IS THERE A RELATION BETWEEN SELENIUM IN BLOOD AND SUBJECTIVE SYMPTOMS SELF-RELATED TO DENTAL AMALGAM?

Paul Johan Höl, Nils R. Gjerdet, Rune Eide, Jan S. Vamnes, Rolf Isrenn (Dept. of Odontology, Dental Biomaterials, University of Bergen, Bergen, Norway).

E-mail address of corresponding author: Paul.Hol@odont.uib.no

ABSTRACT

Animal studies have shown that selenium plays an important part in the protection against mercury toxicity. This is often explained by the formation of a HgSe-complex bound to selenoprotein-P in blood. The aim of the present study was to investigate if the selenium levels were affected in persons who reported general health problems associated with dental amalgam fillings. The selenium concentrations were determined in whole blood samples (B-Se) of 80 individuals by hydride generation AAS. The subjects comprised four groups: 19 healthy controls without amalgam experience (median Se-concentration: 123.0 microgram/l (ug/l)); 21 healthy controls with amalgam fillings (130.3 ug/l); 20 patients who claimed symptoms from dental amalgam (119.2 ug/l); 20 patients who have had amalgam fillings removed due to suspected symptoms associated with amalgam (124.7 ug/l). The B-Se concentrations was statistically significant lower in subjects who claimed symptoms of mercury amalgam illness, than healthy subjects with amalgam (p=0.05). This difference was more evident between the individuals with more than 35 amalgam surfaces (p=0.003).

INTRODUCTION

It is established that dental amalgam fillings release detectable amounts of mercury to the body. Several studies have demonstrated that there exist a positive correlation between the amount of amalgam and the presence of mercury in blood, urine and tissue (Björkman, 1995). The potential clinical effects of the mercury release are debated. It appears that no generally accepted criteria exist for so-called “amalgam-related illness” which includes mainly subjective symptoms and general ill-being of the persons involved.

It is assumed that the Hg²⁺ cation is the proximate toxic species of both mercurous and mercuric compounds (Clarkson, 1997). Hg²⁺ reacts with a variety of ligands. The biochemistry of inorganic mercury in the mammalian body is dominated by its reaction with sulfhydrile groups. It is found in cells and tissues attached to thiol-containing molecules as cysteine, glutathione, metallothionein and some enzymes.

Selenium (Se) reacts with Hg in the bloodstream by forming complexes containing the two elements at an equimolar ratio when selenite and inorganic mercuric are co-administered (Yoneda, 1997). Selenite is effluxed from red blood cells after being taken up selectively and reduced by glutathione (GSH), and then the reduced form of Se forms equimolar (Hg-Se) complex with Hg in the plasma. The equimolar complex binds selectively to a plasma protein, selenoprotein P (Sel P), to form a (Hg-Se)-Sel-P complex. (Suzuki,1998).

Se levels have been mentioned in connection with amalgam removal procedures (Schrauzer, 1995).

The aim of the present study was to investigate if the Se levels in blood are affected in persons who report general health problems self-related to the presence of dental amalgam fillings. Moreover, to establish a relationship between the exposure variable i.e. the number of amalgam surfaces, and Se in blood.

Material and methods

Eighty individuals in four groups were investigated (table 1):

1. Nineteen healthy volunteers without amalgam experience.

2. Twenty-one healthy volunteers with amalgam fillings.

3. Twenty patients with symptoms allegedly caused by their dental amalgam fillings.

4. Twenty patients who had removed their amalgam fillings because of concern about illness caused by mercury released from their dental restorations.

Groups	Male/female	Age (min/max)	no. of amalgam surfaces (lower/upper quartiles)
1. Never had amalgam	10/9	21 (19/25)	0
2. Healthy with amalgam	6/15	43 (28/58)	40 (29/49)
3. Alleged symptoms from amalgam	5/15	46 (25/65)	35 (25/45)
4. Removed amalgam	8/12	50 (24/65)	48 (23/58) (before removal)

Table 1. Age, sex-distribution and number of amalgam surfaces (medians).

This material has previously been investigated by Vamnes, JS et al. (2000), who collected blood and urine samples for mercury analysis. The persons in the group who never have had amalgam (group 1) were younger than those in the other groups, because middle-aged persons without dental amalgam experience were not obtainable in the Norwegian population. In the healthy group with amalgam, (group 2) 16 of 21 persons were health-workers (nurses and doctors). For the three groups with amalgam-experience age, sex and number of amalgam surfaces or previous surfaces were not statistically different. The median time since removal of amalgam in group 4 was 31.5 months (range 12 to 96 months). The patients with amalgam-related illness (group 3) were selected consecutively from those who were referred to the Dental Biomaterials Adverse Reaction Unit at the Faculty of Dentistry, University of Bergen, for the evaluation of possible side-effects from their amalgam restorations. Prior to admission to the unit, they had been examined by their physician and dentist who reported their symptoms on a reporting form. All patients in group 3 and 4 were convinced that the presence of amalgam was an important etiology of their illness. In order to participate, the subjects with amalgam illness should report at least three of the following, general subjective symptoms: General fatigue, impaired memory, concentration problems, muscle or joint pain, digestion disorders, vertigo, headache or oral symptoms (metallic taste, burning sensations, salivation problems). The dental and oral status were recorded for all participants. The overall exclusion criteria were occupational exposure to mercury, other heavy metals, or solvents, abuse of alcohol or drugs or high consumption of marine food.

The blood samples were collected in acid-cleaned polypropylene tubes and frozen (-20⁰C ) until analyzed. The blood samples (1.5 ml) were introduced into Teflon vessels together with 4 ml of conc. nitric acid and 2 ml of 30 % hydrogen peroxide and digested by the microwave-technique (Milestone 1200 MEGA, Sorisole, Italy). Before the analysis, Se was reduced from Se⁺⁶ to Se⁺⁴ by adding 0,5 ml conc. HCl to 0,5 ml of the sample solutions, and placed in a water bath at 100⁰C for 45 min. Analysis of total Se was performed by hydride generation atomic absorption spectrometry (Perkin Elmer 372 equipped with an MHS-20). Sodium borohydride was used as a reducing agent. All samples were at least analyzed in duplicate. The detection limit for the method, defined as 3xSD in a blank solution (N=10) was 0.6 ng, or 8.9 ug/l in a sample. The accuracy of the analytical method was monitored by a reference material.

RESULTS AND DISCUSSION

The concentration of selenium in blood (B-Se) was statistically significant lower in the subjects who claimed symptoms of amalgam illness, compared with healthy subjects with amalgam (Figure 1). The difference between the individuals of group 2 and 3 with more than 35 amalgam surfaces was more evident. Healthy individuals with amalgam revealed a significant positive correlation between Se-B and amalgam surfaces (Spearman’s r = 0.47, p = 0.038), whereas persons reporting amalgam related illness, revealed a negative correlation (Spearman’s r = -0.51, p = 0.023).

Figure 1. Se-concentrations in whole blood, ug/l (medians, whiskers represent quartiles). Open bars represent subgroups with > 35 amalgam surfaces.

The levels of Se-B for Norwegian citizens are among the highest reported in Europe (Schrauzer, 1995, Meltzer, 1993), which was confirmed in this study. All four groups have normal Norwegian Se-B concentrations, but the difference in concentration levels in the amalgam groups needs to be explained.

If we refer to the formation of HgSe in the bloodstream, this reaction depends on the availability of both mercury and Se (as Se^2-) in the blood. Akesson (1991) found a significant association between B-Hg and P-Se (r = 0.2, p = 0.001) in a study of 244 dental personnel and 81 matched referents. Dental personnel had higher U-Hg (p < 0.0001), P-Hg (p = 0.03) and P-Se (p = 0.0007) than referents.

In this study Se-B correlates with the number of amalgam surfaces, but not with Hg in whole blood.

Drasch (1996) found that at a lower Hg-conc. (< 700 ug/kg) in the kidney there is excess of Se sufficient to bind all mercury passing the kidney. For Hg-conc. > 700 – 1000 ug/kg (persons with amalgam fillings in more than 8 teeth can reach this level (Schupp, 1993)) additional Se is utilized to maintain a stable 1:1 ratio, so that further increasing Hg-conc. may bind Se passing through the kidney. Can this explain our findings regarding the correlation between amalgam surfaces and Se concentration in blood? If the free Se-concentration decreases in the kidneys, will this affect the Se-concentration in blood? Drasch assumes that if the formation of a 1:1 Hg-Se complex actually takes place, any increase of mercury that is not accompanied by adequate Se supplementation will allow the unbound or “free” mercury to react with enzyme SH- residues causing partial or total enzyme activity inhibition.

Sometimes Se supplements are advocated in case of “amalgam poisoning”. It could be speculated whether such supplements could contribute to higher levels of Se in the patient groups. The levels are however, lower in the amalgam illness group compared with the control groups. On the other hand, many of the persons in the healthy control group with amalgam are in health-oriented professions and thereby more concerned about taking nutritional supplements, including Se, but we have no indications that this is the case.

Another possibility is that a metabolic interaction exist between Hg and Se.

Thus, it is indicated that persons with ill-health, in this case related to dental amalgam, might have a different Se-metabolism compared with healthy people.

REFERENCES

Akesson I, Schutz A, Attewell R, Skerfving S, (1991), Arch. Environ. Health 46(2):102-109.

Björkman, L (1995), Studies on dental amalgam and mercury. Exposure, accumulation and effects. Thesis. Stockholm, Karolinska Institutet.

Clarkson TW (1997), Crit. Rev. Clin. Lab. Sci. 34(4): 369-403.

Drasch G, Wanghofer E, Roider G, (1996), J. Trace Elements Med. Biol. 10(4): 251-254.

Meltzer HM, Bibow K, Paulsen IT (1993), Biol. Trace Element Re. 36(3): 229-241.

Schrauzer GN (1995), In: Status quo and perspectives of amalgam and other dental materials. (Friberg LT, Schrauzer GN, eds.), Stuttgart, Thieme, pp. 106-118.

Schupp, IB (1993), Untersuchungen an menchlichen Organen zur Frage der Quecksilberbelastung durch Zahnamalgam und weitere Faktoren, Thesis. München, Ludwig-Maximilians Universität.

Suzuki KT, Sasakura C, Yoneda S (1998), Biochim. Biophys. Acta. 1429(1):102-112.

Vamnes JS, Eide R, Isrenn R, Høl PJ, Gjerdet NR. (2000), J. Dent. Res. 79: 868-874.

Yoneda S, Suzuki KT (1997), Toxicol. Appl. Pharmacol. 1997;143(2):274-280.