Dental Fillings (Amalgams)

Back to Videos

Share this with your friends


Since the early nineteenth century, dentistry has relied mainly on amalgam (approximately 50% metallic mercury (Hg)) for filling teeth. Scientific research has shown that Hg is constantly released from amalgams, mainly as Hg vapour (Hg),which is inhaled, absorbed, metabolized to ionic Hg (Hg2+) and distributed throughout the body, mainly bound to proteins. Dental amalgam is the major source of the body Hg burden. Toxicological research on amalgam Hg has indicated deleterious effects on the immune, renal, reproductive and central nervous systems, and oral and intestinal bacteria. Results do not indicate that amalgam fillings are safe. Oral DL-2,3-dimercapto-succinic acid, magnesium salt (DMSA); 2,3-dimercapto-l-propane-sulphonic acid, sodium salt (DMPS); N-acetyl-L-cysteine (NAC) and potassium citrate B.P. (K Cit.) were studied for Hg chelating ability in patients who had, or until recently had, amalgam fillings. Based on the increase in urinary Hg concentrations after single doses, compared with controls, the order of efficacy was: DMPS plus K Cit., NAC plus K Cit. and DMSA (each producing an increase of 163%), then in descending order, DMSA plus K Cit., DMPS, NAC and K Cit. Very significant (p < 0.01) correlations were demonstrated between post-chelation urinary and post-chelation sweat Hg concentrations with alt agents. Both these parameters may be good indicators of total body Hg burden. The advantages of employing combined chelating agents were examined and some clinically useful and convenient methods of assessing and reducing Hg burdens suggested.
Keywords: dental amalgam, amalgam mercury, mercury vapour, urine and sweat mercury provocation tests, oral chelating agents, DMSA, DMPS, NAC, K Cit.,


Background to Amalgam Usage and Toxicity
Dental amalgam contains approximately 50% mercury (Hg) with variable amounts of silver (Ag), tin (Sn), copper (Cu) and sometimes zinc (Zn). It hardens rapidly, becoming a solid solution, not a chemical compound. Over many years, it was erroneously taught in dental schools throughout the world that amalgam was a stable alloy which did not release Hg in the mouth. However, in an average subject with eight occlusal amalgams, it has been calculated that approximately 120 Mug of Hg are released into the mouth per day owing to mechanical wear, vaporization and dissolution [ 1, 2]. The amount of Hg absorbed into the body in this situation is estimated to be 10 Mug (range 3-17 Mug) per day [ 3, 4]. All other sources (food, water, air) total approximately 2.6 Mug/day of absorbed Hg [ 4, 5]. It is established that mercury vapour (Hgo) is continuously released from amalgam fillings [ 6-9]. The rate of release is enhanced by chewing [ 2, 6, 7, 9], tooth brushing [ 8], amalgam polishing [ 10], and after hot drinks [ 11]. Intraoral Hgo levels correlate, before and after chewing, with the number and type of amalgam fillings [ 2, 7], plasma Hg levels correlate with the number and total surface area of amalgams [ 12], and Hg levels in human milk [ 13, 14], faeces and urine [ 15] correlate with the number of amalgams.

The evidence suggests that dental amalgams constitute the major source of Hg exposure in the general population [ 4, 5, 16]. Human autopsy studies demonstrate significantly higher Hg levels in the brains and kidneys of subjects with dental amalgams [ 17]. A specific 'no observed effect level' cannot be established for Hgo [ 4]. This is probably the most important form determining human exposure to amalgam fillings. The highly lipid soluble vapour enters the blood from the lungs and oral mucous membranes, traverses cell membranes, including the blood-brain barrier and placenta, rapidly partitions between plasma and red blood cells and becomes widely distributed. Intracellular oxidation of Hgo by the catalasehydrogen peroxide complex forms the divalent ion, Hg2+, the proximate reactive toxic species, which combines covalently with nearby sulphhydryl groups (--SH), in, for example, haemoglobin, reduced glutathione (GSH), protein cysteine groups, etc. thus causing enzyme and cellular dysfunction. Hgo dissolved in saliva may also be oxidized to Hg2+ and be swallowed and partially absorbed in the gastrointestinal tract. The urine and faeces are the major routes of excretion for the covalent compounds of Hg2+ arising as above [ 18-22]. Additionally, small amounts of methyl-Hg (Hg+) are produced in the mouth from amalgam Hgo, possibly via Hg2+ [ 23], and can bind to --SH groups.

Body tissues have various retention half-lives for Hg2+ and organic (methyl or ethyl) Hg, (Hg +), ranging from days to years. Half-lives may differ with chronic exposure as a result of compromised cellular function (e.g. kidney Hg turnover decreases with age and duration of exposure)[ 5, 21, 22]. Because blood and urine Hg levels remain relatively low during Hg exposure from dental amalgam, they are poor diagnostic indicators of the very high Hg levels which accumulate in some body tissues as a result of such exposure [ 21, 22]. Furthermore, there is poor correlation between the urinary excretion of Hg and the occurrence of demonstrable evidence of poisoning, and, in some cases, failure to excrete Hg is a factor in the development of poisoning [ 24, 25].

The choroid plexus, an important part of the blood-brain barrier, acts as a sink for Hg and other toxic metals [ 26, 27]. Additionally, it has been shown that Hg is selectively concentrated in human brain regions (medial basal nucleus, amygdala and hippocampus) involved with memory function, and it has been postulated that Hg may be implicated in the aetiology of Alzheimer's disease (AD) [ 28, 29]. Further work has suggested a connection between exposure to inorganic Hg (Hg2+) and AD [ 30, 31, 32].

Also giving rise to concern is the finding that Hg released by dental amalgam can enhance the prevalence of resistance to multiple antibiotics in the bacteria of the primate normal flora [ 33].
Ongoing research on the pathophysiological effects of amalgam Hg has focused on the immune system, the renal system, oral and intestinal bacteria, the reproductive system and the central nervous system. Research evidence does not support a belief in the safety of amalgam [ 22].

Use of Oral Chelating Agents in Diagnosis and Treatment of Hg Overload

Therapeutic chelating agents are compounds which can be administered to a patient for the reduction of toxic metals (or their compounds). They form metal complexes with these toxic metal ions in vivo, which are readily excreted in the urine, faeces, etc. so reducing body levels of the metals to a less dangerous range [ 34].

A chelating agent, by definition, forms a ring compound with a metal [ 35]. All the therapeutic compounds discussed below may not form ring compounds with the particular Hg with which they are reacting; hence, the term complexing agent may be a better term. However, the term chelating agent will continue to be used here in the broad sense.
The chelating agents evaluated in this study are: DL-2,3-dimercapto-succinic acid, magnesium salt (DMSA); 2,3-dimercapto-l-propane-sulphonic acid, sodium salt (DMPS), N-acetyl-L-cysteine (NAC), and potassium citrate B.P. (K Cit.).

DMSA, DMPS [ 34-37] and NAC [ 34, 38] have all been employed in the in vivo chelation of Hg. Because of the ease with which the --Hg--S--linkage is formed with --SH containing compounds, these three, but particularly DMSA and DMPS, can be used with advantage for chelation of inorganic (Hg2+) and organic (Hg+) Hg [ 34]. K Cit., a salt of citric acid, can be used to alkalinize the urine to reduce kidney damage owing to dimercaprol chelated Hg [ 39], and citric acid can form complexes with traces of heavy metals [ 40]. Additionally, the urinary excretion of the metal aluminium (Al) can be enhanced by citric acid [ 34].

DMPS has been shown to be capable of inducing Hg urinary excretion which was directly proportional to the total body Hg burden in rats given mercuric chloride (HgCl2) or Hgo [ 44], and thus DMPS could prove to be a useful biological monitoring or diagnostic agent for total body burden of inorganic Hg [ 41-43]. Additionally, after treatment with DMPS, urinary excretion of Hg is significantly greater in human subjects with amalgams than in similarly treated subjects without amalgams [ 42, 44]. At least two-thirds of this excretable Hg in the urine of those with amalgams appears to be derived from the amalgam and amounts of Hg excreted correlated with total amalgam surface area [ 44].

In animal experiments DMSA has been found to be more effective than DMPS for removal of Hg from the body and brain [ 35], with the exception of the kidney, for which DMPS was more effective [ 35, 41]. DMSA removed more organic Hg whereas DMPS removed more inorganic Hg. A combination of DMSA and DMPS removed Hg from most organs [ 35].

Safety of the Agents being Studied

DMSA and DMPS: these have been in use as effective chelating agents particularly for Hg, lead (Pb) and arsenic (As) for many years in Russia (and the USSR), China, Japan, Germany and the US. They are used orally and parenterally, DMPS more by the latter route than DMSA [ 34-37, 41-44]. Therapeutically effective doses of the agents have proved to be well tolerated, relatively non-toxic and with a wide therapeutic index. They are readily excreted by the kidneys with no nephrotoxicity being reported [ 35-37, 41, 45, 46]. They can increase the excretion of Cu, and that of Zn somewhat, but other trace elements do not appear to be significantly affected when therapeutically reasonable doses are employed [ 35].

NAC: the following side effects have been reported after the use of NA as a mucolytic, by direct instillation or nubilization, of up to 20% solutions, through a face mask or mouthpiece, from one to six hourly, in cases of cystic fibrosis of the pancreas and other affections where mucolytic therapy is required: bronchospasm, nausea, vomiting, stomatitis and rhinorrhoea. It should be used with caution in elderly patients with severe respiratory insufficiency or in asthmatic patients [ 47].

K Cit.: Potassium salts should be given cautiously to patients with renal or adrenal insufficiency, acute dehydration or heat cramp [ 48].

With all of the above background in mind, it was thought desirable to investigate possible ways of assessing human body Hg load, and methods of removing it, with as little inconvenience to the patient as reasonably possible.

Much more has been reported about the parenteral route for chelation of mercury than the oral route. It was therefore decided, in the human situation,
( 1) Compare the abilities of oral DMSA, DMPS and NAC to increase the urinary excretion of Hg, from whatever source (amalgams, diet, environment--amalgams being the major source) [ 4, 5, 16].
( 2) Determine at what time after oral dosing the Hg excretion rate is maximal with the above agents.
( 3) Investigate the possible ability of oral K Cit. to increase urinary Hg excretion.

( 4) Evaluate the effect of oral K Cit., given at the same time as each of the other agents, on urinary Hg excretion.

( 5) Decide from the data how best to apply a convenient oral test, in the clinical situation, for the better diagnosis of those patients who may be in need of chelation treatment for an abnormal body Hg burden.

( 6) Gain useful clinical information from the overall results which could lead to effective oral chelation courses being employed to reduce abnormal body Hg burdens as efficiently as possible.

It was also decided to measure sweat Hg concentrations, following the dosing with each chelating agent, and the combinations with K Cit. It was considered worthwhile to determine whether any relationship existed between post-chelation sweat Hg concentrations and urinary Hg concentrations. Pre- and post-chelation sweat tests would have been preferred but, owing to the cost (patients were paying for themselves) and the need for the patients to be at the laboratory for an hour for each sweat test, only the post-chelation sweat data were to be obtained in as many cases as possible at this stage.

The correlations between the number of amalgams with plasma, breast milk, faeces and urine Hg concentrations have been well recorded [ 12, 13, 14, 15] and were not intended to be re-examined in this study for urine or sweat. However, patients' amalgam status has been recorded.


A total of 191 polysymptomatic patients with amalgam fillings (74.6%) and some who had, until recently, had amalgam fillings (25.4%) were admitted to this study. The latter group had not had chelation following amalgam removal to reduce the residual heavy metal load. Haematology/serum biochemistry screens via the Biolab Medical Unit and the London Clinic, London W1 were obtained for all patients considered for the study. Patients with medical histories of cardiac, renal, hepatic or psychiatric disease, those taking medically prescribed xenobiotic treatment, or those with clearly abnormal haematology or serum biochemistry results were excluded from the study.

Of the patients, 12 (6.3%) failed to complete the required procedures, leaving 179 who completed the requirements. Ages ranged from 18 to 82 years, with an average of 43.4 years. The group comprised 106 women (59.2%), average age 42.9 years, and 73 men (40.8%), average age 44.2 years.

NAC as 500 mg powder in gelatin capsules, and DMSA as 500 mg powder in gelatin capsules, were obtained from Biocare Ltd, Lakeside, 180 Lifford Lane, Kings Norton, Birmingham B30 3NT, UK. DMPS as a 5% (w/v) sterile solution in 5-mi ampoules (Unithiol) were manufactured by the Chemical Pharmaceutical Department, Ministry of Industrial Medicine, St Petersburgh, Russia, and kindly supplied by Professor Dr Med. Dent. Gerd S. Hausmann, Munich, Germany. K Cit. was obtained from Thornton & Ross, Huddersfield, UK.

Determination of Time of Maximal Urinary Hg Concentration (Mug 1[sup-1])

Six patients (three male and three female), randomly selected from a group of 18, were used for each chelating agent. The average age was 43.2 years with a range of 25 to 64 years. They were not taking any medication or other compounds which could have affected urinary Hg excretion and had no diagnosed medical conditions. When the same patients were used for more than one chelating agent, a period of at least seven days separated the tests.

Early morning control urines were collected in 50-mi sterile Universal containers. The patients were instructed to empty their bladders completely after the sample collection. Each group of six then immediately received either DMSA (30 mg kg[sup-1] in capsules), DMPS (10 mg kg-1 in solution from ampoules diluted with sterile water to 100 ml), NAC (30 mg kg-1 in capsules) or K Cit. (5 g dissolved in 200 ml of sterile water) orally, and test urine samples were collected, as above, at hourly intervals for up to five hours. Immediately after collection all samples were labelled, refrigerated and analyzed within 24 h of collection. The time of maximum urinary Hg concentration was observed from the means of the results of Hg concentration vs. time (Fig. I).

Evaluation of the Individual Chelating Agents, and with K Cit., as Shown by Urinary and Sweat Hg Excretion
Having determined the time of maximal urinary Hg excretion following oral dosage with each chelating agent, patients, immediately after collection of an early morning control urine, were given, orally, either DMSA (30 mg kg-1); DMPS (10 mg kg-1); NAC (30 mg kg-1); and/or K Cit. (5 g). These were given on a random basis. Some results, however, had been collected during preliminary work using DMSA, prior to the use of the other agents in the study. This made the DMSA group larger than the other groups. Test urines were collected at the laboratory, 3 h after dosing with DMSA, NAC, K Cit., and when the first two agents were given with K Cit., and at 2 h after DMPS alone and with K Cit. All urine samples were refrigerated immediately after collection and analysed within 24 h.

Sweat patches were applied when the patient reported at the laboratory, immediately following the collection of the test urine samples. Patches were removed after 1 h and immediately analyzed for mercury.


The procedure for the collection of sweat samples for trace element analysis was as previously described [ 49].

The technique for the measurement of total Hg in all samples was atomic absorption spectrophotometry with cold vapour hydride generation [ 50]. The initial measurements were carried out on a Pye Unicam PU9000 atomic absorption spectrophotometer (ATI Unicam Ltd, York Street, Cambridge CB 1 2PX, UK), and later measurements on a Hitachi Z-8200

Polarized Zeeman atomic absorption

spectrophotometer (Hitachi Scientific Instruments, Hogwood Industrial Estate, Finchampstead, Wokingham, Berkshire RG11 4QQ, UK). Both were attached to a Pye Unicam PU9060 continuous flow vapour system. Analyzer settings were as suggested in the manufacturer's recommendations and analytical reference book. All reagents were of analytical grade (Merk (UK) Ltd, Hunter Boulevard, Magna Park, Lutterworth, Leicestershire LE17 4XN). Inter- and intra-batch precision, measured using Nycomed QC material (Nycomed (UK) Ltd, Nycomed House, 2111 Coventry Road, Sheldon, Birmingham B26 3EA) was <10% in both cases.

Statistical Methods Employed

The Mann Whitney U-Test was used to analyze differences between the pre-and post-chelation urinary Hg concentration data for each of the agents employed. Correlation coefficients were obtained using Microsoft Excel for Windows 95 (Microsoft Ltd, Microsoft Place, Winnersh, Wokingham, Berkshire RG11 5TP, UK).


The single dose studies showed that peak urinary Hg excretion times were as follows: DMPS and K Cit., at 2 h; DMSA and NAC, at 3 h (see Fig. 1).

The details of the groups given the various agents are as follows (see Fig. 2 and Table 1).

K Cit.; Urinary tests; 18 patients, 10 men and eight women; age, range 20-65, average (av.) 45.6 years; av. number (no.) of amalgams = 6.8. Of these patients, 12, five men and seven women, also had post-chelation sweat tests at the same time as the test urine sample; age, range 27-65, av. 41.3 years, av. no. of amalgams = 7.0.

DMPS; Urinary tests; 20 patients, nine men and 11 women; age, range 27-76, av. 44.7 years; av. no. of amalgams = 9.6. Of these patients, 12, four men and eight women, also had post-chelation sweat tests, as above; age, range 27-76, av. 42 years; av. no. of amalgams = 9.8.

DMPS plus K Cit; Urinary tests; 16 patients, nine men and seven women; age, range 25-62, av. 43.1 years; av. no. of amalgams = 8.8. Of these patients, 13, six men and seven women, also had post-chelation sweat tests, as above; age, range 25-62, av. 44.3 years; av. no. of amalgams = 9.6.

NAC; Urinary tests; 22 patients, seven men and 15 women; age, range 24-75, av. 43.2 years; av. no. of amalgams = 7.0. Of these patients, 14, four men and 10 women, also had sweat tests, as above; age, range 24--75, av. 43.6 years; ay. no. of amalgams = 6.9.

NAC plus K Cit.; Urinary tests; 16 patients, six men and 10 women; age, range 26-60, av. 41.2 years; av. no. of amalgams = 9.4. Of these patients, 14, five men and nine women, also had post-chelation sweat tests, as above; age range 26-60, av. 41.7 years; av. no. of amalgams = 9.7,

DMSA; Urinary tests; 65 patients, 24 men and 41 women; age, range 18-82, av. 42.7 years; av. no. of amalgams = 8.0. Of these patients, 42, 17 men and 25 women, also had post-chelation sweat tests, as above; age, range, 18-70, av. 41.2 years; av. no. of amalgams = 8.5.

DMSA plus K Cit.; Urinary tests; 22 patients, eight men and 14 women; age, range 23-66, av. 44.9 years; av. no. of amalgams = 8.0. Of these patients, 18, eight men and 10 women, also had post-chelation sweat tests, as above; age, range, 23-66, av. 45.3 years; av. no. of amalgams = 8.2.

Mean urinary and sweat test Hg concentrations, as well as statistical data, are set out in Fig. 2 (urinary data) and Table 1 (urinary and sweat data).
Fig. 2 shows that the mean pre (control) and post (test) chelation urinary Hg concentrations (Mug-1), for each agent and combination studied, at the previously determined peak times (Hg concentration using K Cit. was determined at 3 h), exhibited highly significant differences (p < 0.01). Each agent and combination showed a positive ability to increase mean urinary Hg concentration. DMPS plus K Cit. (sets of results, n = 16), NAC plus K Cit. (n = 16), and DMSA (n = 65) were the most effective, showing a mean increase in urinary Hg concentration of 163% in each case; in descending order of increase these were DMSA plus K Cit. (n= 22), 146%; DMPS (n= 20), 135%; NAC (n = 22), 131%, and K Cit. (n = 18), 83%.

Table 1 shows that, in the case of each chelating agent and combination, the mean post-chelation urinary Hg concentration was seen to be highly significantly (p<0.01) increased compared with its control value.

With control urinary Hg concentration (Mug 1-1) vs. post-chelation sweat Hg concentration (ppb) (the sweat test having been started at the same time as the subsequent test (post-chelation) urine sample was taken), results show that there were significant correlations between the above parameters, only with K Cit. (p < 0.03) and DMPS (p <0.02).

Post-chelation test urinary Hg concentrations (Mug 1-1) vs. post-chelation sweat Hg concentrations (ppb) showed highly significant correlations (p <0.01) with all the agents and combinations employed.


Although it had been established that the peak urinary Hg excretion for K Cit. was at 2 h (Fig. 1), it was decided to give it at the same time as each of the other three agents being studied. This allowed evaluation of the joint combinations together, over the same times, which was considered desirable for proper evaluation of possible true synergism. The results (Fig. 2) show K Cit. to be a useful agent in increasing urinary Hg excretion and that it improves urinary Hg excretion produced by DMPS and NAC. The contribution of K Cit. to the latter two agents, however, was not synergistic (i.e. results were not seen to be more than the sum of the effects of the agents used separately), or even fully additive with that of DMPS (determined at 2 h) or NAC, and indeed appeared to reduce the efficacy of DMSA (the latter two determined at 3 h).

It appears from these results that the use of K Cit. separately, rather than with DMSA, may be more effective in the promotion of urinary Hg excretion, and, in the cases of DMPS and NAC, K Cit. may be beneficially used with the latter two agents.

DMPS has been shown to be more effective th0n DMSA in increasing the excretion of kidney Hg [ 35, 41] and K Cit. is known to facilitate the safer passage of chelated Hg through the kidney [ 39], and has been employed as a urinary alkalinizing agent and mild diuretic for decades [ 51]. Thus the fact that K Cit. increased the urinary Hg excretion when combined with DMPS was not surprising, and the improved result when K Cit. was combined with NAC may indicate that the latter agent is also effective in removing kidney-bound Hg. DMSA has been shown to be more effective than DMPS in the removal of Hg from the body (with the exception of the kidney) and brain [ 35]. The reduced Hg excretion seen after 3 hours with DMSA plus K Cit., as compared with DMSA alone, is something which will be further investigated.

As shown in Table 1, the highly significant (p <0.01) beneficial increases in urinary Hg concentrations, compared with control results, for each agent and combination used demonstrate a range of useful oral methods for reducing body Hg levels. However, DMSA and K Cit. appear to be better employed separately.
Significant correlations between control urine and post-chelation sweat Hg concentrations occurred only with K Cit. (p < 0.03) and DMPS (p < 0.02). The latter chelating agent is reported to exert its major effect on kidney Hg [ 35, 41], which could also apply to K Cit., in light of its properties [ 39, 40, 51]. Both agents, separately, may have less effect on sweat Hg excretion than DMSA, NAC or the combined agents employed, and this may account for the results obtained. Further work with these agents, including: pre- and post-chelation sweat Hg excretion, will be needed before firm conclusions can be drawn.
The highly significant (p < 0.01) correlations for all agents, and combinations, between post-chelation urine and post-chelation sweat Hg concentrations is a very interesting finding. This indicates that both these concentrations appear to be relevant and useful parameters for the evaluation of the efficacy of the oral agents employed to reduce body Hg load. Further, it is suggested, by analogy with the parenteral use of DMPS [ 41, 42, 43], that these results may be directly proportional to the total body Hg load. Additionally, results from Fig. 2 and Table 1 indicate that the oral use of DMPS (10 mg kg-I), with or without K Cit. (5 g) 2 h before test urine collection, or DMSA (30 mg kg-1) alone, or NAC (30 mg kg-1), with or without K Cit. (5 g) 3 h before test urine collection, respectively, would provide 'useful diagnostic urinary Hg provocation tests' for assessment of body Hg burden when compared with control urine Hg concentrations. A sweat test for Hg, started at the same time as the test urine collection, would also provide additional useful data on the total body Hg burden. Thus 'urine plus sweat mercury provocation tests' could provide useful diagnostic information on body Hg load which arises mainly from amalgam fillings [ 4, 5, 16].
Adverse side effects arising from the single dose procedures described above were monitored and reported in a small percentage (approximately 5%) of patients and included mild gastrointestinal discomfort, mild fatigue, mild mental 'fuzziness', mild headache and mild diuresis. These side effects usually cleared within 6 h of the dose and were, in most cases, likely to have been due to heavy metal mobilization. No case of hypersensitivity to any of the agents used was seen.


Various oral methods, using the agents studied above, have been employed with good results, as measured by reduction of Hg levels in follow-up laboratory testing and clinical improvement in patients' Hg-related symptoms [ 52]. Broadly, dosage, length and frequency of courses, as well as duration of treatment, are dictated by the medical condition of the patient, including the severity of Hg-related symptoms, size of the body Hg load, and results of regular haematology, serum biochemistry and appropriate Hg provocation tests, as described above. The following, therefore, will act as a general guide only:

( 1) Oral DMSA (from 2 to 30 mg kg-1 daily, divided into three doses) for 2 to 7 days. Repeat twice with breaks of 3 weeks, then reassess and repeat as considered necessary to deal with the Hg load. High fluid intake (2.5 to 3 1/day) is advisable while on DMSA. K Cit., 1.5 to 3 g (adult dose) in a glass of water three times a day after food, for 2 to 7 days between the DMSA courses may be advisable to aid the clearance of heavy metal chelates through the kidney [ 39], particularly where the renal history has not been normal. Dosage, length and number of courses, of each agent, must be decided on the basis of laboratory tests, the patient's clinical condition, and tolerance of the symptoms produced because of the mercury mobilization. If hypersensitivity is suspected (e.g. skin rash, etc.) the DMSA should be stopped.
The periodic laboratory tests should also include data on the patient's trace elements status (e.g. Zn, Cu, Fe, Chromium (Cr), etc.).

( 2) Oral DMPS (2 to 10 mg kg-1 daily, as one dose in water) for 2 to 7 days. Then, as above, including monitoring, except K Cit. (dosage as above) can be used during the DMPS courses, and will be beneficial in increasing Hg excretion.

( 3) Oral NAC (2 to 30 mg kg-1 daily, divided into three doses) for 2 to 7 days. Then, as above, including monitoring. K Cit., as in ( 2), will be beneficial in increasing Hg excretion.

( 4) K Cit. (dosage as before) may also be used in courses of up to 7 days, between courses of DMPS or NAC, to improve Hg clearance. Serum potassium levels should be monitored with repeated K Cit. courses. In some cases other citrate salts, such as sodium citrate, may be judged to be more appropriate clinically.

It must be appreciated that the successful removal of body Hg burden requires sustained treatment along the above lines, often for several months. Hg binds very tenaciously to many tissue nucleophilic sites and it requires the persistent spaced application of external competing Hg binding ligands to displace the endogenously bound heavy metal. A satisfactory result is achieved when urinary Hg concentration is not more than 1 to 2 Mug 1-1, and does not rise to more than about 3 to 4 Mug 1-1 after at least two urinary Hg provocation tests, using oral DMSA (30 mg kg-1), at intervals of 2 to 3 weeks.

Retesting after the load appears to have been sufficiently reduced is advisable after 3 months. If Hg seepage is seen, from the results of urinary provocation tests, introduce 0.5 to 1 g DMSA, orally daily, on two to three consecutive days a month for three months, then reassess. Repeat this treatment if necessary, and continue with laboratory tests every 3 months for at least two tests until the results are satisfactory. Chelation can then be stopped. Following this it is advisable to carry out laboratory monitoring at 6-month intervals, as further seeping of Hg may occur, and must be dealt with for best results to ensue.
Any increase in body heavy metal load (particularly Pb and Hg) owing to environmental and/or dietary sources will also be controlled by the above procedure.


The chelating agents studied were clinically effective orally for increasing the urinary excretion of Hg. When K Cit. was given with DMPS or NAC the urinary Hg concentrations were further increased. Results showed that DMSA and K Cit. are better employed separately to reduce body Hg. Both the post-chelation urinary and sweat Hg concentrations appear to be useful parameters in the clinical assessment of body Hg burdens in patients who have been exposed to Hg from dental amalgam, and other sources. All the agents studied can be employed orally, as discussed above, for the clinical assessment and reduction of raised body mercury burdens in a way that is convenient to the patient.

This study has indicated that further work to investigate pre and post oral chelation sweat Hg concentrations, and to correlate these, in each patient, with pre and post oral chelation urine Hg concentrations, should be done.


The authors wish to thank Professor Dr Med. Dent. Gerd S. Hausmann, Munich, Germany for donating an initial supply of Unithiol (DMPS), and, the Biolab Research and Benevolent Fund (London, UK) for carrying out some of the mine Hg analyses free of charge. We also wish to thank Dr Damien Downing, senior editor, and Dr Stephen Davies, consulting editor, Journal of Nutritional & Environmental Medicine, for helpful discussions concerning the paper.

TABLE 1. Control and post-chelation Hg levels in sweat (Mu g kg-1) and urine (Mu l-1) with different chelating agents
Legend for Table:

A - n
B - Pre-chelation Mean (SD)
C - Post-chelation Mean (SD)
D - p Value
E - Sweat vs. Pre-chelation urine Hg
F - Sweat vs. post-chelation urine Hg

Sweat Hg


Potassium citrate 12 --[a] 4.58 (1.4)[b]
DMPS 12 -- 5.50 (1.7)[c]
DMPS +Potassium citrate 13 -- 6.31 (3.2)[c]
N-Acetyl-L-cysteine 14 -- 5.00 (1.7)[b]
N-Acetyl-L-cysteine +
potassium citrate 14 -- 5.50 (2.8)[b]
DMSA 42 -- 6.52 (4.2)[b]
DMSA + Potassium citrate 18 -- 5.78 (1.9)[b]

Urine Hg


Potassium citrate 18 5.25 (1.6) 9.56 (3.1)[d] <0.01
DMPS 20 5.05 (1.7) 11.88 (6.7[e] <0.01
citrate 16 5.25 (2.0) 13.8.(6.3)[e] <0.01
L-cysteine 22 4.52 (2.1) 10,43 (4.6)[d] <0.01
L-cysteine +
potassium citrate 16 5.13 (2.7) 13.50 (9.6)[d] <0.01
DMSA 65 4.98 (3.5) 13.11 (9.0)[d] <0.01
Potassium citrate 22 5.68 (1.6) 13.95 (5.4)[d] <0.01



Potassium citrate 0.62 (<0.03) 0.62 (<0.01)
DMPS 0.70 (<0.02) 0.62 (<0.01)
DMPS +Potassium citrate -0.02 (>0.62) 0.81 (<0.01)
N-Acetyl-L-cysteine 0.12 (>0.33) 0.70 (<0.01)
N-Acetyl-L-cysteine +
potassium citrate 0.29 (>0.31) 0.86 (<0.01)
DMSA 0.29 (>0.05) 0.56 (<0.01)
DMSA + Potassium citrate 0.19 (>0.45) 0.77 (<0.01)

[a] = No data, [b] Test started at 3 h after oral chelating
dose. [c] Test started at 2 h after oral chelating dose.
[d] Results at 3 h after oral chelating dose. [e] Results at
2 h after oral chelating dose.

r = correlation coefficient.
FIG. 1. Single dose studies to determine times of peak urinary Hg excretion.
FIG. 2. Mean urinary Hg concentrations (Mu 1-1), pre- and post-chelating agent. Chelation period was 3 h except for DMPS and DMPS plus K Cit. when it was 2 h (control; post-chelation).
[1] Gross MJ, Harrison JA. Some electrochemical features of the in vivo corrosion of dental amalgams. Appl Electrochem 1989; 19: 301-10.
[2] Vimy M J, Lorscheider FL. Serial measurements of intraoral air mercury: estimation of daily dose from dental amalgam. J Dent Res 1985; 65: 1072-75.
[3] Vimy MJ, Lorscheider FL. Dental amalgam mercury daily dose estimated from intraoral vapor measurements: a predictor of mercury accumulation in human tissues. J Trace Elem Exp Med 1990; 3:111-23
[4] World Health Organization. International Programme on chemical safety. Inorganic Mercury. In: Environmental Health Criteria 118. Geneva: WHO, 1991; 36.
[5] Clarkson TW, Hursh JB, Sager PR, Syversen TLM. Mercury. In: Clarkson TW, Friberg L, Nordberg GF, Sager PR, eds. Biological Monitoring of Toxic Metals. New York: Plenum, 1988; 199-246.
[6] Svare CW, Peterson LC, Reinhardt JW, Boyer DB, Frank CW, Gay DD, Cox RD. The effects of dental amalgams on mercury levels in expired air. J Dent Res 1981; 60: 1668-71.
[7] Vimy MJ, Lorscheider FL. Intraoral air mercury released from dental amalgam. J Dent Res 1985; 64: 1069-71.
[8] Patterson JE, Weissberg B, Dennison PJ. Mercury in human breath from dental amalgam. Bull Environ Contam Toxicol 1985; 34: 459-68.
[9] Aronsson AM, Lind B, Nylander M, Nordberg M. Dental amalgam and mercury. Biol Metals 1989; 2:25-30
[10] Hibberd AR. Personal Clinical Observation. 1994.
[11] Anthony It, Birtwistle S, Eaton K, Maberly J. eds. Environmental Medicine in Clinical Practice. Southampton: BSAENM Publications, 1997; 204-8.
[12] Molin M, Bergman B, Marklund SL, Schutz A, Skerfving S. Mercury, selenium and glutathione peroxidase before and after amalgam removal in man. Acta Odontol Scand 1990; 48:189-202.
[13] Oskarsson A, Schultz A, Skerfving S, Hallen IP, Ohlin B, Lagerkvist BJ. Total and inorganic mercury in breast milk in relation to fish consumption and amalgam in lactating women. Arch Environ Health 1996; 51: 234-41.
[14] Vimy M J, Hooper DE, King WW, Lorscheider FL. Mercury from maternal "silver" tooth fillings in sheep and human breast milk: a source of neonatal exposure. Biological Trace Element Res 1997; 56: 143-52.
[15] Skate I, Engqvist A. Human exposure to mercury and silver released from dental amalgam restorations. Arch Environ Hlth 1994; 49: 384-94.
[16] Clarkson TW, Friberg L, Hursh JB, Nylander M. The prediction of intake of mercury vapor from amalgams. In: Clarkson TW, Friberg L, Nordberg GF, Sager PR, eds. Biological Monitoring of Toxic Metals. New York: Plenum, 1988, 247-60.
[17] Nylander M, Friberg L, Lind B. Mercury concentrations in the human brain and kidneys in relation to exposure from dental amalgam fillings. Swed Dent J 1987; 11: 179-87:
[18] Horsted-Bindslev P. Mercury and the mental patient. In: Horsted-Bindslev P, Magos L, Holmstrup P, Arenholt-Bindslev D, eds. Dental Amalgam--A Health Hazard? Copenhagen: Munksgaard, 1991; 41-61.
[19] World Health Organization. International Programme on Chemical Safety. Inorganic Mercury. In: Environmental Health Criteria 118. Geneva: WHO, 1991; 47-64.
[20] Magos L. Mercury metabolism and toxicology. In: Horsted-Bindslev P, Magos L, Holmstrup P, Arenholt-Bindslev D, eds. Dental Amalgam--A Health ttazard? Copenhagen: Munksgaard, 1991; 11-32.
[21] Goering PL, Galloway WD, Clarkson TW, Lorscheider FL, Berlin M, Rowland AS. Symposium overview, toxicity assessment of mercury vapour from dental amalgams. Fundamental & Appl Toxicol 1992; 19: 319-29.
[22] Lorscheider FL, Vimy MJ, Summers AO, Mercury exposure from "silver" tooth fillings: Emerging evidence questions a traditional dental paradigm. FASEB J 1995; 9: 504-8.
[23] Sellars WA, Sellars R, Liang L, Hefley JD. Methyl mercury in dental amalgams in the human mouth. J Nutr Environ Med 1996; 6: 33-6.
[24] Goldwater LJ. The Toxicology of Inorganic Mercury. Annals New York Acad Sci 1957; 65: 498-503.
[25] Goldwater LJ, Ladd AC, Jacobs MB. Absorption and excretion of mercury in man; significance of mercury in blood. Arch Envir Health 1964; 9: 835-41.
[26] Friedheim E, Corvi C, Graziano J, Donnelli T, Breslin D. Choroid plexus as protective sink for heavy metals? Lancet 1983; 1: 981-2.
[27] Zheng W, Perry DF, Nelson DL, Aposhian IIV. Choroid plexus protects cerebrospinal fluid against toxic metals. FASEB J 1991; 5: 2188-89.
[28] Thompson CM, Markesbery WR, Ehmann WD, Mao YX, Vance DE. Regional brain trace element studies in Alzheimers disease. Neurotoxicology 1988; 9: 1-7.
[29] Wenstrup D, Ehmann WD, Markesbery WR. Trace element imbalances in isolated sub cellular fractions of Alzheimers disease brains. Brain Res 1990; 533: 125-31.
[30] Khatoon S, Campbell SR, Italey BE, Slevin JT. Aberrant guanosine triphosphate-B-tubulin interaction in Alzheimers disease. Ann Neurol 1989; 26: 210-15.
[31] Duhr E, Pendergrass C, Kasarskis E, Slevin J, Haley B. Hg2+ Induces GTP-tubulin interactions in rat brain similar to those observed in Alzheimers disease. FASEB J 1991, 5: A456.
[32] Duhr EF, Pendergrass JC, Slevin JT, Haley BE. HgEDTA complex inhibits GTP interactions with the E-site of brain B-tubulin. Toxicol Appl Pharmacol 1993; 122: 273-80.
[33] Summers AO, Wireman J, Vimy M J, Lorscheider FL, Marshall B, Levy SB, Bennett S, Billard L. Mercury released from dental "silver" fillings provokes an increase in mercury and antibiotic resistant bacteria in oral and intestinal floras of primates. Antimicrob Agents & Chemother 1993; 37: 825-34.
[34] Jones MM. New developments in therapeutic chelating agents as antidotes for metal poisoning. Critical Revs in Toxicol 1991; 21: 209-33.
[35] Aposhian HV. DMSA and DMPS-water soluble antidotes for heavy metal poisoning. Ann Rev Pharmacol Toxicol 1983; 23: 193-215.
[36] Graziano JH. Role of 2,3-Dimercaptosuccinic acid in the treatment of heavy metal poisoning. Med Toxicol 1986; 1: 155-62.
[37] Friedheim E, Corvi C. Meso-DMSA, A chelating agent for the treatment of mercury poisoning. J Pharm Pharmac 1975; 27: 624-26.
[38] Lund ME, Clarkson TW, Berlin M. Treatment of acute methyl mercury ingestion by haemodialysis with N-acetyi-cysteine (Mucomyst) infusion and 2,3-dimercaptopropane sulphonate. J Toxicol Clin Toxicol 1984; 22: 31.
[39] Wade A, Reynolds JEF, eds. Martindale, The Extra Pharmacopoeia, 27th edn. London: Council of the Royal Pharmaceutical Society of Great Britain, 1977; 331.
[40] Wade A, Reynolds JEF, eds. Martindale, The Extra Pharmacopoeia, 27th edn. London: Council of the Royal Pharmaceutical Society of Great Britain, 1977; 740.
[41] Cherian MG, Miles EF, Clarkson TW, Cox C. Estimation of mercury burdens in rats by chelation with dimercaptopropane sulphonate. J Pharmacol Exp Ther 1988; 245: 479-83.
[42] Godfrey M, Campbell N. Investigation of sodium 2,3-dimercapto-propane-l-sulphonate (DMPS) as a diagnostic test to confirm chronic accumulation of mercury in humans. J Adv Med 1994; 7:19-30.
[43] Daunderer M. Mobilisation test for environmental metal poisoning. Forum des Praktischen und Aligemendn-Arztes 1989; 28: 88.
[44] Aposhian HV, Bruce DC, Alter W, Dart RC, Hurlbut KM, Aposhian MM. Urinary mercury after administration of DMPS: correlation with dental amalgam score. FASEB J 1992; 6: 2472-76.
[45] Graziano JH, Siris ES, LoIacono NJ, Silverberg SJ, Turgeon L. 2,3-Dimercaptosuccinic acid as an antidote for lead intoxication. Clin Pharmacol Therapeut, St.Louis 1985; 37: 431-8.
[46] Graziano JH, Lolacono NJ, Meyer P. Dose-response study of oral 2,3-dimercaptosuccinic acid in children with elevated blood lead concentrations. J Pediatrics, St. Louis 1988; 113: 751-7.
[47] Wade A, Reynolds JEF, eds. Martindale, The Extra Pharmacopoeia, 27th edn. London: Council of the Royal Pharmaceutical Society of Great Britain, 1977; 573.
[48] Wade A, Reynolds JEF, eds. Martindale, The Extra Pharmacopoeia, 27th edn. London: Council of the Royal Pharmaceutical Society of Great Britain, 1977; 1265.
[49] Howard JMH. Serum, leucocyte, sweat and hair zinc levels; a correlation study. J Nutr Med 1990; 1: 119-26.
[50] Littlejohn D, Fell GS, Ottaway JM. Modified determination of total and inorganic mercury in urine by cold vapour atomic absorption spectrometry. Clin Chem 1976; 22: 1719-23.
[51] Wade A, Reynolds JEF, (Eds) Martindale, The Extra Pharmacopoeia, 27th edn. London: Council of the Royal Pharmaceutical Society of Great Britain, 1977; 1270.
[52] Hibberd AR. A clinical ecology programme for assessment and treatment of suspected amalgam compromised patients. Biomed Newsletter. Birmingham: Biomed Publications. Revised October 1996; 12: 1-4.
Adapted by PHD DCC PHC FPS MRPHARM.S,( n1) and MPHIL PHD( n2)


Sweden has banned mercury amalgam dental fillings, effective January, 1997, after determining that at least 250,000 Swedes have immune and other health disorders directly related to the mercury in their teeth. Denmark will ban amalgams beginning in January 1999.

In 1991, Germany's Health Ministry recommended to the German Dental Association that no further amalgam fillings be placed in children, pregnant women, or people with kidney disease, and in 1993 this was extended to include all women of child-bearing age, pregnant or not. Austria is also phasing out mercury fillings.

By contrast, the American Dental Association (ADA) says replacing amalgam fillings from non-allergic patients for the purpose of removing toxic substances from the body is "improper and unethical." The Canadian Dental Association (CDA) insists that there is no scientific evidence linking medical illness symptoms to mercury fillings, except relatively rare allergic sensitivity to mercury. (The number of persons with a specific and detectable sensitivity to mercury may not be so small. According to a Health Canada report, as many as 15 percent of people with amalgam fillings show signs of sensitivity to mercury. Some American researchers claim that at least 20 percent of people with amalgam fillings are "mercury toxic.")

What gives? Are the Europeans and Scandinavians hysterical Cassandras, in a sweat about nothing, or are the North American dental associations concerned about things other than patient health? Are mercury amalgam tooth fillings dangerous or not?

Amalgam tooth fillings are an alloy of 50 percent mercury, 35 percent silver, 13 percent tin, 2 percent copper, and a bit of zinc. Mercury toxicity was known in the 19th century, but amalgam's cheapness, ease of placement, and durability kept it popular. Dentists argue that mercury fillings last longer than resin composites, and are more gentle to tooth pulp. Composites also require more skill and time to place.

Unfortunately, mercury is a poison that penetrates all living cells of the human body. It is more toxic than lead, cadmium and arsenic. The smallest amount of mercury that won't damage human cells is unknown. Autopsy studies show a correlation between the number of mercury fillings and mercury levels in the brain and kidneys. Research also indicates that amalgams have an adverse effect on the immune system's T-lymphocyte count.

Scrap dental amalgam is classified hazardous waste by the American Environmental Protection Agency, and by law must be stored in unbreakable, sealed containers, and handled without touching. Dr. Sandra Denton, M.D., who specializes in treating chronic mercury toxicity, asks: "What is it about the mouth that makes this same stuff nontoxic?" Referring to American Dental Association (ADA) claims that amalgams have been proved safe in studies, Dr. Denton challenges them to produce such studies. They have not. "On the other hand," says Denton, "research documenting mercury toxicity is voluminous." She has collected some 3,000 articles and several books on the topic.

A Danish study found that Multiple Sclerosis (MS) patients had eight times higher levels of mercury in their cerebrospinal fluid than healthy controls. An article in the Journal of Forensic Medicine & Pathology states: "Slow retrograde seepage of mercury from root canal or Class V amalgam fillings ... may lead to multiple sclerosis in middle age." Dr. Hal Huggins of Colorado Springs, Colorado, a dentist who has MS himself, treats MS victims and people with other chronic health problems by removing mercury amalgam fillings as well as with detoxification and nutritional supplementation. He claims that 80 to 85 percent of his patients improve significantly.

Despite Huggin's successes, the U.S. Multiple Sclerosis Society opposes mercury amalgam removal, stating that they have found no scientific correlation between amalgams and MS. Dr. Huggins counters that if his results are to be written off as "anecdotal" or "placebo effect", then he has the largest collection of sustained recurring anecdotal placebo responses in the world.

Antibiotic resistant bacterial disease has become a significant and growing public health problem over the past decade. Studies show that genes protecting bacteria against mercury poisoning often bundle together with other genes that give bacteria antibiotic resistant qualities. If amalgam fillings stimulate and maintain populations of mercury-resistant bacteria, it's no major stretch to suggest that they might also be an agent in developing antibiotic-resistant bacteria. Research by Dr. Anne O. Summers, et al. at the University of Georgia shows such a relationship in monkeys. Dr. Summers put mercury fillings into the molars of monkeys. Within five weeks bacteria in the animals' intestines became resistant not only to mercury, but also to common antibiotics like penicillin, streptomycin, and tetracycline.

Another monkey study by Dr. Stuart B. Levy at Tufts University found that before having mercury fillings, an average of one percent of the monkeys' oral and nine percent of their intestinal Enterobacteriacae were antibiotic-resistant. After receiving mercury fillings, 13 percent of oral and up to 70 percent of intestinal bugs became antibiotic resistant. The ADA responds by reiterating its stand that mercury fillings are safe, and arguing that animal studies "cannot be viewed as affecting humans."

It is well-established that elemental mercury vapour emits from amalgam tooth fillings during chewing, brushing, and eating hot and/or acidic foods. Most of this vapour is inhaled allowing efficient absorption across the alveolar membrane in the lungs. Mercury easily crosses the blood/brain barrier -- the brain and nervous system's main natural defense against many toxic substances. It can bind strongly to sulfur-containing proteins in nerve tissue (which may explain the association with MS--a disease of the nerve sheaths), and deposits in virtually all body tissues and organs. In experiments on mercury fillings in sheep, Dr. Murray Vimy, a dentist at the University of Calgary, proved that mercury migrates from the teeth into nearly all body tissues, especially the brain, kidneys, and liver.

The average dentist handles two or three pounds of mercury annually. According to Consumer Reports, up to 10 percent of dental offices have mercury vapour levels exceeding 50 micrograms per cubic metre of air --the upper limit considered safe for eight-hour workplace exposures. Dr. Sandra Denton cites a study at the University of North Texas that found neuropsychological dysfunction in 90 percent of dentists tested. Female dental personnel have a higher spontaneous abortion rate, higher incidence of premature labour, and elevated perinatal mortality, which has been substantiated by the EPA to be characteristic of women chronically exposed to mercury vapour. Stillbirths are significantly correlated with maternal blood mercury levels. Methyl mercury, the organic form of mercury that forms after oral ingestion of mercury, is 100 times more toxic than elemental mercury. Methyl mercury easily crosses the placental barrier and builds up 30 percent higher red blood cell levels in the unborn child than the mother.

The CDA counters that with billions of mercury amalgam fillings placed, there is no apparent epidemic of ill health effects. However, others argue that so many people have mercury fillings that no effective control group exists. Former Health Canada biologist Mark Richardson, who researched the scientific literature on mercury toxicity in preparing a risk assessment, notes that it is people wanting to maintain the status quo who conclude that there is no evidence that mercury toxicity is a health problem. He refers to the tobacco industry's stalwart insistence that studies linking smoking to lung cancer are unscientific. Richardson's report, under consideration by Health Canada, recommends limiting the number of mercury fillings per person.

Stubborn reluctance of dental associations to acknowledge the health risk of mercury toxicity from amalgam fillings may indeed have much in common with tobacco company tactics. If diseases like Multiple Sclerosis, Chronic Fatigue Syndrome, and Multiple Chemical Sensitivity are linked to mercury exposure from tooth fillings, significant potential exists for individual or class action lawsuits against dentists. Indeed, the German Dental Association has stated that if the government recommends further limitations on amalgam use, it will advise its members to stop using amalgams completely due to increasing risk of legal liability. The truth will eventually out, and if mercury fillings are indeed eventually proved harmful, a history of foot-dragging will not bolster the dental community's case in court.

Dr. Murray Vimy is certain that every time you chew, brush, or grind your teeth you absorb mercury. However, he councils against panic and suggests that mercury fillings be replaced with non-mercury materials like resin composites, porcelain, or gold, as needed. There is some risk that mass replacements could expose the patient to more mercury than if old fillings were left alone.


By Charles W. Moore

Charles Moore is a freelance writer living in rural Nova Scotia who specializes in health issues.