Mercury released from amalgam restorations does not give rise to toxic effects on the nervous system of children

Clinical scenario

The director of a low-income paediatric clinic has decided to invest in dental materials for its dental service. He is aware of the lower costs of amalgam restorations, but has concerns about the possible toxic effects on children's nervous systems. He decides to do a bibliographic search to help him with his decision.

Clinical question

The PICO (see below) question developed was: in children of between 6 and 10 years of age (Population) do amalgam restorations (Intervention) compared with composite restorations (Comparison) increase the risk of neuropsychological disorders (Outcome)?

Search strategy

Four databases, namely PubMed, Lilacs (Latin American & Caribbean Health Sciences Literature), Cochrane Library and Trip Database (www.tripdatabase.com) were searched using the following terms:

“neurotoxicity syndromes” [mesh] and “dental amalgam” [mesh] and (“mercury” [mesh] or “mercury poisoning, nervous system” [mesh]). limited to English or Spanish, child: 6–12 years and randomised controlled trials (see Table 1).

Table 1 Search strategy

No articles were found in Lilacs or the Cochrane Library, one systematic review was found via the Trip database, and six articles were found in PubMed. Of the latter six, three papers were considered relevant. These are summarised in Table 2.

Table 2 Summary of relevant papers

Discussion

Silver–mercury restorations have been used since 1830 but the controversy about the material's safety and its adverse effects on general health still continues. It is a fact that vapour of mercury is released from dental amalgam restorations and it is absorbed by the patient. The issue is, does this situation cause neuropsychological damage, particularly in children, who could be more vulnerable to the toxic effects of mercury than adults?

There are some confounding variables, such as eating and toothbrushing habits, fish consumption, gum chewing habits, etc., which influence daily mercury release and absorption from amalgam restorations. In order to reduce bias because of the variability between subjects, the studies selected for consideration here are randomised controlled trials.

The results of both studies^{1, 2} provide consistent evidence of no adverse effects on the nervous systems of the children who received amalgam restoration. Both clinical trials were well-designed, randomisation was correct and the tests were of sufficient power to detect clinically important neurocognitive effects.

Because of the nature of the intervention, blinding was not possible for dentist and patients but psychometrists were instructed to be blinded. The followup period for these studies could be insufficient for two reasons: amalgam restorations generally remain in place longer than the period followed in these trials (Kaplan-Meier median survival times were 12.8 years for amalgam restorations) and the delayed effects later in life, if they exist, are unknown.

Statistical considerations

The power of the test in New England Children's Amalgam Trial² (80%) could be considered not strong enough to detect small effects. In fact, the researchers calculated the sample size to detect a 3-point difference between treatment groups based on the existing evidence^{3, 4} which shows that, in children, a 10–15-μg/dl increase in blood lead level is associated with a 3-point decline in intelligence quotient. Because analysis of exposure data using the intention-to-treat principle did not consider the varying amounts of amalgam restorations in the children of the treatment group, the authors analysed the data (in another paper),⁵ stratifying subjects by surface–years of amalgam or by urinary mercury excretion, adjusting for test score, randomisation stratum, age, sex, family socio-economic status, hair mercury concentration and blood lead level. In the Casa Pía trial¹, De Rouen and colleagues considered a power of 97% for the tests. The sample size for the study was selected to ensure adequate power for detecting a small but near-uniform effect of 0.3 standard deviations for the three neurobehavioral outcomes considered, while maintaining control of the overall type-I error. For the data analysis in this trial, they developed a new statistics procedure, which allows comparisons with longitudinal data on multiple outcome variables and facilitates the detection of differences between treatments at the earliest possible time.⁶ The results in the three papers found no significant statistical association between amalgam restorations and nervous systems disorders in children.

Recommendations

As the release of mercury from an amalgam restoration is at its peak just subsequent to placement in the cavity, declining to a much lower, steady-state level by 10–15 days,^{7, 8, 9} the replacement of successful restorations is not advised. Composite resins, a commonly used option, also suffer from a lack of research into adverse effects and so cannot be considered free of risk. Furthermore, amalgam is more a cost-effective restoration material: composite resins have been shown to be 1.7–3.5-fold more expensive than amalgam to generate one tooth year.¹⁰ These considerations are particularly important in making decisions in Public Health.

Recently, a preliminary report by the European Scientific Committees on Consumer Products, on Health and Environmental Risks, and on Emerging and Newly Identified Health Risks http://ec.europa.eu/health/ph_risk/committees/04_scenihr/docs/scenihr_o_011.pdf, concluded that that there are no increased risks of adverse systemic effects exist and that dental amalgam is a safe material to use in restorative dentistry. Future research should be directed towards analysing adverse effects of amalgam in the less than 1% of the population genetically vulnerable to mercury toxicity or allergic to mercury.⁶

Clinical bottom line

There is no significant statistical association between the changes in the neurobehavioral and neuropsychological scores considered and the exposure to amalgam restorations in children. Under conditions similar to these trials, there is no reason other than aesthetics to discard amalgam as a choice to restore posterior teeth in children.