Manuscript

Authors: Khier, Al Bagshy, Al Mugait

Title: Dental Amalgam- Myth Vs Reality,

Webmedcentral DENTISTRY 2010;1(10): WMC00904

The authors review the very important question of the impact of iatrogenic low dose mercury exposure (mainly from amalgam) on public health.

Because of the spread and expected rising world wide use of dental amalgam, even subtle effects may cause a significant health problem. Therefore, the results of this review will have an important impact on future decision-making policies by governmental institutes and health agencies.

In fact, mercury is assumed to be the sixth most toxic in a universe of 6 million substances (Nascimento and Chartone-Souza, 2003) and the most toxic non-radioactive element. Any additional exposure to this element is associated with negative health effects.

Because of the possible political, financial and legal consequences of possible adverse effects, it must be assumed that many studies present biased information. Khier et al did use studies which try to downplay the toxicity of dental amalgam. Thus they can not find that three europaen countries have already banned dental amalgam (Finnland, Danmark, Sweden). In the case of of Sweden, the health inssurrancies did not cover the costs of dental amalgam use for patients since almost over 10 years.

The review is based mostly on literature, which are used and provided by the American Dental Association, which is the leading dental association of the world, to downplay the toxicity of dental amalgam. It is not easy for authors, to find all relevant studies to dental amlagam and to see the methodology of each. Thus it may happen that completely false conclusions where drawn, which was not the intention of the authors. Below some examples

The authors cite the study of Mackert & Berglund (1997) as an proof that the mercury body burden from dental amalgam is small. But beside the estimations by Mackert& Berglund (1997), which are based on a theroretical model, experimental and autopsy studies show clearly that amalgam bearers have elevated mercury (Hg) levels in biomarkers and until 10- times(!) more Hg in body tissues, like kidney and brain. Therfore, more than 80% of human Hg load is derived from dental amalgam. Thus, for the major part of the population amalgam contributes more to mercury burden than fish consumption [Drasch 1992, 1994, 1997; Kingmann et al. 1998, Lorscheider et al. 1995, Mutter et al. 2004, Nylander 1987, Pizzichini et al. 2003]. Therefore, it is not appropriate to state that the level of mercury exposure from dental amalgam is not high. Interestingly, dental boards also still claim that amalgam contributes only little or negligibly to the mercury load of man [Dodes 2001; Jones 1999; Larkin 2002; Wahl 2001a, 2001b, 2001c, 2002, 2003].

In contrast to methyl-Hg, there are no studies on mercury vapour that establish a safety limit. Furthermore, there is mounting evidence that mercury concentrations in blood and urine do not adequately represent the actual mercury body load [Danscher et al. 1990; Drasch 1997; Hahn et al. 1989, 1990, Hargeaves et al. 1988; Holmes et al. 2003; Lorscheider et al. 1995; Opitz et al. 1996; Vimy et al. 1990; Weiner & Nylander 1993]. Furthermore Drasch et al. [2001, 2002, 2004] showed, that 64% of mercury-exposed workers in gold mines in the Philippines with clinical-neurological signs of mercury intoxication had concentrations of mercury below the safety limit of 5µg/l.

For lead it is shown that health problems arise at blood levels far below of the presently accepted safety limits The same can be assumed for mercury, which means that safety values which up to now were regarded as reliable can neither be fixed for Hg nor for Pb.

No kidney damage?

The study by Mortada et al. (2002) is the only amalgam-study to have used a real control group with no previous exposure to dental amalgam (But perhaps before birth: via mercury from maternal amalgam fillings). The amalgam group had on average only 4 amalgam fillings for only approx. 4 years. And they found significantly increased markers of kidney damage.

Regarding the relevance of Urinary-Hg-levels and mercury body load or clinical symptoms: see remarks above. Indeed, fish consumption (this belongs to the species of consumed fish- see controversial data from the Faroe- Isles and the Seychelles) may protect from its mercury toxicity. Furthermore, the toxicity of methyl mercury (Me-Hg), which is bound to cystein in fish, seems to be far lower (only approx. 1/20) than Me-Hg-Cl or Me-Hg-J usually used in experiments [Harris et al. 2003]. Marine fish represents also a significant source of selenium and essential omega-3-fatty acids, which protects against mercury toxicity. Nevertheless, Me-Hg-Cl shows less neurotoxicity for the growing nervous system than mercury vapor from dental amalgam [Frederikson et al. 1996]. Another study also points to smaller neurotoxicity of Me-Hg from fish compared to iatrogenic Hg-sources (Amalgam, Thiomersal) [Holmes et al. 2003]. Here in contrast to iatrogenic mercury sources of the mothers, no correlation between maternal fish consumption during pregnancy and the risk of autism for their children was found.

Why should only a small proportion of individuals be affected by chronic low-level Hg exposure? The mentioned studies did not suggest this. Rather, the data by Mortada and Echeverria suggest that a relevant proportion could be affected. Given the frequent use of amalgam, this could indeed represent a serious health risk for the public. In animal experiments an impairment of the kidney function due to amalgam fillings could also be found [Boyd et al. 1991, Galic et al. 2001; Pollard et al. 2001].

Other studies found neurobehavioral effects of low dose mercury exposure: Mercury exposure below the safety limits, results in measurable changes in cognitive or neurobehavioral functions [Bittner et al. 1998; Echeverria et al. 1995; 1998; Siblerud 1989, 1992; Siblerud et al. 1993; 1994, Yoshida et al. 2004]. Color discrimination is impaired by a low Hg-exposure [Urban et al. 2003]. Dental staff shows neuropsychological symptoms [Echeverria 2002, Aydin et al. 2003; Ngim et al. 1992; Ritchie et al. 2002] or pathological muscle biopsies [Nadorfy-Lopez et al. 2000]. Visual evoked potentials in Hg exposed staff (among them dentists) show significant changes compared to controls [Urban et al. 1999]. A meta-analysis showed neuropsychological impairment in 686 persons exposed to Hg vapor compared to 579 controls [Meyer-Baron et al. 2002].

Lindh et al (2002) showed that a significant proportion of individuals, who presumably suffer from their dental amalgam, abouth 75% report a significant and long lasting improvement of their symptoms after amalgam removal. Prochazkowa et al. (2004) found an improvement after amalgam removal in 70% of individuals with autoimmune diseases (including MS), Low-dose mercury exposure is indeed considered a cause for autoimmune diseases by some authors, too [Bartova et al. 2003, Berlin 2003, Hultmann et al. 1994, 1998; Pollard et al. 2001; Prochazkova et al. 2004, Stejskal & Stejskal 1999; Stejskal et al. 1999, Sterzl et al. 1999, Via et al. 2003].

Low dose mercury exposure, as from dental amalgam, may be a substantial pathogenetic factor for Alzheimer´s disease (http://iospress.metapress.com/content/j33p87808363493j/)

To review the possible impact of mercury on autism accurately, the authors should include a broader range of references and information:

Pichichero et al ¹ argued that ethylmercury administered through vaccines was eliminated rapidly from the blood and rapidly excreted in stool. A recent analysis (Holmes et al. 2003) of mercury excretion in the baby hair of autistic children and controls provides an alternative interpretation. When combined with a closer examination of the stool findings, it may be believed that Pichichero et al. significantly overestimated the degree of mercury excretion. The rapid rate of mercury elimination from the blood and slow rate of excretion in stool suggest that ethylmercury from vaccines is rapidly taken up and retained in infant tissue. Higher tissue retention may help explain, why autistic children are more vulnerable to early mercury exposure and offer indirect support for the autism-mercury hypothesis (Bernard et al. 2001).

Most methyl mercury is eliminated from the body through stool and ethyl mercury from vaccines most likely follows the same path. Both mercury species must be removed from the blood to enter the feces, but elimination from blood simply means that the mercury has gone elsewhere, not necessarily eliminated. The recently published analysis of autistic and control baby hair suggested that while mercury was available to the hair follicles through the blood for excretion in normal infants, the baby hair of autistic infants contained very little mercury, only 0.47 micrograms (µg) per gram (g) compared to 3.63 µ/g in controls. The subjects in the study by Pichichero et al. (none of them autistic and therefore with no evidence supporting reduced excretion) provide another perspective on mercury excretion rates. They measured mercury concentrations in stool of 22 infants exposed to thimerosal in vaccines, ages two and six months, and found a range of 23-141 nanograms (ng) per gram of dry weight of stool, with a mean of 82 ng/g. The authors interpreted these levels as positive evidence of mercury elimination.

But these mercury concentrations are extremely low, not nearly enough to allow rapid excretion. Infant dry weight stool volumes have been measured at between 1-3 grams per kilogram (kg) per day, and so, with a mean weight of roughly 4 kg for a two month old infant, infant stool volumes may be expected to fall in the range of 4-12 grams (dry weight) per day. Taking the data from Pichichero et al, and using these simple assumptions, one come of other stylized estimates of excretion time for the 187.5 mcg of ethylmercury that infants received by six months of age before thimerosal was removed from vaccines.

Stool Hg concentration Daily Hg excretion Days to excrete 187.5 µg

(ng/g) (mcg/day) (days)

Minimum: 23 0.09-0.28 670-2,083

Mean: 82 0.33-0.98 191-568

Maximum: 140 0.56-1.68 112-335

In the best case, early vaccine exposure takes nearly four months to eliminate, but for normal children with reduced excretion, these concentrations suggest that the elimination process would take six years. For autistic infants, with evidence of reduced excretion in hair and additional fetal exposures (from maternal amalgam filling, fish consumption and Rho D immunoglobulin injections) (Holmes et al. 2003) these excretion times were likely far longer.

Khier et al. should also include the report by Madsen et al. published by the American Academy of Pediatrics in their journal Pediatrics¹, and a subsequent analysis by Hviid et al., JAMA 2003, which also has some important methodological flaws:

Madsen et al. and Hyiid et al compare the number of newly recorded autism cases prior to 1992, when thimerosal-containing vaccines were used, with those after 1992, when such vaccines were no longer produced in Denmark. The authors claim to observe a rise in autism rates after removal of thimerosal, and thus conclude that thimerosal plays no role in the etiology of autism. An in-depth analysis of the report reveals three major problems with the analysis and methodology.

• Autism counts were first based on hospitalized, inpatient records and then changed in the middle of the study period to include outpatient records. This new outpatient registry was introduced in 1995. Therefore, their purported increases after 1994 can be explained entirely by the registration of an existing autism population that did not require hospitalization. The authors minimize this discrepancy and do not adjust for it in their chart (Figure 1), yet in a prior study using the same Danish data, outpatients exceeded the inpatients by a ratio of 13.5 times, and represented over 93% of total cases. This huge gap clearly invalidates their inpatient data, the corresponding time period from 1970-94, and any evidence for a rising trend of autism in Denmark. The authors claim that inpatient admissions were rising also, but the “data [were] not shown”. They did not explain this omission, the only bit of credible data in their possession, since it compared equivalent populations.

• Additional discrepancies in the autism case counts make the trend assessment unreliable. After 1992, the registry added in patients from a large Copenhagen clinic, which accounted for 20% of the case load in Denmark (Byrd 2002). The patients from this clinic were excluded prior to 1992. Their inclusion in subsequent years would drive apparent increases in rates from 1992-1995 that was yet another form of registration effect.

• The diagnostic category used by the Danish psychiatric system changed after 1993 from “psychosis proto-infantilis” of ICD-8 to “childhood autism” of ICD-10. Psychosis proto-infantilis (code 299) is a category that has never been used in published autism surveys outside of Denmark. ICD-8 contained another, clearly more suitable code, 295.8 for “infantile autism”, which provided diagnostic criteria similar to current criteria used in ICD-10 and DSM-IV. The Pediatrics report mentions the diagnostic change in passing but fails to quantify its effect. In another paper using the same inpatient registry,(Bertrand et al. 2001) two of the investigators in the Pediatrics report note that the psychosis proto-infantilis category includes inpatient cases that do not fulfill the criteria for autism (which would further reduce the value of this case finding tool), while also noting the that ouptatient cases of autism in Denmark would not be captured.

• The autism trend data are described as an “incidence study”, a marker of quality in an epidemiological analysis. But the report is in no way a proper incidence study. It relies instead for its definition of the “incidence” of autism on the date when cases were entered into the new registry of outpatients. Many of these children were between 7-9 years old, and most were over 4 years old, when recorded as part of an increasing “incidence” trend. Yet the onset of autism must occur, by definition in the diagnostic criteria, before three years of age. Recording an “incidence” event at, say, seven years of age is clearly incorrect. Yet the authors record many such events to report an increase in registrations (especially after 1994) that they misleadingly describe as increasing incidence. The most widely used approach to assessing autism trends is to use year of birth as the “incidence time.” Failure to report the birth cohort incidence means that this study’s autism rates cannot be fairly compared with incidence levels observed in other countries.

• A recent study from the same group reported Danish autism rates for children born in the 1990s of 6 per 10,000. This falls below the rates of autism reported in the U.S. (over 30 per 10,000) by more than 80%. While emphasizing their illusory increase, the authors never mention that their rates are actually quite low. Although the estimates confirm that these Danish rates are very low in the 1990s compared to the U.S. or the U.K. (Ball et al. 2001) the authors fail to provide the most basic statistics that might enable a full comparison with other reports. These crucial omissions suggest a clear bias toward elevating the perception of Danish autism rates later in their study period.

• The report also estimates inpatient rates for the pre-1993 “psychosis proto-infantilis” at well below 1 per 10,000. If these were true rates for autism, these would be among the lowest rates measured anywhere in the world at any time period. This low rate would also contradict the single published survey of autism rates from Denmark, which indicated an autism rate of over 4 per 10,000 as far back as the 1950s. (Carbon et al. 2002)

1. The mercury exposure levels described in Madsen et al. are likely to be overstated. The authors describe a level of mercury exposure to Danish infants of 125 micrograms (mcg) by 10 months of age between 1970-92, a period in which they claim (without justification) that autism rates were low. All exposures came from the monovalent pertussis vaccine manufactured by Statens Serum Institut, which, according to the paper, provided the vaccine coverage rates reported therein.

• These mercury levels of 125 mcg are substantially lower and later than those scheduled in the U.S. in the 1990s, 187.5 mcg by six months.

• These ethyl mercury exposures --at 50 mcg per dose for the 9 week and 10 month injections--are the highest amounts ever described in any single vaccine dose.

1. The context for the early mercury exposures was completely different in Denmark when compared to any other country, and particularly compared to the U.S. and U.K., where autism rates are being watched most closely. The Danish report describes a different world of vaccine exposures and ignores exposures that are present today that were not present in Denmark in the 1970s. Autism onset has been reliably associated with exposure to viruses. In the cases where increasing thimerosal exposures have accompanied autism increases, numerous additional confounders were present that were not present in Denmark.

• Between 1970-92, the only childhood vaccine given in Denmark until 5 months of age was the monovalent pertussis vaccine.

• In the United States in the 1990s, children were exposed to multiple doses of diphtheria, pertussis, tetanus, polio, hepatitis B and haemophilus influenza B (Hib) vaccines before five months of age.

• In the United Kingdom, injections before age 5 months included multiple doses of meningitis C, polio, diphtheria, tetanus, Hib, and pertussis vaccines.

• Denmark did not administer thimerosal-containing Rho D immunoglobulin during pregnancy (see Holmes et al. 2003).

Public health authorities (now teamed with a Danish vaccine manufacturer) have published this study

To further evaluate a possible link between mercury and autism, Khier et al. should include the following references (animal and human studies) in their analysis, which all found an association between autism (speech disorders) and mercury:

1. Bernard S, Enayati A, Roger H, et al: The role of mercury in the pathogenesis of autism. Mol Psychiatry 2002;7(Suppl 2):S42-S43.

2. Geier DA, Geier MR. A comparative evaluation of the effects of MMR immunization and mercury doses from thimerosal-containing childhood vaccines on the population prevalence of autism. Med Sci Monit 2004;10:PI33-39.

3. Geier DA, Geier MR: An assessment of the impact of thimerosal on childhood neurodevelopmental disorders. Pediatr Rehabil 2003;6:97-102.

4. Geier MR, Geier DA. Thiomersal in childhood vaccines, neurodevelopment disorders, and heart disease in the United States. J Am Phys Surg 2003;8:6-11.

5. Morgan DL, Chanda SM, Price HC, Fernando R, Liu J, Brambila E, O'Connor RW, Beliles RP, Barone S Jr. Disposition of inhaled mercury vapor in pregnant rats: maternal toxicity and effects on developmental outcome. Toxicol Sci 2002;66: 261-73.

6. Blaxill MF, Redwood L, Bernard S. Thimerosal and autism? A plausible hypothesis that should not be dismissed.Med Hypotheses. 2004;62(5):788-94.

7. Blaxill MF. Study fails to establish diagnostic substitution as a factor in increased rate of autism.
   Pharmacotherapy. 2004 Jun;24(6):812-3; discussion 813-5.

8. Grether J, Croen L, Theis C, Blaxill M, Haley B, Holmes A. Baby hair, mercury toxicity and autism.
   Int J Toxicol. 2004 Jul-Aug;23(4):275-6.

9. Hornig M, Chian D, Lipkin WI. Neurotoxic effects of postnatal thimerosal are mouse strain dependent.Mol Psychiatry. 2004 Sep;9(9):833-45.

10. Lin-Wen Hu, J. Bernard and Che: Neutron Activation analysis of Hair samples for the Identification of Autism. Transactions of the Amercan Nuclear Society, v89, November 16-20, 2003

Another study, which was not mentioned by Khier et al. suggests maternal amalgam fillings can also be a risk factor for the development of autism in children [Holmes et al. 2003].

It is well known that Hg from maternal amalgam fillings reaches placenta and fetus [Ask et al. 2002]. It is also known from autopsy studies that the mercury levels in organs and brains of babies and children correlates linearly with the number of amalgam fillings of their mothers

These findings may indicate that autistic children may have an increased mercury level in their brains in spite of the decreased mercury level in their hair (Holmes et al. 2003) which is confirmed by another study on humans [Lin-Wen Hu, J. Bernard and Che: Neutron Activation analysis of Hair samples for the Identification of Autism. Transactions of the Amercan Nuclear Society, v89, November 16-20, 2003]. On the other side, higher infant hair mercury levels were associated with faster progress to developmental milestones [Grandjean et al.1995]. Experimental research has shown that the neurotoxic effects of thiomersal depends on individual susceptibilities, which are genetically determined.

In contrast to the conclusions by Khier et al., the studies mentioned above suggest that chronic low level exposure to mercury may lead to serious health problems. To evaluate the toxic effects of mercury exposure from dental amalgam, it is necessary to conduct a prospective, randomized, controlled trial (in the case of Alzheimer´s disease with a duration of 70 years). It is very important to identify possible sources of bias in most studies. It must be assumed, that pharmaceutical companys, health officionals, governmental institutions and dental boards, which have surely more influence to the media and medical research than unprofitable patient organizations, are very interested to denie any negative effects of “low-dose” mercury exposure from dental amalgams.

Other studies

Some other studies, not mentioned by the authors, found negative effects through “low dose” mercury exposure:

[Akiyama et al. 2001, Pizzichini et al. 2000, 2002, Pizzichini et al. 2001, 2003, Olivieri et al. 2000, 2002, Drasch et al. 2000, Björkman 1991; Ely 2001, Pendergrass & Haley 1996; Haley 2002, Duhr et al. 1993, Palkiewicz et al. 1994; Pendergrass et al. 1997; Pendergrass & Haley 1995, 1996, 1997; Leong et al. 2001; Cedrola et al. 2003, Godfrey et al. 2003, Berlin 2003; Dunsche et al. 2003a, 2003b; Martin et al. 2003; Wong & Freeman 2003, Guttman-Yassky et al. 2003, Gerhard et al. 1998a, 1998b; Gerhard & Runnebaum 1992, Rowland et al. 1994, Gerhard et al. 1998b, Sheiner et al. 2003, Podzimek et al. 2003, Frustraci et al. 1999, Lorscheider & Vimy 2000,Baasch 1968; Ingalls 1983, Ingalls 1986, Issa et al. 2003, Ahlrot-Westerlund 1989, Siblerud 1992, Siblerud & Kienholz 1994, [Huggins et al. 1998, Prochazkowa et al. 2004, Engel 1998, Casetta et al. 2001; Bangsi et al. 1998]. Bates et al. 2004 (Because the study cohort consisted primarily only of healthy persons at the time of joining themilitarity, which was selected by the process of military scrutiny, the odds ratio may be underestimated. One problem in this study is that there was no really amalgam free control group and the dental status before entrance to the military was unknown), Pamphlett et al. 1998, Pamphlett & Waley 1996,Albrecht and Matyja 1996, Adams et al. 1983; Schwarz et al. 1996,Rehde & Pleva 1994, Stejskal & Stejskal 1999, Kidd 2000, Lindh et al. 2002, Godfrey&Campbell 1994; Melchart et al. 1998, Stromberg & Langworth 1998. A risk analysis and review of the literature from 1997-2002 performed on behalf of the Swedish government comes to the conclusion that for medical and environmental reasons, “amalgam should be banned as fast as possible” [Berlin 2003].

In sum:

On the basis of the available literature, partly listed above, one reach a different conclusion to that of Khier et al. Therefore the authors should reevaluate the literature published so far and reconsider their conclusions. They can also use a recent review about dental amalgam for deeper independent information:

http://www.occupmed.com/content/6/1/2.

see list of publication below