I n t r o d u c t i o n
Is amalgam health-promoting?

The below described Ahlquist et al study, was made on a group of women in Gothenburg, Sweden, from 1988-1995. One of the few epidemiological studies on the subject, it has been frequently cited by expert groups and odontologists especially in Germany, adhering to the view that amalgam fillings pose no threat to health and that the material may even have some health benefits. The methodology of said study was described and critizised by Bo Walhjalt, Sweden, in Heavy Metal Bulletin No. 3, 1997 ("The need for reassessment of mercury assessments"). In the media, the study has inspired headlines such as: "Amalgam health-promoting". In the article below, Mats Hanson thoroughly investigates the background and methodology of the study. The article was first published in TF-bladet, journal of the Swedish Association of Dental Mercury Patients, No. 2, 2002. Monica Kauppi

Misuse of women in Gothenburg

by Mats Hanson, Ph.D. Sweden (see also separate information)

Epidemiologic studies seldom give clear results, often because factors which could affects the results have - intentionally or unintentionally - been overlooked.

Connections to industry?

Everyone can easily find out that the risks with potentially hazardous products and conditions in society are often dismissed by using presumably scientific, epidemiological studies, and later it turns out that those behind the studies have connections to the industries producing the products, defend earlier positions or have positions in society where it is more or less clear that it is expected that the current position should be defended. The tobacco industry is a well known parade example. Some respectable medical journals now demand that the authors should state dependencies and financial sources.

We have long experience with "research" intended to dismiss health damaging effects of amalgam. Initially it was denied that any mercury at all was released from amalgam: "...even with the use of modern physical and morphological analyses has any form of systematic degradation of amalgam fillings been detected, not even during decades." (Bergman, Glantz, Nilner, Olsson, Läkartidn. 79, 1982, 879).

Evidence as early as 1874

The evaporation of toxic amounts of mercury from amalgam was demonstrated already 1882! (Talbot). When mercury eventually was found in the brain in amounts proportional to the number of fillings in the mouth, it was denied that any health effects could occur. "Mercury in the brain - so what!" stated the prof. of dental toxicology, K.S. Larsson, at the Medical Research Council State of the Art Conference in 1992, a conference which was introduced by Harald Löe, head of National Institute of Dental Research, USA, with the words: "This conference is intended to give arguments to the dentists and assurances to the public."

150 years of odontological malpractise

The degradation of science to an instrument for defending 150 years of dental treatment with amalgam was noticed already 1956 in a book by a german dentist (H. Struntz: Treason against your health): "Influential dental circles sabotage a resolution of the amalgam issue." and the Council for Planning and Coordination of Research (FRN) stated that: "During the evaluation FRN has been forced to recognize an extensive inclination of researchers in the amalgam area to neglect the scientifically critical attitude."

Healthier with amalgam

A number of studies, based on a population study of women in Göteborg, have been presented and widely published in the media as a support for the idea that you will be healthier the more amalgam you have in your teeth, or at least that women in Göteborg are not worse off if they have much amalgam, compared to little. The question is: Have the researchers behind these studies adhered to a scientifically critical attitude?

The studies started in 1968-9 and were intended to illuminate the health status of a representative sample of women, also dental health. The dental status were presented in a number of papers by Halling, Bengtsson et. al. and can be found i a dissertation by Halling 1987, most of the papers have been published in Swedish Dental. Journal. Some of the papers contain important information, essential for the relevance of later papers on amalgam load and health.

Halling's papers are only referred to in two later studies related to amalgam and then only as: " From a panoramic survey the number of remaining teeth, restorations, crowns, pontics and endodontically treated teeth were assessed. For each dentolous woman the number of toth surfaces filled with amalgam was registered." (Ahlqwist et al, 1988) and in Ahlqwist et al 1993 that the number of single women were somewhat overrepresented in women groups which did not participate in the investigations.

What information is provided by Halling?

Halling made a compilation of results from all patients and the additional information given by a more careful study of 75 patients from the various age groups. These were then considered representative for all women. In the larger group the dental status was assessed by panoramic x-ray.

Table: Dental status in patients

Thus about 5 healthy, unrepaired teeth in 38-year old and some single tooth in 60-year old. 50-year old had 2-3 unrepaired teeth. At age 38, 4 % completely lacked own teeth (in the latest data, i.e. persons having most of their dental care in the 80-90ies there are fewer persons completely lacking teeth, something usually meaning more crowns and root fillings instead of extractions).

46-year old: 11 % lacked own teeth
50-year old: 18 % "
60-year old: 40 % "

Root fragments etc (impacted teeth) were found in every 5th patient (in the 75-person investigation). It is likely that this is more common in older persons compared to younger ones, but there is only information on the total 75-person group of mixed ages.

Crowns without screwposts: 1-2 per person.
Crowns with screwposts: 2-3 per person..
In e.g.. 50-year old 23 % of remaining teeth had crowns; 3,5 per person.

Bridges: Number of supporting teeth 2 per person, more common in higher ages.
Pontics ~ 1 per person.

Endodontically treated teeth: 3-4 per person. In 50-year old 21 % of remaining teth root-filled.

Number of teth with osteolytic lesions (focal infections) about 2 per person (in the 75-person-group, mixed ages).
Isolated areas (not adjacent to teeth) of osteolysis, osteosclerosis (today called cavitation, NICO) + "foreign bodies": 1 in every 4th person. (75-person-study).

The difference between the number of remaining teeth in the various age-groups (14,6 teeth at 50) and a useful bite consists of bridges, partial prothesis in one or both jaws, partial in one and full denture in the other up to a completely artificial dentition.

In is not clear whether Halling has considered "amalgam crowns", i.e. teeth almost completely built up by amalgam as crowns or amalgam fillings.

Endodontically treated teeth often have an anchor consising of an easily corroded, gold-plated brass screw (in Sweden). The gold layer is thin and just promotes corosion. Zinc in the brass alloy dissolves first. Pins of gold alloy also occur but since these were more expensive they were more rarely used. If the dentist was greedy or had used up the brass pins he could use other useful items, e.g. pieces of paper clips (we know of many such examples).

Materials characteristic of the period

From Halling's material it is possible to reconstruct what "typical" 38-year old, 50-year old etc. had in their mouths.
"Typical" 38-yeal old: (106 in the whole study) 21,9 own teeth remaining, 2-3 of these with crowns, about 3 endodontically treated teeth, about 16 teeth with fillings, 12 of them with amalgam. Every 10th to 12th person has a bridge (gold alloy at this time).

"Typcal" 50-year old (302 in the study): 14,6 remaining own teeth, 12 of them with fillings, 9 of these with amalgam, 3,5 crowns, most of them with screwposts, + an additional root filled tooth, pieces of old roots in every 5th person. Teeth with osteolysis (bone destruction) about 2 per person, number of teeth supporting a bridge 2 per person. Every 4th person had isolated areas of osteolysis, "foreign bodies" etc.

"Typical" 60-year old (211): 8,4 remaining own teeth, root fragments etc. as by 50-year old, 2-3 crowns per person, endodontically treated teeth ~ 2 per person, thus at most 6,4 own, not root-filled teeth, some of them can serve as supports for bridges. Bridges and partial dentures in 30 %.

Questions accumulate

Then the obvious question is whether it is really possible to find enough women with more or equal to 20 amalgam fillings (460 it is stated in the statistics) and 193 with 0-4 fillings to have a statistical material to draw any conclusions.

Even more questionable is the star-shaped figure in which it is shown how much more of symptoms are present in 50 year old women with 0-4 fillings, compared to women of the same age with more or equal to 20 amalgam fillings (Ahlqwist et al 1988). The number of women represented in the figure is not stated. Was it possible to find any number of 50 year old women in Sweden (in the 1980ies) with 0-4 amalgam fillings and not the jaws built up with root-filled teth, crowns, bridges and dentures?. If they existed they should have been rare and especially it should not be possible to find any substantial number among 302 randomly selected women in that age. And what did they have earlier in their missing teeth?

Misleading information

The authors have gone out in the media with the information that it was an impressively large study, 1024 women, and that the results showed that women with few fillings had more symptoms that those with many amalgam fillings. The numbers compared were 460 + 193 = 657. Almost as a foot.-note it is stated that when the figures were corrected for age most of the associations disappeared. Remaining were joint pains, diarrhea and poor appetite. Corrected also for socio-economic group among the women themselves or for their husbands also diarrhea disappeared but irritability was added (cruel husband?). Common practice in epidemiological studies is to first correct for confounding factors like age before any conclusions at all are drawn - not the reverse. The symptoms displayed in the study are hardly directly indicative of amalgam poisoning. The authors conclude that the study does not support that amalgam fillings in the teeth should be removed if there are no symptoms. We agree, but for completely different reasons, namely that a major proportion of dentists are unable to remove amalgam without heavy mercury exposure for the patients (and dentists). Careless amalgam removal might precipitate symptoms in previously healthy persons.

Strong connection

Before realizing that the old data could be used to relate symptoms to number of amalgam fillings, a study was published by Bengtsson et al, 1987 where a strong correlation was found between the same symptoms and low blod pressure (less or equal to 120 mm). A figure is presented for a subgroup which used beta-blockers and looks precisely as the one which later was presented for symptom-amalgam. Almost all symptoms which are more common in persons using beta-blockers, compared to a group using diuretics (not mercury diuretics presumably), are also more common in women with few amalgam fillings, according to later publications. Strange!? To the best of my knowledge there are not especially many amalgam patients which use beta-blockers.

Weaknesses in the investigation

The medical examination consisted of, in addition to standard laboratory tests, a questionaire where the women were asked to answer whether any of the symptoms from a list of 30 had troubled them any time during the most recent 3 months. The authors were apparently aware of the weaknesses of such an examination and write: "Conclusions will be drawn only when dramatic differences are presented between different groups of age and sex." (also males participated in this part).

Such differences are a pronounced reduction of general tiredness after 50 year of age, reduction of abdominal pains and headache with age, increase of joint and muscle pains with age and a pronounced peak for increased weight and sweating in women, 50 years of age (the nearest younger group was 38 and had not yeat entered menopause).

Meagre exposure data

These are the basic data on which a number of studies of amalgam relating to health have been based. One study relate these meagre exposure data with cardiovascular disease (infarcts), diabetes, cancer and early death (data extracted from the swedish cancer registry and hospital records) (Ahlqwist et al, 1993). None of these diseases have any clear connection with the common problems experienced by amalgam patients. In this study the statistics is based on the number of tooth surfaces with amalgam (more or equal to 20 or 0-4 surfaces), not number of amalgam fillings. In another place in the same paper (Table 3) the number of amalgam surfaces appear and in Table 4 the number of amalgam fillings (Table 4 lists the p-values for the data in Table 3). Has a dentist really written this? Do they really know what they base their statistics on?

In Ahlqwist et al, 1995, the number of tooth surfaces with amalgam reappear and that this had been established by panoramic x-ray and photos in 1968-9. Nowhere in Halling's papers is there any measurement of tooth surfaces with amalgam. Under the heading "results" in Ahlqwist et al, 1995 the text states the number of amalgam fillings. The results in the tables state number of surfaces.

Serum samples were drawn in 1968-9 and in 1980-1. 20-30 years later they measure methyl mercury and inorganic mercury in these samples and relate to amalgam. To the best of my knowledge there is not one study published in which mercury has been measured in serum samples, decades old. A minimum of quality control had been that someone, anywhere in the world, anytime had measured Hg immediately in serum samples, saved these for 20-30 years and then measured again with exactly the same method to see if mercury had been lost or absorbed. The samples from 1968-9 were taken in open glass tubes, in 1980-1 in plastic vacuum tubes. Since the samples never were intended for Hg measurements there was certainly not a thought about contamination or absorption. G. Drasch, well known trace element expert in Munich, states (Hock, Drasch et al., 1998, p.61) that you can save blood samples for trace element analysis for at most 3 days at 4°C. "It is mandatory to measure mercury levels within a few days of blood collection because storage of mercury-containing specimens in plastic test tubes over several months leads to deposition of mercury in the test tube walls and a loss of mercury in the specimen."

In the Hg-study (Bergdahl et al, 1998) the number of amalgam surfaces reappear and the patients, 46 resp 58 years of age 1968-9 and 1980-1 are reported to have a mean of 21 resp. 15 amalgam surfaces, with a range 0-55 and 0-53 surfaces. The diffrences in Hg-content are minimal. In several of the stated ranges (variations between highest and lowest values), the lowest values are lower than the sensitivity of the measurement method according to the description in the text. How have these values been obtained?

The same weaknesses apply to a study where lab-values and disease are related to Hg-level (Ahlqwist et al 1999). Again the number of amalgam fillings reappear and their relation to serum-Hg. A review paper in Läkartidningen (Bengtson et al, 2001) only refer to number of fillings, also for the studies where number of surfaces or amalgam filled surfaces have been reported in the original papers!

Lavstedt and Sundberg published a similar study on the relation between amalgam and symtoms but the only groups where any difference was seen were between persons with amalgam and those with no own teeth. The authors state "in the clinical registration only the material in the artificial crown was noted but not the presence of possible amalgam fillings at the crown margins, made because of secondary caries. Such fillings were relatively common at the time of the registration" (1970).

Amalgam under crowns and bridges

None of the studies indicate that amalgam regularly occur under crowns and bridges; gold bridges are actually cemented with amalgam, removable dentures are clasped to remaining, amalgam-repaired teeth. "Gold crowns" are almost regularly a shell of gold alloy over an amalgam core. If a bridge is placed (always made of metal at the time of the studies) existing amalgam fillings are not removed. Bridges over the front teeth consist of pocelain fused to a metal skeleton, containing easily oxidized and toxic metals in order to obtain a strong binding.

We have, at various meeting, repeatedly tried to get C. Bengtsson and M. Ahlqwist to explain what the patients really had in their mouths. Ulf Bengtsson wrote a letter to Ahlqwist (1988) where he asked the following questions:

A. Does the control group contain women who do have 0-4 amalgam fillings but also have other types of restorations like bridges, crowns etc?
B. It is very common that gold crowns are placed on an amalgam core. Can such gold crowns on amalgam occur which have been recorded as non-amalgam?
C. Are there in the control group amalgam fillings in contact with adjacent gold crowns or bridges of metal?
D. It is obvious that it is unusual with such intact dentition as the control group gives the impression of. Is it possible that the control group can have just as bad or worse dental health relating to fewer own teeth, but that the replacements, because of extensive damage, are of other materials than amalgam?
E. Can the control group, historically and without considering when the fillings were made in relation to the time of examination, have had as many or more amalgam fillings than the group with more than 20 amalgam fillings?

Reply from M. Ahlqwist:
A. Yes, it happens.
B. Might occur, but it is then rare and can not affect the results.
C. Not specifically investigated.
D. There is a relationship between the number of teeth and the number of amalgam fillings, meaning that
those with fewest number of teeth often have the fewest number of fillings. This is the reason why we have included the number of teeth as a background factor in our statistical analyses.
E. Theoretically, yes. But not in practice. Since we have followed the women during a 12 year period before the last examination and during that period there were few changes.

Ahlqwist et al seem to presuppose that there is a linear relationship between the number of amalgam fillings (or tooth surfaces with amalgam?) at the time of examination and health, without recognizing hidden amalgam fillings, earlier several sets of amalgam which have been drilled away, often without protection.

Our experience is that it is often a gold-amalgam combination, root fillings with gold-plated brass screw-posts and amalgam on top and even a gold-shell on top of this battery, poisonous root-filling materials (e.g. N2, endomethasone), polishing amalgam without any protection at all, placement of metal-ceramic constructions in already amalgam-filled jaws etc. which precipitate acute illness and prolonged symptoms (years or decades), symptoms which do not disappear or reduce until all metals have been removed. Hg is slowly eliminated and damage is slowly repaired or compensated for. During the amalgam debate in the 1930ies it was estimated that a person could be considered mercury-free and unexposed 4-5 years after a mercury exposure.

Dental health corresponds to general health

Further arguments were presented on the web (19/3 2001, www.amalgamskadefonden.se) by a researcher:
"Do they have good exposure information? Is there any evidence for the absence of distorting effects - confounding? Has the study enough strength to exclude negative effects with any degree of certainty? The relevant exposure parameter might be accumulated life-time exposure. The relation between amalgam load at the time of measurement and life-time exposure will be very low. Good dental health in general relates to good general health. Correction for socio-economic factors and age made most correlations disappear and from this the authors draw the conclusion that persons with many amalgam fillings at least did not have an increased number of symptoms. The authors do not appear to realize the extent of the problems with confounding. As long as a confounding factor of unknown magnitude appear to exist, no conclusions can be drawn. Discussion about this are lacking; it seems as if the authors mean that the reverse relations support the conclusion that amalgam is harmless, although these reverse relations actually demonstrate that the study contains weaknesses which exclude any conclusions."

Per Dalén points out that you can compare with the epidemiological research on smoking and lung cancer; 40000 persons and a 10 years follow-up was required to confirm a casual connection which convinced most persons, however, not the tobacco industry. Regarding amalgam, where the exposure situation is much less clear, a study of 1/2 million might be required. "Epidemiological methods which require an enormous material to give scientifically meaningful results are certainly not useful under these circumstances. Despite this, they are still used in order to show that various products which produce mass exposure are harmless..... The most important knowledge base is simply the many case reports which show that symptoms of long duration improve when amalgam has been removed."

References
Ahlqwist M, Halling A & Hollender L Rotational panoramic radiography in epidemiological studies of dental health. Comparison between panoramic radiographs and intraoral full mouth surveys. Sw. Dent J 10 1986 79-84

Halling A & Bengtsson C Dental status of Swedish middle-aged women as found in a population study in Göteborg, Sweden Sw Dent J 5 1981 1-7

Bengtsson C, Edström K, Furunes B, Sigurdsson JA & Tibblin G Prevalence of subjectively experienced symptoms in a population sample of women with special reference to women with arterial hypertension Scand J Prim Health Care 5 1987 155-62

Ahlqwist M, Bengtsson C, Furunes B, Hollender L, Lapidus L Number of amalgam tooth fillings in relation to subjectively experienced symptoms. Results from a population study of women in Gothenburg, Sweden Commun Dent Oral Epidemiol 16 1988 227-31

Lavstedt S & Sundberg H Medicinska diagnoser och sjukdomssymtom relaterade till amalgamfyllningar Tandläkartidn 81 1989 81-8 (Medical diagnoses and symptoms related to amalgam fillings)

Tibblin G, Bengtsson C, Furunes B, Lapidus L Symptoms by age and sex. The population studies of men and women in Gothenburg, Sweden Scand J Prim Health Care 8 1990 9-17

Ahlqwist M, Bengtsson C, Lapidus L Number of amalgam fillings in relation to cardiovascular disease, diabetes, cancer and early death in Swedish women Commun Dent Oral Epidemiol 21 1993 40-44

Ahlqwist M, Bengtsson C, Lapidus L, Lindstedt G, Lissner L Concentrations of blood, serum and urine components in relation to number of amalgam tooth fillings in Swedish women. Commun Dent Oral Epidemiol 23 1995 217-221

Ahlqwist M, Bengtsson C, Lapidus L Epidemiologiska aspekter på amalgamets möjliga påverkan på hälsan Tandläkartidn 86 1994 184-188 (Epidemiological aspects on the possible influence of amalgam on health)

Bergdahl IA; Schütz A; Ahlqwist M; Bengtsson C; Lapidus L; Lissner L; Hulten B Methylmercury and inorganic mercury in serum - Correlation to fish consumption and dental amalgam in a cohort of women born in 1922. Environmental Research; 77 (1) p20-24 APR 1998

Ahlqwist M, Bengtsson C, Hakeberg M, Hägglin C Dental status of women in a 24-year longitudinal and cross-sectional study. Results from a population study of women in Göteborg. Acta Odontol Scand 1999 57(3), 162-167

Ahlqwist M; Bengtsson C; Lapidus L; Gergdahl IA Schütz A Serum mercury concentration in relation to survival, symptoms, and diseases: results from the prospective population study of women in Gothenburg, Sweden. Acta Odontol Scand 1999 Jun;57(3):168-74

Bengtsson C, Ahlqwist M, Bergdahl IA, Lapidus L, Schütz A Inget samband mellan antal amalgamfyllningar och hälsa. Epidemiologiska erfarenheter från populationsstudie av kvinnor i Göteborg. Läkartidn 98 2001 930-3 (No connection between number of amalgam fillings and health. Epidemiological experience from a population study of women in Göteborg)

Increased blood mercury levels in patients with Alzheimer's disease. Hock C; Drasch G; Golombowski S; Mullerspahn F; Willershausenzonnchen B; Schwarz P; Hock U; Growdon JH; Nitsch RM Journal of Neural Transmission; 105 (1) p59-68 1998

Print article!