Effective Health Care reports. But the survival of fillings in these studies is unlikely to be matched in the conditions of routine dental practice. However, these are the only studies that make reliable comparisons of the longevity of different materials.

Amalgam is the intra-coronal restorative material of choice

Overall, amalgam was found to last the longest of any of the available restorative materials (see number of 'tooth years' in the table below).

But amalgam is silver in colour, and if aesthetics are important, patients may prefer restorations which are tooth-coloured. Composite restorations, one of the main alternatives, had a shorter life-span, whether or not a dentine bonding system was involved. The newer generation of dentine bonding agents, which use some form of acidic primer, have better retention rates than earlier generations.

Amalgam fillings are safe

There have been concerns over the safety of amalgam, most of which appear to be unjustified.

The British Dental Association has recently concluded: 'To date, extensive research has failed to establish any links between amalgam use and general ill-health. Those countries which are limiting the use of amalgam are doing so to lower environmental mercury levels.'

The Department of Health's committee on toxicity reviewed the evidence on the safety of amalgam in response to an expert report from the European Commission, and concluded that dental amalgam is free from risk of systemic toxicity and only a very few cases of hypersensitivity occur.

Several restorative materials were reported as having low survival rates. The use of cermet cements, the earlier composite and glass ionomer sandwich techniques in class II cavities, and gallium all had high failure rates and cannot be recommended.

The survival of inlays using ceramics, gold and composites was examined in the review. Overall, no significant difference was found between porcelain and composite inlays, although some types of porcelain did give better results.

A review of economic studies was also carried out for Effective Health Care and was supplemented by information provided by dentists on the time taken to carry out restorations. This allowed a comparison of the cost- effectiveness of filling materials to be made.

The results (see table) show that amalgam is much more cost-effective than composite and inlays across all time periods because it is cheaper and has better survival rates.


There is a lack of clear scientific criteria for dentists to use to decide when a restoration has failed and should be replaced. As a result, there is wide variation in treatment decisions by any one dentist, and between different dentists.

There is a difference between identifying how long a restoration could last if objective outcome measures were used and how long it is allowed to last when individual practitioners use their own criteria. It is claimed that the likelihood of having a restoration replaced is more than doubled when a patient changes practitioner.

Appropriate criteria for replacement of restorations are needed, and dental schools should train dentists in using standardised definitions of what constitutes a failed restoration and to adopt appropriate maintenance policies. This would protect the public against unnecessary procedures, reduce costs and improve the quality of professional decision making.

The dental manufacturing industry is constantly promoting the use of new materials. These are marketed and introduced into practice typically without reliable and comprehensive trials. This has created a high level of uncertainty about the merits of different materials. While this review has helped to give a clearer picture of what filling materials might be the best buy, introducing dental materials into clinical practice should be subject to any new NHS regulations designed to promote the quality of healthcare.

In addition, the good results achieved in the better studies demonstrate that routine clinical practice may be producing sub-optimal results. This raises the issue of how clinical practice can be improved so that restoration longevity in all settings approaches the best that can be achieved - and what the resource implications of this may be.

Co-ordinated research in primary dental care is needed to assess the effects of clinicians' skill, tooth type, cavity type and material type on restoration survival, taking into account evolving disease patterns.

Dental amalgam is the intra-coronal restorative material with the longest duration and, from the NHS's perspective, is of lower cost.

Unless there is a contra-indication (usually aesthetics or pregnancy) it is recommended for routine use wherever possible.