In 2006 JAMA (Journal of the American Medical Association) published a study claiming that mercury is safe to put in the mouths of children in the form of dental amalgam http://jama.ama-assn.org/cgi/content/full/295/15/1775. The article was titled, "Neuropsychological and Renal Effects of Dental Amalgam in Children: A Randomized Clinical Trial," and it was co-authored by David C. Bellinger, PhD, MSc; Felicia Trachtenberg, PhD; Lars Barregard, MD, PhD; Mary Tavares, DMD, MPH; Elsa Cernichiari, MS; David Daniel, PhD; Sonja McKinlay, PhD, in JAMA, 2006;295:1775-1783.

Here's why that claim is not valid.

The study has some interesting peculiarities about it. The authors knew going in that only a small segment of the population is susceptible to the kind of effects they were supposedly searching for in the general population. It's like they knew that gold had been found in Sacramento but they decided to start their search in New York, Chicago, Los Angeles, and elsewhere.

There is research evidence that the subpopulation that is susceptible to mercury poisoning can, according to the research, be identified in more than one way: For one, there is the "BDNF polymorphism". More specifically the signature is a V66M polymorphism in the gene encoding brain-derived neurotrophic factor; see http://toxsci.oxfordjournals.org/cgi/content/abstract/81/2/354). Also, over the long haul this segment of the population tends to show evidences of neurodevelopmental and other neurodegenerative disorders. What is more, both of these factors have been linked to mercury toxicity in multiple ways.

The authors show that they knew about special sensitivities when writing their report for JAMA if not when they were designing their study originally. Oddly, however, they chose to study the general population and to speak in terms of averages (like going to New York to look for gold in Sacramento) as if the information about toxicity and its special impact on susceptible individuals did not exist. They studied two groups of 267 children each. One group got amalgam fillings and the other got only composite fillings. The critical measure, according to them, was change in total IQ scores over time. Although this is a powerful measure of mental functioning, it is a crude way to assess the impact of mercury poisoning.

Another peculiarity is that they only measured urinary mercury, which is indicative of children dumping the poison. “Children assigned to the amalgam group had a significantly higher mean (SD) urinary mercury level 5 years after baseline than did children assigned to the composite group..." This finding is interesting because it contradicts their claim that mercury does not leach out of the amalgam much. Their own results, however, show that a lot of mercury is getting into the blood and vital organs of all the children.

The main flaw in the construction of their study is that they did not take seriously the vast amount of research showing that only about 2% of the general population is extremely susceptible to mercury toxicity over the short-term. This is roughly the proportion susceptible to autism and other neurodegenerative disorders. As a result, less than 2% of their sample, presumably about 5 kids in each group, might be expected to show an impact in the design they constructed. In this respect their study was done as if in a vacuum, as if no other work on the identification of the relevant genetically distinct subpopulation existed.

The design is also flawed in the measures applied. Should anyone reasonably expect Alzheimer’s, MS, or any long-term genetic related neurodegenerative disorder to show up in a 3 to 5 year period as measured by an IQ test? Would any of these disorders be detectible in a sample of 581 subjects over such a short span with an IQ test? The premise for the study is implausible. To detect the kind of change they were looking for, a vastly more powerful design with a greater sample size (about 15 to 25 times greater) might detect some differences. A more intelligent approach would be to screen for susceptibility to mercury toxicity to start with and then run the same sort of tests on a sample of children known to be susceptible to mercury poisoning.

At any rate, we still learn from their study that mercury put in our mouths by our dentists is poisoning our general population. The study only shows that the measures applied did not detect harm beyond showing vastly increased amounts of urinary mercury in children treated with dental amalgam tooth restorations. This shows that the mercury was pervasively affecting every organ reached by the blood. They also found a highly significant difference between boys and girls in albumin in the urine: “.. mean (median) albumin was higher for girls than for boys (36.9 [7.3] vs 18.3 [3.4]; P = .02).”

Why is albumin measured? Because it is related to liver health. From another source at http://www.transweb.org/qa/qa_txp/faq_albumin.html : “Liver disease, kidney disease, and malnutrition are the major causes of low albumin. A diseased liver produces insufficient albumin.” The higher level in girls than boys shows that they are handling the challenge of the mercury toxin more aggressively. They are also less likely to get autism.

The authors of the JAMA study concluded: “...our findings indicate that for US children exposure to elemental mercury secondary to the restoration of dental caries with mercury amalgam is unlikely to cause a reduction in IQ of at least 3 points." This is exactly like saying, we found no evidence of a 3 point difference in IQ over a 3 to 5 year period caused by mercury poisoning from dental amalgam in our small sample, so mercury is not harmful to the general public at large.

IQ test scores are not very sensitive to mercury poisoning and the population sampled is not the genetically susceptible subgroup of the population. Trying to measure mercury poisoning in the manner applied is like trying to predict autism, Alzheimer’s, and neurodegenerative disorders in the general population by measuring changes in IQ over time. This is an inappropriate way to detect mercury poisoning. The design involves a crude measure applied to the wrong sort of data sample.

What is more the whole study is irrelevant as a refutation of better designs that have shown positive results on the essential question: Is dental amalgam mercury harmless? Answer: No. Does getting rid of dental mercury improve health? Answer: Yes. End of story.

The authors say: “Our study was designed to answer a specific question on the safety of amalgam restorations as the standard of care for US children” but it appears that it would be more accurate to say that it was designed in such a way as to minimize any contrasts between the impact of dental amalgam and composite restorations.

The authors give away an important falsehood that is buried in their report: They say “the factors that might produce enhanced sensitivities are unknown” but this is not true. Their own list of references includes one of the studies showing that BDNF polymorphism (see their reference number 5) enables identification of a subpopulation especially susceptible to mercury poisoning. We know that persons susceptible to autism and neurodegenerative conditions are also more susceptible to mercury toxicity. As a result, it is already established that for those persons dental amalgam is a really bad idea. What is more for anyone in the general population, long-term effects of mercury toxicity are bad and well established. Dental amalgam, therefore, is a bad choice for anyone.

Mercury is not safe to put in your mouth. This is evidently surprising to dentists who put lots of mercury in people’s mouths. The authors say that 70,000,000 fillings are placed each year with 50% mercury. This is reprehensible when there is a good alternative composite material to use for tooth restoration that does not contain mercury. At 750 micrograms per mercury filling on the average, with about 50% mercury content on the average in each filling, dentists are placing approximately 12 metric tons of mercury in the mouths of their patients every year (I did the math). They warn us to beware of the contamination coming from fish and water supplies, but the mercury they are flushing from dental labs and their own offices is the primary source of the contamination of our water and fish.

A far better study than the JAMA paper is the one titled “Chronic low-level mercury exposure, BDNF polymorphism, and associations with cognitive and motor function” by Diana Echeverriaa, James S. Woodsa, Nicholas J. Heyera, Dianne S. Rohlmanc, Federico M. Farinb, Alvah C. Bittner, Jr., Tingting Lib, and Claire Garabediana. This paper appears in one of our toxicology journals and you can read the abstract at http://toxsci.oxfordjournals.org/cgi/content/abstract/81/2/354. The upshot of it is that dental mercury is especially devastating for folks who are susceptible to mercury poisoning. We know how to identify those folks and if we did that we would see that dental mercury is a huge factor in all kinds of neurodevelopmental disorders. There is no doubt about this.

Why then does JAMA say otherwise? Why does the ADA say otherwise?

Well, like my 2.9 year old grandson says, "I'll figure it out." It has something to do with the fact that 12 metric tons per year of mercury are being put in the mouths of the general public. Do you suppose this might have something to do with the fact that the ADA trade magazines and JAMA defend dental amalgam as safe? The fact is that the dentists themselves and the American medical community are the main creators of the mercury problem whose existence as a community they have chosen to deny (although many have gotten off that bandwagon). If 70% of the mercury burden in the general population is coming from dental amalgam (see Lorcheider et al. cited in my recent Milestones book with S. Oller and L. Badon through the link below) do you really think that tuna fish sandwiches are as harmful as the silver/mercury fillings dentist(s) put in our mouths?