DENTAL AMALGAM; A Report from an Ad-Hoc Working Group presented to the Commission and the Medical Devices Experts Group

Draft, June 1997

Comments by IAOMT-Sweden

The president of the Norwegian patient organization "Forbundet Tenner og Helse", Dagfinn Reiersøl, has written very penetrating comments on this report (23 July 1997, abbreviated as FTH below). We are in essential agreement with the many important points raised in his critique.

5.1 General remarks

As stated on page 19, the report draws heavily on five reviews which were selected on the grounds that they should:

a) cover the scientific literature comprehensively,

b) be commissioned by an official body and,

c) be publicly available

These reviews were of course not peer reviewed, nor were they published in scientific journals. Such official documents belong partly to the political sphere and must be viewed accordingly. They are not consensus documents in any current sense. As we know, the search for truth in controversial areas is always marked by some degree of intellectual conflict between people who have more or less openly taken sides. If one side is not fairly represented in an official group of scientists commissioned to review the amalgam problem, bias is inevitable.

Speaking of representation, we have failed to find even a single person among the members of the present EU Ad-Hoc Working Group who has been associated with research on amalgam from a critical perspective. The majority seem to be dentists or people affiliated with the dental industry. This is not an inappropriate comment, because the report will fulfill the above criteria (a-c) and can be expected to be used as a source in later reviews.

Peer reviewed articles

In addition to the five reviews just discussed, the draft report lists 218 references to papers that are supposedly "peer reviewed in a scientific journal" (p. 19). There is a notable exception in ref. #52, Edwardson (1995) which is an "Alzheimer's Disease Society Factsheet". This would be a minor departure from the rule if the reference carried little weight in the report, but on page 28 Edwardson's factsheet is cited together with Saxe et al. (1995) as underpinning the following statement: "Currently the weight of evidence is that there is no link between Alzheimer's disease and mercury". This superficial treatment of a complex and important research field can hardly be called scientifically responsible (1). The Saxe et al. study examines the statistical relationship between number of amalgam surfaces and intellectual abilities of 129 elderly nuns, and is not a study of Alzheimer's disease. Its design is in fact such that the findings are almost completely irrelevant to the problem of dementing disease.

The Swedish reviews

We are well acquainted with the two Swedish reviews (1992 and 1994), and will give more detailed attention to them. It is interesting that the 1994 review is only available in Swedish, with a four-page English summary. Even if some of the group members can read Swedish, this means that the great majority will only have a superficial idea of the contents and scientific merits of this review, which is extensively quoted in the report.


A recurring pattern in the rather intense activities in the amalgam controversy during the 1990s is that official groups keep quoting reviews by other official groups, creating an impression of consensus which has no counterpart in the actual scientific situation. The 1992 Swedish review is a striking example of this. The proceedings of this "state of the art conference" starts with an Introduction by Dr. Harald Löe, the then director of the US National Institute of Dental Research, who had also been involved in the 1991 NIH-NIDR Technology Assessment Conference (2). Dr. Löe quoted verbatim from the conclusions of this conference (3):

"Available data do not justify discontinuing the use of any currently available dental restorative materials or recommending their replacement."

With this he evidently struck a keynote at the very start of the Swedish conference, because in the conclusions drafted at the end of the next day the following is found (4):

"Available data do not justify discontinuing the use of silver-containing dental amalgam or recommending their replacement."

Dr. Löe was one of four members (and also a moderator) of the Scientific Panel appointed by the Swedish Medical Research Council to summarize the conference. The only difference between the two sentences just quoted is probably explained by the fact that the 1992 Stockholm conference was not intended to cover other restorative materials than amalgam.

For the historical record it should be noted that of the scientists originally invited to the Stockholm conference not even one had a record as a critic of amalgam use. When this became known, political pressure was exerted on the organizing committe, which (reluctantly, it is said) invited Dr. Murray J. Vimy, University of Calgary. He had no influence on the summary statement from the conference, and his highly critical open letter to the Secretary General of the Swedish Medical Research Council is available on the Internet (5).

Some of the thirteen papers read at the Stockholm conference are brief, ordinary research papers of rather indifferent quality. Important and difficult areas, notably epidemiology, and the psychosomatic aspects of amalgam-related illness were therefore not actually reviewed at this so-called state of the art conference.

It would be very hard to maintain that the 1992 Stockholm conference was not biased in favour of continued amalgam use. This conference was formally commissioned by a scientific body and not by a government agency, which makes its lack of balance all the more remarkable. It was not a consensus conference, and the contents of the official conclusions were dictated by the four people in the "scientific panel".


The 1994 Swedish review was written by an ad-hoc group commissioned by the National Board of Health and Welfare. This group had six members, one of whom was also present at the 1992 conference (Prof. Göran Möller). There were also two secretaries, one of whom is now a member of the EU ad-hoc group (Prof. Jan Ekstrand). In 1994 none of the eight persons forming this Swedish group had published any work betraying a critical attitude to the use of mercury amalgam for implantation in the human body.

In the usual manner, earlier reviews were summarized as background material. In addition to this, the report lists 248 references to scientific papers, two thirds of which were published in the 1990s. The relevant literature is growing rapidly, and the list of references is clearly selective and far from complete. One example is the treatment of Alzheimer's disease, which is rather superficial in the present draft review (6).

The following statement appears in the English summary (page 107):

"Scientific studies show that dental amalgam does not contribute to cardiovascular disease in women."

This is a rather more categorical statement than would be expected in a scientific review when discussing a disease of complex and largely unknown origin. The present ad-hoc group has accepted this statement and quotes it without qualification (5.3.2, last paragraph). The sole basis of this bold conclusion is actually a study by Ahlqwist, Bengtsson and Lapidus (1993), which is also cited in the present draft report (ref. 3). This study was criticized in detail by FTH (page 3-4), and we agree completely with the arguments put forward there. An unbiased reviewer would hardly have asserted that the absence of an effect of a toxic substance can be established on such slender and ambiguous evidence.

The authors of the Swedish 1994 review repeatedly give short shrift to studies in which the findings indicate that dental amalgam may have side effects outside the oral cavity. This somewhat hypercritical attitude is obviously selective. Quite doubtful evidence is frequently accepted which appears to explain alleged side effects of amalgam by some other hypothesis. Thus the mere fact that the patients may be given other diagnoses than mercury poisoning is regarded as somehow enlightening. The diagnoses may be somatic (page 48) or psychiatric-psychosomatic (page 49ff.), but the important thing is that the causes are still unknown in the great majority of cases. A diagnosis is not an explanation at the causal level unless it is based on causal knowledge. There is obviously some degree of confusion on this point behind several of the conclusions drawn in this review. The present draft report also shows evidence of the same kind of thinking, see e.g., last paragraph, and, first paragraph (page 42).

A diagnosis is of course first of all a descriptive label. Ideally it is linked to solid knowledge about the causes of the disease, but this is true only of a small minority of medical diagnoses. In the medical culture much is made of the empirical knowledge of treatment, course and prognosis of various diseases, and the art of diagnosis is an interesting form of pattern recognition. At the same time the lack of causal knowledge is being played down, and is given very little thought in everyday medical activities. It is therefore interesting how negatively and even arrogantly many doctors (and dentists) react to unorthodox hypotheses on possible causes of diseases. Whether they are well read in the research literature or not, such professionals may instantly dismiss unconventional suggestions with an air of annoyance. This may be due to misplaced pride in the medical science which has failed to elucidate the causes of most diseases ("if it were as simple as that, we should have known long ago"). Conventional medical thinking often favours certain standard kinds of provisional and incomplete hypotheses about causes (genetics, diet, stress, psychosomatics, etcetera), and competition tends to be resented. But in addition there is always the tacit assumption that a definite diagnosis implies a definite cause which is the same in all cases. This is completely unfounded, and a major block to scientific progress, particularly in the field of chronic diseases. Medical research would become much more flexible if we adopted the commonsense idea that there may be more ways than one to acquire a disease.

The review section on psychosocial and psychiatric aspects (pages 49-58) is one of the longest, but is unfortunately riddled with question-begging and circular reasoning based on the ambiguities of diagnosis just discussed. Psychiatric diagnoses are much more artificial and ill-defined than somatic diagnoses (the DSM classifications notwithstanding), and causal knowledge in this area is very poorly developed. A frequent theme in psychiatric studies of amalgam-related illness is that several of the patients can be given psychiatric diagnoses, and that therefore mercury from amalgam can have nothing to do with their complaints. This is plainly a non sequitur of a rather embarrassing kind, which is however eagerly accepted by many dentists and physicians with a stake in the amalgam controversy.

We have devoted more space to this review than to the others because the great majority of the members of the ad-hoc group are unable to study it in detail. From a scientific point of view this review is biased and unreliable, which can hardly be evident if only the English summary is considered.

5.2 Mercury, its release from dental amalgam and fate

"The potential for dental amalgam to be a source of mercury exposure was shown by Stock (1939). Until about 15 years ago there was little interest in these findings since any adverse effects from this source were considered unlikely given the very small amounts of mercury involved" (page 20f.).

The paper cited is obviously misdated. Alfred Stock published this paper in 1928, and all the bibliographical details except the year are correct (ref. 188). If 1939 had been the actual year of publication, the statement that "there was little interest in these findings" would have contained a grain of truth. This year marked the end of "the second amalgam war", which had then raged in Germany for more than ten years and was fuelled by widespread interest in Stock's findings! The reason why it ended was of course the world war which started the same year.

This amalgam war only 60 years ago in a scientifically leading country was of course a major event in the history of dentistry. It is hard to believe that none of the ad-hoc group amalgam experts have even an inkling of the truth about this (7).

5.3 Toxicity, mercury and dental amalgam

The occupational exposure of dental personnel is briefly mentioned in various places throughout the present report, but is not actually reviewed. This is an important area with substantial and partly alarming findings, and we would gladly supply hundreds of references. A handful are given in the below footnote (8), and more can be found in FTH. The findings range from signs of polyneuropathy and poorer performance in various psychological tests to a significant excess incidence of malignant brain tumours. Other general health complaints attributed to dental amalgam

"Some studies reported that patients improved after their dental amalgam fillings were replaced by another dental filling material. However, these reports have not been controlled for potential placebo effects." (page 42)

This is one of the crucial points in the amalgam controversy. Removal of amalgam is analogous to stopping a medication, as is routinely done when side effects occur. If the removal is done properly by a skilled dentist, symptom reduction is often strikingly good in cases of amalgam-related illness. This is confirmation that the amalgam (or rather the metals leaking from it) actually gave rise to side effects in these cases, and the parallel to the logic used when handling side effects of drugs is obvious.

There is no requirement to "control for placebo effects" when side effects of drugs are being studied. This is significant, because drugs are distrusted by many people, and side effects are very much expected to occur. According to the (unproved) assumptions underlying placebo-controlled clinical trials there should be numerous cases in which the alleged side effects of drugs were actually psychosomatic phenomena, or even somatizations of anxiety. The conventionally "correct" way of sorting this out should be to give placebos to some of the patients experiencing side effects, while the others are continued on the drug. This is not being done, and it would probably be hard to find anyone who is enthusiastic for such a trial protocol. The reasons are obvious. Adverse effects of drugs is a sensitive area, and it would be very damaging to the goodwill of both the pharmaceutical industry and the medical profession if the "It's-All-In-Your-Head" stereotype became associated with drugs in this way, suggesting an unwillingness to accept responsibility. Therefore common sense is used in matters of side effects, and we never hear any complaints that this very important medical field is managed in an "unscientific" way.

Why should the study of the side effects of dental amalgam be treated differently? It would be interesting to hear the arguments, because it is generally taken for granted that the usual principles should apply which refer to experimental treatments of well-defined diseases. Amalgam-related illness can not be regarded as "a disease" in the ordinary sense, but is a rather bewildering collection of relatively rare individual reactions to a toxic substance, exactly like side effects of drugs. The inevitable lack of "cardinal symptoms" and a consistent "picture" is in fact often (wrongly) used against the idea that amalgam-related illness is a chemically induced health problem.

A physician trained in the art of diagnosis will smilingly shake his/her head at an "impossible" list of symptoms like the one shown in Table 3 (page 41) of the draft report. The very same physician will feel entirely at home with an elaborate list of the side effects of a drug, such as is found by the hundred in commonly available reference books and in package inserts. We have appended a list of the side effects of indomethacin which amply illustrates the point just made.

This is a question of different perspectives, apparently, and both are needed because the problem of side effects is different from the problem of disease. Each should be used only in the proper circumstances. We are interested in arguments why the disease perspective should be preferred in the amalgam field, but to us the side effect perspective seems very natural, scientifically as well as from the point of view of common sense.

If amalgam removal works in such a way that a substantial proportion of patients are permanently relieved from long-standing symptoms it is in fact absurd to attribute this to placebo effects. It is equally absurd to cast doubt on amalgam removal because it does not make every patient better. On pages 42-44 the draft report quotes approvingly from a medley of sources in which such arguments are put forward. (See also page 60f.) The impression inevitably conveyed between the lines is of course that the idea of removing heavy metal implants from the body is disturbing to the ad-hoc working group.

The 1994 Swedish review goes particularly far in attributing striking health effects of amalgam removal to placebo mechanisms. One of the best published studies of the long-term effects of amalgam replacement (9) was discussed at length on page 97f. (in an out-of-the way chapter), and the results after a follow-up time of four years were described as remarkable. However, since about half of the 100 patients were in some sense selected, the whole study is written off as inconclusive and is not even included in the main list of references of this review. (The "selected" patients did not in fact show more improvement than the others.) The obvious implication is that placebo effects are a more likely explanation than an actual effect of heavy metal removal.

In an introductory chapter, the 1994 Swedish group discussed amalgam removal and stated that: "nobody has been able to show in a systematic way how permanent these improvements have been" (page 14, our translation). This contradicts even their own reading of Redhe's book in a most remarkable way.

The idea that amalgam replacement might act as a particularly powerful placebo is explicitly advanced on page 56f in the 1994 Swedish review. In contemporary medical culture placebos are somewhat mythical things with remarkable powers, but nobody seems to have actually seen any scientific evidence supporting this. There is even a dearth of published case reports, which leaves the whole concept in the sphere of medical lore and urban legends. This is particularly embarrassing if we are supposed to take it on trust that many patients are permanently relieved from long-standing symptoms by the application of nondescript placebos.

There is a very interesting recent book on the placebo question, "Der sogenannte Placeboeffekt" by Gunver Sophia Kienle (10). Dr. Kienle cuts the whole problem down to size in a rather surprising but wholly convincing way. The origin of the modern placebo doctrine was only 40 years back, and the tradition still relies heavily on a much quoted 1955 paper by H.K. Beecher, which turns out to be remarkably untrustworthy on closer analysis. There is in fact no scientific evidence for strong and universal placebo effects.

5.3.9 The effect of dental amalgam on the immune system

According to the present draft report the 1991 WHO report stated that:

"A special problem in the risk assessment of mercury is the fact that mercury can give rise to allergic and immunotoxic reactions, which are partly genetically regulated. There may well be a small fraction of the population that is particularly sensitive, as has been observed in animal studies. A consequence of an immunological etiology is that it is not scientifically possible to set a level for mercury, e.g. in blood or urine, below which mercury-related symptoms will not occur in individual cases, since dose response studies for groups of immunologically sensitive individuals are not yet available."

After this rather weighty quote, the draft report cites other trusted reviews (11) to the apparent net effect of defusing the WHO report. This is unfortunately not warranted by the scientific situation, because the authors of the WHO report were not arguing at the same level as the other authors cited. If anything is likely to give rise to unpleasant surprises in the future, it is attempts to "clear" dental amalgam in this way, relying on inadequate empirical studies of immunological problems (12).

6. Biocompatibility of Alternatives to Dental Amalgam

Slightly more than eight pages are devoted to the biocompatibility of alternatives to dental amalgam. This is an important subject. The most interesting aspect of the group's treatment of these problems is the striking change of attitude when the mercury amalgam problem is temporarily out of focus. The literature on alternatives is reviewed critically according to the "side effects model" and without any attempts to gloss over problems. Dentistry uses at least a dozen metals other than mercury, and even those are reviewed in the same critical vein.

7. Risk Assessment

As amply shown by FTH, the ad-hoc group totally lacks expertise in risk assessment and grossly underestimates the margin of safety required according to modern toxicological standards. We agree completely with this critique. It is more than remarkable that the group's lengthy discussion of the risk assessment of amalgam which was drafted in 1997 contains no reference to the Richardson and Allan (1996, reference in FTH) paper.

7.5 Alternatives to dental amalgam

The toxicity and other risks associated with some of the alternative materials are of course not negligible, but compared to mercury the effects of these toxic constituents are very much limited in time. As in chapter 6 the reviewers leave no doubt about their critical attitude to modern alternatives to amalgam.

7.6 The benefits of dental amalgam

This section contains an eulogy of amalgam which ends with the following two paragraphs (page 79):

"7) Dental amalgam is a plastic material. It is mixed as a paste and inserted into a prepared cavity. Unlike cast gold or most ceramic restorations, laboratory procedures are not required. Dental amalgam restorations can be placed in one visit at the chairside demonstrating the ease of use and the convenience of the material both for the patient and the clinician."

"8) Overall dental amalgam is the least expensive of the permanent restorative materials in terms of direct cost, frequency of replacement and requirement of professional time."

Is this ad copy or a scientific review? In any case the reader will know at a glance what the authors' preferences are.

11. Conclusions

"2. In recent years toxicological and biocompatibility aspects of dental amalgam have been reviewed extensively and risk analyses carried out. There are no data at present to indicate that mercury from dental amalgam fillings will cause an unacceptable health risk to the general population."

"4. No systemic toxic effects have been shown to be related to the release of mercury from dental amalgam fillings. In particular, evaluation of the literature indicates that no systemic toxic effects would be expected to arise from exposure to mercury at levels associated with the presence of dental amalgam fillings." (page 103)

"Taking the evidence that our group has reviewed, the benefits of restoring teeth with dental amalgam outweighs significantly the documented risks."

"10. There are only a limited number of case reports of adverse reactions related to dental amalgam fillings, even though this material has been placed in billions of teeth. Except for hypersensitivity to components of dental amalgam there are no scientifically substantiated reports that self-reported symptoms ascribed to toxic effects of dental amalgam are relieved by the removal of amalgam fillings." (page 104)

Leaving the other statements aside, it is abundantly clear that the last sentence above is simply untrue. There are certainly tens of thousands of people or more who know this from personal experience. Special pleading accompanied by talk about placebo effects and the rigorous demands of the scientific method will do nothing to alter this, but may help to undermine the goodwill of science.

As our Appendix on the side effects of indomethacin shows, there are drugs on the market which have a rather unpleasant record of side effects. This is considered acceptable, but of course only on the condition that the side effects are recognized for what they are, with no attempts at belittling or explaining away. The minority of patients who react adversely would never accept being treated as possible psychosomatic cases, or being subjected to double-blind procedures. The drug manufacturers know this, and use their common sense in order to prevent the build-up of an explosive situation.

The risk assessment of amalgam is one thing, and helping the people who get side effects is another. This simple lesson can obviously be learnt from the drug example. In the long run it is impossible to maintain that the risks are acceptable if the victims of side effects are not being treated well. Under no circumstances can the future of amalgam be safeguarded by denying the problems. It is very much to be regretted that the EU Ad-Hoc Group continues the strategy of denial that the dental profession has long been following. Time is running out in this era of public awareness, and much is to be gained from a change of attitude.



Birger Gran, M.D. Per Dalén, M.D.

(To the beginning of the dokument)


(1) The ADA makes use of the Saxe et al. study on their website at under the title "Study Finds No Link Between Amalgam and Decrease in Mental Abilities".

(2) Cited as "USA National Institutes of Health (1992)" in the present draft report

(3) Page 11 of the Proceedings book from the Swedish conference.

(4) Page 29 of the State-of-the Art Document, a separate booklet with the same title as the Proceedings volume.

(5) URL:

(6) Some additional references:

Basun H, Forssell LG, Wetterberg J and Winblad B (1991). Metals and trace elements in plasma and cerebrospinal fluid in normal ageing and Alzheimer's disease. J Neural Transm Park Dis Dement Sect 4:231-58.

Duhr E, Pendergrass C, Kasarsis E, Slevin J and Haley B (1991). Hg2+ induces GTP-tubulin interactions in rat brain similar to those observed in Alzheimer's disease. FASEB J 5:456.

Ehmann W D, Markesbery W R, Alauddin M, Hossain T I M and Brubaker E H (1986). Brain Trace Elements in Alzheimer's Disease. Neurotoxicology 7:197-206.

Pendergrass JC, Haley BE, Vimy MJ, Winfield SA and Lorscheider FL (1997). Mercury vapor inhalation inhibits binding of GTP to tubulin in rat brain: similarity to a molecular lesion in Alzheimer diseased brain. Neurotoxicology 18: 315-324.

Thompson CM, Markesbery WR, Ehman WD, Mao YX and Vance DE (1988). Regional brain trace-element studies in Alzheimer's disease. Neurotoxicology 9:1-8.

Wenstrup D, Ehman WD and Markesbery WR (1990). Trace element imbalances in isolated subcellular fractions of Alzheimer's disease brains. Brain Research 533:125-131.

(7) Frykholm (1957) is cited as a reference in the relevant paragraph, but he was obviously aware of the fact that an "amalgam war" started in 1926, and his reference to Stock (1928) is correct.

(8) Ahlbom A, Norell S, Rodvall Y and Nylander M (1986). Dentists, dental nurses, and brain tumours. Brit Med J 292: 662.

Echeverria D et al (1995). Behavioral effects of low level exposure to Hg0 among dentists. Neurotoxicol Teratol 17: 161-168.

Jokstad A (1990). Mercury excretion and occupational exposure of dental personnel. Community Dent Oral Epidemiol 18: 143-148.

Nylander M (1986). Mercury in pituitary glands of dentists. Lancet i: 442.

Nylander M, Friberg L, Eggleston D and Björkman L (1989). Mercury accumulation in tissues from dental staff and controls. Swed Dent J 13: 235-243.

Pohl L and Bergman M (1995). The dentist's exposure to elemental mercury vapor during clinical work with amalgam. Acta Odont Scand 53: 44-48.

Ritchie KA, Macdonald EB, Hammersley R, et al. (1995). A pilot study of the effect of low level exposure to mercury on the health of dental surgeons. Occup Envir Med 52: 813-817.

Shapiro IM, Cornblath DR, Sumner AJ et al. (1982). Neurophysiological and neuropsychological function in mercury-exposed dentists. Lancet i: 1147-1150.

Skare I et al(1990). Mercury exposure of different origins among dentists and dental nurses. Scand J Work Envir Hlth 16: 340-347.

Zanders D et al (1992). Mercury exposure of male dentists, female dentists, and dental aides. Zentralbl Hyg Umweltmed 193: 318-328.

(9) Redhe, O (1991). Sjuk av amalgam. R-Dental AB, Falun, Sweden. ISBN 91-7970-884-6.

(10) Stuttgart, New York; Schattauer, 1995, ISBN 3-7945-1687-7, 99 pages.

(11) 1992 NIH report; Swedish Medical Research Council, 1992; Swedish Report, 1994.

(12) The following references are examples of relevant material not covered by the ad-hoc group:

Bigazzi PE (1992). Editorial: Lessons from animal models: the scope of mercury-induced autoimmunity. Clin Immunol Immunopathol 65:81-84.

Hultman P, Johansson U, Turley SJ, Lindh U, Eneström S and Pollard KM (1994). Adverse immunological effects and autoimmunity induced by dental amalgam and alloy in mice. FASEB J 8: 1183-1190.

Katsunuma T, Iikura Y, Nagakura T, Saitoh H, Akimoto K, Akasawa A and Kindaichi S (1990). Exercise-induced anaphylaxis: improvement after removal of amalgam in dental caries. Annals of Allergy 64: 472-475.

Lindqvist B and Mörnstad H (1996). Effects of removing amalgam fillings from patients with diseases affecting the immune system. Med Sci Res 24: 355-356.

Stejskal VDM, Cederbrant K, Lindvall A and Forsbeck M (1994). MELISA - an in vitro tool for the study of metal allergy. Toxic in Vitro 8: 991-1000.

Tibbling L, Thuomas KÅ, Lenkel R and Stejskal V (1995). Immunological and brain MRI changes in patients with suspected metal intoxication. Int J Occup Med Toxic 4: 285-294.

(To the beginning of the dokument)