Pharmacology & Toxicology 1992. 70, 308- 313.

Dental Amalgam and Mercury Asbjorn J okstad", Yngvar TlIomassenl), Erik Byel), J ocelyne Clench. Ass'l and J an Aaseth"l

11Department of Anatomy, Dental Faculty, University of Oslo, P,Q Box 1052 Blindern, N·0316 Oslo 3. l!Nalional Institute of Occupational Health, Oslo, "\lorwegian Institule for Air Research, Lillestmm , and "Hedmark Central

Ho>pital. Elverum. Norway

(Received October 21,1991: Ae<.:epted December 18. 1991)

Ab~·lracr. The mercury concentrations in blood (HgB) and urine (HgU) samples, and in exhaled air (HgAir) were measured in 147 individuals from an urban Norwegian populalion. using cold vapour atomic absorption spectrometry. The >tudy aimed to estimate the mercury exposure from the dental reslorations, hy correhlting the data 10 the presence of amalgam re>tor~tions. Mean values were HgB = 24.S nmoll1. HgU _ 17,5 nmolll and HgAir = O.S j.4glm', HgU correlated with HgAir. and both HgU and HgAir with the number of amalgam restoralions, amalgam restored surfaces and amalgam restored occlusal surfaces. 11gB sho,",'ed poor correlalion to HgU and HgAir and the presence of amalgam reslorations. A differenlialion of the mercury absorplion due lO exposure from dental amalgams and from the dietary intake, necessitates measurements of both organic and inorganic mercury in the pla,ma, and in the erYlhrocytes. The results suggesl that individuub with muny umalgam restoration:;, i.e., more lhan 36 restored surfaces, absorb 10 12 ~!g l'lglday.

In order to estimate the impact of the exposure to mercury

in the cnvironment on general health it is necessary to

identify. characterize and quantify all contributing SOUf(;es.

One potential source have been shown to be the degradation

of dcntal restorations made from amalgam (Brune & Evje

1985). Dental amalgams consi~t of approximately 45-5(),~'O

mercury, 25 35% silver, 230% copper and 15- 30"/0 tin (Phil­

lips 1986). Various stratcgies have been used to estimate

the exposure levels of mercury from amalgam n:storations,

based on in l'itro experiments (Brune & Evje 1985: Okabe

1987: M arek 1989). or measurements in exhaled air or in

intraoral air (Gay 1'1 ai, 1979: Vimy & Lorscheider 1990:

Berglund 1990). Estimations have also been made from

measurements of mercury levels in the urine (Frykholm

1957: Olstad 1'1 al. 1987). blood (Kroncke ct al. 1980: A brah­

am cl al. 19R4). plasma (Molin 1990), or saliva (Ot! 1'1 al. 1984). Few of these studies report the mercury concen­

trations in both blood and urine or exhaled or intraoral

air samples, and the estimates of the daily contribution of

mercury from amalgam restorations arc comradictorary.

Thc present study aimed to assess the mercury contcnts

in blood and urine, as \.I'ell as the men;ury amounts in

exhaled air in a segmclH from an urban Norwegian popula­

tion. A further aim was to estimate the mercury exposure

from the dental restorations, by correlating these data to

the presence of amalgam restorations.

Materials and Methods

Th~ prescnt study was a sub· study of a larger invesligalion on environmental e~posure to contaminants from traffic in an area within Oslo, Nomay Klench·Aas 1991). One hundred and sixty­lWO individuals, randomly chosen from cohorts of the larger ma­leriaL were invited to participate in the present study. After a wrinen consent. each participant visited a (;entral clinic for a series of tests . The group finally consisted of 147 individuals with ages ranging from 3- 87 years (median _ 33 years). Ninety-four of the participants (63.9"/0) were females.

The consent permiHed one of the authors to oblain data on the dental status from lheir own dentiSls. The denIal stalus was assessed in most (;ases on the basis of X· rays and the patient's record. This mode of assessment was often not possihle for a variely of rea:;ons. The prevailing reason was a lack of regular atlendance al a particu­lar dental practice. Therefore, the participants with no dala avail· able from their dental record '""ere invited 10 be examined at a university dental clinic. The examination wa:; made by an e.~peri·

enced clinician, using a dental mirror and a probe, Recordings were made of the total number or amalgam restoralions. the number of amalgam restored toolh surfaces and Ihe number of amalgam re· stored occlusal surfaces. No information was avai labk on the po,,· ible prior presence of amalgum for the adult participants with no amalgam re:;toralions. No aHempt was made to 4uaJilatively s(;ore the restorations by larnish, porosities or marginal degradation.

Blood Silmples were collected from the cubital vein in vacutainer tuh~s, tested free of mercury contamination < I nmo! Hgil, with beparin as antico~gulant. The tOlal blood mercury concentration (HgB) was determined ~fter hemolyzalion ,

Urine was collected in polystyrene bouies. The participants pro· vided morning urine sample~. to reduce the effects of daily nuctu· ation ofexcretcd mcrcury (Piotrowski 1975). All samples were re · frigeruted 10 4' , and kept cool unt il analysis . The HgU was deter· mined by cold vapour atomi~ absorplion spe~trometry (Ebbestad 1975), applying a modified LDC mercury monitor model 1205 , The detL'(;tion limit is 0.5 nmolil, and the precision of the method is 2,0",{, R,S.D at the 50 nmol Hgli lewl HgB was determined after a nitri~ I pcrchloric acid digest ion of the sam pies. The detection limil for mercury in whole blood is mostly dependenl on the mercury content of the ~cids . Throughout this study the delection limit of the melhod was kepI below 2 nmolll blood , All :;amples were analyzed in duplicate. Quality assuran(;e mate rials from Nycomed, Norway (Seronorm Trace Element Whole Blood and Urine) were used as control ,umples. The mere!.!ry concentrations found in these materials was in good agreement with the producers certificate (within +5%). HgU was also adjusted for urine now rale, by relating the values to the creatinine concentr:! tions.

The pulmonary air samples werc collected arter the following procedure: Normal breathing was performed for several minutes before a deep brealh. After holding Ihe breath for 20 sec .. about half was blown out through the mouth, und the remaining ai r blown lOla a 31 plastic air bag, The air bag consisted of an opening valve. in lo which a cardboHd mouth-piece was inserted . This procedur~ ensures the expiration ofinlraoral air first. followed by pulmonary

•

•

•

•

,

,

•

•

DENTAL AMALGAM AND MERCURY 309

air which was collected in the plastic bag. The air samle was ana­lyzed for mercury (HgAir) immediately after sampling, using the LDC monitor. One air sample was measured from tach patient using a continous air now of 300 ml l min . through the LDC moni ­tor, The detection limit was 0.1 I4g / 1 with a 50% variance and 2% precision at this leveL Calihration was made against air standards containing known amounts of mercury vapour.

Correlations octwccn the variables were computed using Pearson correlation coefficients. The correlation coefficients between HgU, '-1gB and I-IgAir versus the participants ' age and amount of dental amalgam were compUied both without and with subgrouping the val uc >. I nit ial examinations o f SC~ tterplots of the daw showed that the relatio[1!;hips between the amount ofamalg~m verSlis HgU and HgAir w~re non· linear. Furthermore. the variance o f the residuals decrea>cd with less amo unt o f Hmalgam. i.e. there wa, an inhomog­eneity of the varian<:~ . The <:orrelations between the Hg values. gender. age and the ,1mOI,mt of amalgam wa> therefore Hiso esti ­mated in it multiple linear regres>ion model. ~fter log transformation of Ihe Hg values. However. no regre,sion model> for the Hg levels were made from the data recorded in the present >tody. The value of sllch models would he limitcd due to the lack of recording important factors affecting the Hg release from amalgam. e.g. the age and quality of the restorations. their technical qualily. pre>cnce of other melals and other environmental conditions intra·orally. ,," well a s other va riables such as smoking. frequency of fish meals ~nd use o f medicat ion .

Results

The dental status was recorded in [15 participants, while 32 failed to report the name of their dentist and did not attend the dental examination. The dental amalgam restora­tions varied between 0--69 restored surfaces (median = 24), 0- 20 restored occlusal surfaces (median = 10), and 0--29 res­torations (median = I J). Twenty-two participants (19.1'%) did not have any amalgam restorations.

Blood assays were submitted from 133 individuals. Samples were not received from the young children (n = 5), and from 9 adults unwilling to participate in the blood sampling. The "1gB concentrat ions varied between 3 and 71 nmol/l, with mean HgB = 24.8 nmolll (sd = 11.8, median = 23 nmol / I) (fig. I). Urine samples were analyzed for all the participants. except in one case due to a breakage of the plastic container (n = 146). The HgU values varied between 2 and 80 nmol/l , with mean HgU = 17.5 nmol/l (sd = 16, median = I I nmolll) (fig. I). HgAir samples were received from all 147 participants. The HgAir varied from amounts at the detection limit of 0.1 ~tg/ml up to 9.8 fJ.g / m ' . The mean HgAir was 0.8 fJ.g i m1 (sd = 1.3, median = 0.3 Ilg/ml) (fig. I) .

The frequency dis tribution curve for the HgB values (fig. I), having a gaussian like shape, differed from the negative skewness of Ihe distributions of J-I gU and HgAir values. Furthennore. the HgB values fai led to correlate with Ihe HgU values, and correlated only sligh tl y with the HgAir concentrations (table I). Relatively good correlations were obtained betwecn the HgU. the HgAir values.

The correlation coefficients between the diffcrem indices of the amount of dental amalgam and HgU, HgB and HgAir varied only slightly (table I). Recalculating the correlation coefficients after subgrouping the participants according to

Frequency {$}

30

'0

° Frequency ($) 30

'o f----

10 1--

° Frequency (%) 50

40

30

20

10

o

nmoJ/1

nmol/l

2 3 ug/m3

Fig. I. The frequency distribution of the mercury concenJrations in urine S<lmp1cs (HgU) (n ... 146) (upper panel) . whole blood samples (HgRJ (n = 133) (middle panel). and in exhaled pulmonary air samples (HgAirJ (n .. 147) (lower panel) from habitants of Oslo_ No adjustmenl h ilS been made for the creatinine contcnt or the specific gravity of Ihe urine ,ample.

the amount of amalgam yielded approximately the same correlation coefficients, with no changes of the probabilities .

Fig. 2 shows a scatterplot of the Hg concentrations fo r all the participants. while fig . 3 illustrates the average con­centrations within 4 groups categorizt."\.l according to the number of surfaces restored by amalgam. Adjusting Ihe HgU valut:s with the creatinine concentrations did not influ­ence the overall mean or frequency distribution of the HgU values. Nor did the creatinine-adjusted HgU values influ­ence the correlation coetlicients. except identifying a poss·

- - ,

310 ASBJ0RN JQKSTAD £T AL

Table / .

Cross-correJat ion between the dependent and independent variables. Correlation I:akulated by Pearson 's linear correlation index, using the individual values. Probabilities ca lculated by Students' H.:S!. one tailed ,ignificance.

Mercury in Number of amalgam restorations

urine blood

Urine

. •

TO((lI

surfaces

0.549 1' < 0.001

Blood 0.086 0.18!; P= O. I64 1' = 0,027

Air 0.369 0 .159 0.422 P < OJXlI P = 0.033 P < O.OO[

iblc effect of gender. Therefore, on ly the unadjusted HgU values were used in the statistics.

The HgB values increased sligh tly with increasing amounts of amalgam. as scen on the scatterplot (fig. 2). and on the mean values after subgrouping the participants (fig. 3). The correlation coefficicnts were in both cases relatively

low. No differences were obtained with regard to gender or age (P > O.05) (table I).

T he HgU values correlated with the thn:e indices of thc amount of amalgam and with the participants' age

(P < O.ool). Lower HgU values were generally found among the oldcst participants, compared to the others, irrespective

of the amount of amalgam (fig. 4). The total variance ex­plained by a multivariate analysis using gender, age and

number of restored surfaccs in the model was R ' = 36%. A log-transformation of the HgU values, and using the same modcl. increased the R2 to 47'Yo. Approximately the same

R' values were obtained when using thc mean values of subgroups instead of the individual HgU values.

T he HgAir in the pulmonary air samples correlatcd to

the amount of amalgam (P < 0.(01). A weak correlation was also seen to age. Gender did not seem to innuence the HgAir

conccntrations (table I). Thc total variancc explained by a multivariate analysis using gender, age and number of re­stored surfaces in the model was R' = 20%. A log-transfor­mation of thc HgAir valucs, and using the same model,

increased the R2 to 39"1.,. \\'hetl using the mean values of subgroups instead of the individual HgAir values the R' increased furthe r to 41 %.

Discllssion

The dental status of the participants in the present study

(;3n be considered rcprescntative for the dental health of urban Non.vegians ( Bjertness 1990). Also the HgU and HgB values compare well 10 the data from the fcw previous

demographic studies on the HgU (Lie 1980) and HgR (Ma­thiesen 1980; Syversen 1982) amounts in non-occupationally exposed Norwegians. There arc no previous rcports on

measurements of the mercury concentrations in pulmonary air samples in a Norwegian population .

The diffcrcnces in correlation between the different meas-

Occlusat surfaces

0.561 P< O.ool

0.127 1' = 0.098

0.389 1' < 0.001

Resto· rations .

0.524 P < 0.001

0.165 P= 0,046

0.397 P < O,OOI

Age Gender

- 0.327 - 0.172 P < O,OOI P = 0.019

0.084 0.043 P = 0,168 P = 0.311

- 0.159 - 0.132 1' = 0 ,028 P = 0.056

ures of thc amount of amalgam and thc mcrcury concen­trations in urine, blood and ai r were not large. This is in accordance with previous studies (Berglund 1980; Zander

et af. 1990; Akesson el af. 1991). The whole blood mercury concentrations were insignifi­

cantly innuenct.-u by the number of amalgam restorations. Our data thus support the theory that the whole blood mer­cury renect ingested methyl mercury (Me Hg). rather than in ·

organic mercury (Birkc et al. 1972; Skcrfving 1974). Furthcr · more, in a recent study it is shown that the major part of total men;ury in non-occupationally exposed Norwegians is

present as mcthyl mercury (Bulska et al. 1992). This interpre· tation is in accordance with the majori ty of previous reports ( Kroncke et af. 1980; Ott el af. 1984), but conflicts with two

other reports (Abraham el al. 1984: Patterson el a/. 1985).

The reason may be due to the diffcrences in analytical tech· niques or estimation procedures, as well as the use of rela­tively small samples in the studies. Moreover, lack of control

of other potential mercury sources may have biascd the rc· suits, c .g. dietary intake, smoking, chewing habits, medi­cation, and alcohol consumption . On the other hand, our av­

erage diffcrcnce of6 nmolll between the amalgam.free group and the group with extensive amalgam restorations (fig. 3) compares well with the data of Abraham et af. (1984) and Snapp (1989). Furthermore, plasma mercury has becn re­

ported to correlate with the number of amalgam surfaces (Molin 1990). Apparently. Me l-ig, which is mainly located in

the erythrocytes, is high enough to camouflage the fluctu­ations of inorganic mercury in plasma in the whole blood analyses (Bulska et (If. 1992). Thc results thus indicate that in

order to diffcretlliale the exposure from dental amalgams and

food , separate measurements of the organic and inorganic mcrcury should be made in plasma. and in the erythrocytes.

T he urinary mercury values, on the other hand, correlated

to the amount of ama lgam, which is consistent with p revious reports (Olstad el al. 1987; l o kstad 1987). Occupational

HgU exposure is renected by increased urinary mercury levels, when assessed on a group basis (Skerfving & Berlin 1985). The recommended upper limit of the Hgo content in an occupational atmosphere, (25 )lg/mJ. WHO study group

1980), imply an average individual uptak e of about 100

)lg/day (Aaseth & Barregard 1989). As faecal and urinary

,

•

•

DENTAL AMALGAM AND M ERCU RY 311

H g 6 U

N M 0 3 L 1 L

H g A

" g I, M

• • ••

• • •

• • •• .. • • .. • • • • • •

• •• •• .. • • • "''''****** • **"' ••••• •• *********** •• ••• ***** • *****Ir**** ***********Ir** • • ***/r

0 10 20 30 " 50 60 AMALGAM RESTORED SURFACES

6

• • , •

• • • •• • • • •

• • • , • • • • • • • •

***** ***** ********** ****** ************ *************** •

o *.**** **** ** '" ****** *** o 10 20

AMALGAM 30 40

RESTORED SO 60

SURFACES

•

70

•

70

Fig. 2_ Scaltcrpiol of th~ mercury concentrations in urine (HgU). whole blood (HgB) and in plilmon~ry air ,amp1es (1-lgAir) a, a function of (he number of dental amalgam restored surfaces.

excretion of mercury after exposure to Ht arc of the same magnitude, this exposure corresponds \0 a urinary excretion of about 50 pg/day, or approximately 50 ).Ig i l (Za nder et al. 1990). In the present sludy, the mean J-IgU excretion in the group with many restorations was 6 I-I g/1, i.e , 12% of the value reflecting the estimated upper hcalth-based limit. Ho\\'ever, the HgU valucs in the present study showed large varialions, possibly due to differences in excretion kinetics

nmol/ l 50 ~-~~-----="-' r---~

'0 e---l--- - - l--- --' ---~-

30f-- -

20 e--++ 10 f-.r.

o li& 0 I;; . 21-36(n=30)

RII 1-20(n=30) • 36-69(n=30)

Fig. 3. Me~n mercury v~lues in urine (HgU), whole blood (HgB) and pulmonary air sampks (HgAir) from four subgroups. Group I i, the mean of 22 parli~ip~nt, withoul ~malgam resloralions. Groups 2- 4 are equally siled. ~nd show Ihe mean .. ~Iues of the participanls wilh :Jmalgam re>tor~lions, grouped according 10 the number of dcnla! amalgam reslo red surfaces. The verlical bars represent the standard deviations.

(Zander el al. 1990). However, it is also theoretically pos­sible that some of these variations are due to different pal!erns of tooth grinding, and that extensive bruxism may decrease the safely ratio (Siillstcn et af. 199 1).

The low HgU concentrations from participants without amalgam restorations may rcneci ingested McI-lg in food , especially in seafood. which is demethylated before urinary excretion. The analytical method used in the present study, however, is not capable to determinc the Me Hg fraction in the urine.

It is not clear what the HgAi r concentrations in the present sludy represents. Earlier st udies have reported a

HgU (nmollil '0 ----------

(2) (25) 30 ----" •• t-----t.k_~__,,(3l

~­-~. 20 -----' '-=c:::i~:~--

o -----,3C-.2.4---2'6.-"5"0,--5.,.-~8"7 Patiant age group,.

. : 0 surfaces T ; \- 20 surfaces +: 21 - 36 sur faces . : > 36 surfaces

Fig, 4, Urinary mercury (HgU) in relation to the number of dental ~malgam restored surfaces and participanls' ~ge , The age groups ~re : 025 years (n = 20). 26 50 ye~ rs (n .. 66) and 51 87 years (n = 26). The amalgam groups are: 0 surface> (n .. 22), 1 ~20 surfaces (n = 30). 21 36 surfaces (n = 30) and >36 surfaces (n .. 30). The number> in parentheses show the number of Obscf"\lill ions within each >ubgroup.

--,

...................... _---_ ...... _ .... __ .... _----_ .. _-._----------_ .. _._--------_ .. __ .. _------- - ------------------ ------------------

312 ASBJ0R N JO KSTAD ET AL

correlation between the amount of amalgam and mercury vapour present in air samples where participants have breathed into tubes (Gay et al. 1979; Patterson el al. 19H5), or bags (Svare el (If. 1981; Ott cl II/. 1984), where the oral cavity has been flushed and ai r collected with a suction device (Abraham el al. 1984), or by use of intraoral suction probes (Vi my & Lorscheider 1985a & b; Berglund el a/. 1988; Berglund 1990). However, these studies were not aimed to measure the HgAir cOllcentratiOIl in pulmonary air. In the present study all the potential intraoral mercury vapour was exhaled during the first part of the sampling procedure. Thus, the present HgAir values do not represent the intraoral mercury concentrations as a function of time. Theoretically, some mercury may have vapourized during the second half of the exhalation, but this fraction would be minimal. The HgAir values therefore probably represent dissolved Hgo in the blood stream, lung tissues and lung nuids (Cherian ct (if. 1978). The lack of significant corre­lation between HgAi r and HgB in the present study may be explained by the assumption that the predominant mercury in blood, i.c. Me Hg (Bulska ct al. 1992), is bound intracellu­lary or to plasma proteins (Skerfving & Berlin 1985). The wide range of the HgAir concentrations within the groups categorized according to the amount of amalgam (fig. 3),

however, indicates that the pulmonary excretion pattern shows marked inter-individual variation, and it may have been inn uenced by drugs or by alcohol consumption (Berlin 1986).

The first estimates of the daily exposure reported in the literature were based on data of intraoral mercury vapour measurements (Vimy & Lorscheider 1985a & b). Measure­ments of 35 persons with amalgam restorations showed that the intraoral mercury concentrations increased from about 5 ).1g / m3 to about 30 ).lg/ml after chewing, and remained relatively high for some time afterwards. The daily body burden was estimated to be 20 ).lg/day. However. their esti­mations were refuted by other researchers (Mackert 1987; Olsson & Bergman 1987; Berglund 1.'1 a/. 1988; Olsson elal.

1989). These researchers suggested that the daily exposure ranged around 1- 2 ).lg/day. Vimy & Lorscheider (1990) recently presented a reevaluation of their da ta, and eOll­duded that a daily exposure of 10 ).1g / day seemed more correct. However, also the validity of these re-estimations can be questioned, since they arc based on a limited number of measurements. The intraoral mercury vapour measured over 24 hr do not correlate to singular or even short series of measurements (Berglund 1990).

The daily exposure of mercury from dental amalgam may be estimated by using a metabolic model as a function of the whole blood mercury concentra t ion, the fract ion distributed to a unit volume of blood, and the biological half-time (Clarkson I't al. 1988). Three assumptions must in this case be verified . The first is that the body tissue compartments have achieved a steady state. The second assumption is that a single biological half time is sufficient to describe the rctention in a particu lar compartment and third, that a constant fraction of the daily dose is deposi ted

in each compartment. The whole body half lime after a single mercury exposure is 58 days ( H u~rsh 1.'1 al. 1976). while steady state between body tissue compartments and mercury exposure is achieved in approximately 5 times the biological ha lf-time (Clarkson et al. 1988). Since the participants in the present study had many amalgam restorations it is as· sumed that the HgB concentrations were in a steady state. The biological half time of mercury in blood is 3.3 days and 2.1 % of the dai ly dose is deposited in I 1 whole blood (KershaWei al. 1980: Cherian el al. 1978). The model thus suggest that under these circumstances the whole blood mercury values renect 10'1'0 of the daily absorbed mercury (Clarkson 1.'1 (II. 1988). The mean HgB in the amalgam-free group and the group with the extensive number of amalgam restorations differed by 6 nmolll (1.2 ).1gll). Provided tha t this difference is caused by the amalgam restorations on ly, and that the assumptions above arc correct, this calculat ion would suggest a daily exposure of 10'" 1.2 ~lg/l = approxi .

matcly 10-12 ).1g/day. The difference in mean HgU in the amalgam.free group

and the group with the extensive number of amalgam res­torations was 24 nmol / l (4.8 ).1g /1 ). Provided that the differ­ence is solely caused by the dental amalgam, that the HgU approximately represent the 24 hr hour excretion (Zander 1.'1 (II. 1990). and that ro ughly 5()<)lo of the mercury is recov­ered in urine, the mean HgU indicate a daily exposure of approximately 10 ).1g / day. The present results thus suggest that individ uals with mally amalgam restorations, i.e. , more than 36 restored surfaces, are exposed to 10-12).lg Hg/day in addit ion .

Ack nowledgem en ts The authors would like to than k A/S Borregard,

Sarpsborg, Norway, for financia l support.

References

Aaseth, J. & L. Sa rregttrd: K viksolv-risikoevaluering i yrkesmedisi n. Tid,·kr. Nor. I..(Fgi1orell. 1989. 109. 1642 1644.

Abraham. J. E .. C W. Svare & c. W. Frank: The efTe<:t of dental amalgam restorat ions on blood mercury levels. J. Dent . Res. 1'184. 63, 7 1 73 .

Akesson, I. . A. Schutz. R. AtleweU, S. Skerfving & P. GI~ntz : Status of mercury and selenium in dental personnel: Impact of amalgam work and own fillings . Arch. Environ. Health 1'191 , 46.102- 109.

Berglund. 1\ .: Estimation by a 24-hoUf study of the dai ly do,e of in tra-oral mcrcury vapor inhaled after release from den tal amalgam. J . Deni. Res. 1')90. 69, 1646--- 1651.

Berglund, A. , L. Pohl. S. Olsson & M. Ekrgma n: Dt.-tennination of release of intra-oral mercury vapor from amalgam. J . Denl. Res. 1988, 67, 1235 1242.

Berl in. M.: Mercury. In: lIandhook Oil/he toxicology o/mi"/als. Eds.: L. Friberg. G. F. Norberg & V.B. Vouk . El,evier / Nonh Holland Biomedical Press. Amsterdam, 1986. 385-445.

Birke, G .. A. G. lohnels, L. O. Plant in, B. SjOstrand. S. Skerfving & 1'. Westermark: Stud ie, on humans exposed to methyl mercury through fish consumption. A reh. Enriron. Heallh 1972, 2.'i . 77- 9 I .

Bjertness, E.: A multidimcllsioflill s/Ildy of denial h<'ulth in ",hall Norwegian adults. Thesis. Oslo, University of Oslo. 1990.

Brunc. D. & D. M. Evje: Man's mercury loading from a denta l amalgam . Sci. To/al Em/roll. 1985. 44,51 63.

•

DE:-.JTA L AMALGAM AND MERCLiR Y 313

Huish, E., H. Emteborg, D. C. Baxter. W. Frech, D. Ellingsen &. y. Thom8!)SCn: Speciation of mercury in human whole blood by ga~ chromatography with a microwave·induced plasma emission detector system following complexometric extraction and bu· tylation . Analyst 1992, in press.

Cherian, M, G., J . B. Hursh. T. W. Clarkson & J. Allen: Radioactive mercury dimibution in biological fluids and excretion in human subject. after inhalation of mercury vapor, A reI! , EIIl'iroll, Health 1978.33. 109- 114_

Clarkson. T. W. L Friberg, J, B. Hursh & M. Kylander: The prNIiet ion of Intake of mercury ~'apor fro m amalgams. In: 8,(}­IIIXlwl momloring of IIwwls. Eds: T W C larkson. L. Friberg. O. F Nordberg & P Sager, Plenum Press, Ne" Yo rk. 1999. 247 264 .

Cleneh· Aas. J. S .. larssen. A. Iktrtono\a. M . J. Aames. K_ Myhre. C. C. Christensen. 1 L Neo;lein. Y. Thomassen & F LeI')": nIt"

h!'ullh ej(ec/J oj ImfJk fHJllouion <15 mf'lI1"red in the Vtl/erl'lIgu IIrl'lI u{ Os/v, Lillestrom. t\ational ln S!l\ute for Air Research. 1991. Report 7/9 1.

Ebb<.;~I:HJ. U_. N, Gunder>t:n & T. Torgrimsen: A ,impk melhod for Ih ~ d~lcrmilla1ioll of inorganic mercury and melhyl mercury in biological sam ples by Ilameless (,lOmB: absorption. A lomic A h­.<orf"iMf Ne"'s/t'lIl'r 1975. 14, 142 144,

1·T)'kholm. K 0. : On mercury from denta l arnalga m_ Its roxlc and a llcrgic efTa;ts and some comments o n occupational hygiene. l"hcsis. /le/(l Qii,mlo/. seom! 1957. 15. Supplemcnl 22.

Gay. D D . R_ D. COll & J. W Rell1h:,rdt: Chewmg releases mercury from fjllmg.~_ wncl'l 1979, L 985- 91S6,

I hJ r~h. 1. IL T W_ Oarhon. M . G. Chenan. J . J, Vostal & R. V~ndeTMalle : Clc~rance or mercury \ Hg·197. Ilg·203) vapor in· hJ.l ed hy hllman SUbjl'CtS. Arch. HIII'iron. Hellith 1976. 31. 302 309.

Jok Slad . A.: Kvik hol\' cbponering II\' tannlcgcr_ NOr.\k Tamfleg!/"r· ,'If Tid. 1987. 97. 49S- 507 ,

Kersh~w. T. G .. T. \.Ii. Clarkson & r. Dhahi r: The rel~lionship belWtc" blood kvcls and dose of methylmercury in man . Arch 1:;",,,," IIMflh 1980. 3~ . 28 36.

KrOncke. A. K. Ott . A Pet;;chelt. K II Schaller. \i . S/xesi & H. Valentin: ObeT die Queck si lberkonlenlral lonen 111 Blut und Urm \on Personen mit und ohne Amalgamffillungen. P ilch Zulll/' lir:11 Z 1980. 3~. S03 80H.

Lie. A .. \I. Gunderscn & K. J. Korsgaard: Mercury in urine Sell. ~ge and gC()graphk different'e~ m a reference popullition. SC(Jud. I W"rk_ Fm-mJ/J_ 11 .. 01111 1982. 8, 129 I D

M ~ckcrt. Jr . .1_ R.: I'actors affecting e~limation of dental amalgam mcr.:ur> exposure from measurements of mercury vapor levels in mtra,0131 and expired air. j DI.'1Ji Hn, 1987. 66. 1775 1780

Marek. :\1.. The release of mercury from dental amalgam: The mechanism a rid in ritra lestrng. J /Jen/. RF$. 1989. 69. 1167- 1174.

,\ 1 athiesen. B.: /m"ok 01' MI'. kadnrlum og ko-ikks(l/,' fra mrrillgsm/J· ler. Oslo: Stalc Pollulion Control Authorily. 19SO. Report 8180.

\1 01111 . M . Mercur) release from dent,,1 am .. lgam 111 man_ A thesis. S",'J Pmt J_ 1990. Supplement 71

Okllbe. T.: ~lcrcury 111 Ihe st ruclure of dental amalgam_ INn/. .\luIU

1987. 3. 1 8.

Olsson, S. C & M. Bergman: Lener to the editor. J . Dt m. Re$. 1987, 66.1288 1289.

Olsson. S. C. A. Berglund. L. Pohl &. M. Bergman: Model of mercury vapor transport from amaillam restorations in the oral cavity, J, Delli. Rf$. 1989, 68.504-508.

Olstad, M . L.. R. I. Ho lland. N. Wandel & A. Hensten· Penersen: Correlation between amalgam ~storations and mercury concen­trations in urine. J . Denf Res. 1987. 68, 1179- 1182,

Olt . K. H .. F. Loh. A. Kroncke. K. H . Schaller. H. Valentin & D. Weille. Zur Q uecksi lberbelastung durch AmalgamfUllungen Duch lillmo':II. Z. 1984. 39, 199 2(15.

PalleMn. J E .. B. G. Weissberg & P. J_ IXnnison: Mercury 111

human breath from dental amalgams. 811ft F.m·;rmf_ Conltlm To>.'i<'ol. 1985. 34. 459 468.

Phillips. R W_ (Ed .): SkIlJIJI'r$:' $(ielf((' or ,/elfl,,1 tmlla;"ls, 10th ed W S, Saunders Publications, Philudelph ia. 198ti_

I'iolrowski. J. K .. B. P. TroJanow~h & r M. Mogilnlcka: ExcreliOIl kinetics and variability of llrlnary mercur> in workers expo,cd to mercury vapour, fill. Arl'i! Oaflp. t:nl'l'rolJ, Hea/li! 1975. 35. 245 256.

Siillsten. G .. L. Barregi\rd & T. 0 slerberg: Tandgni~,hJ\g hos am~ l ·

gamoorare en orsak til hi)! k\' ichil\"cruhondring? Liif..lItlil/llllJ· gelf 1991, 88. 232 233_

Skerf~lng. So: Methylmercury nposurc. mercury Ie\cls In blood ami haIr and health status irl S"'"Cde~ consuming contaminated fi sh To).";co/vg) 1974. 2.3---23.

Skerfl'lI1g. S. & M Berhn Nordlsk .. expcrtgruppen lOr gransviirde<.­do~umemation 59 Oorgalll~ k hlckSlhcr. A/'befe och lliif<" 1985. Report 20.

Snapp. K . R .. D. B, Boyer. L. C. Peterson & c. W. S'Me: T he conlribution ur denial amalgam to mercury in blood . .I. Delli . R",T. 19!;9. 68, 780 785 ,

S\'arc. C. W. L C Peterson. J, \.Ii, Reinhard t. D. R. Royer. C. W. I-nl1lk . D_ D_ Gay & R_ O. Co~: The effect of dental amalgam~ on mercury Ind, In expired air. J 1)('lIf_ Re.f. 1981. 60 . 1668 1671.

S~\crscn . T L, : Omjaug {Il' lte/s..,iJfkll ''rdrl,rem/'' IIIllgml'lltllbd{/Jl· nin1: i den 1J0rske bejolkmng 0:;10: Slale Pollullon Control Auth· orilY. 1982. Report 35 82.

Vimy. M J. & F L. Lorschcider: lnlra.oral ai r mereury relea".:d from dent~1 "malgam. J Dml H,'s. 1985a , 64.1069 1071.

Vim}". M. J . & E L Lor!.Cheider: Seria l measure1llent~ of inmH.ral air mercury. E~timalion of dai l} dose from .. kntal amalg:\m. j

Delli Re.,_ 19R5b. 64. 1072 1075, Vim)', M. J_ & F. L Lorscheidcr: Dellt.tI am,tlgam m~rellry d:lily

d{)s<: e'aimalcd from intra·oral I',!por mca~uremcnlS: a predictor of mercury accllmulalion in human \i s~ uc~ , j, TrlIce UemelJl E'(p, Mrd.1 990, 3. 11 1 123.

Lander. D .. 1..: . Ewers. I. Freier. 1:.. Jermann. S Westerweller & A. ll rock haus: Untersuchungen t llr Queck ~ll berbe la st u ng der 8cvol· kcrung I. Quc:cksl lberko rl-zent rallonen 1m Urin bei Normalper· sonen ZbI. Hyg. 1990. 190. 315-24

Wo rld He"hh O rganiza tion Recommmt!ed IWII/l/r·basrd Imlll_< /If

ocrllplrliunal npOS'tr" w /relll'l mewl, Genna. 1990. Technical Report 647. 102 114.