Factors influencing mercury evaporation rate from dental amalgam fillings
Lars Bjorkman^ and Birger Lind^ 'Department of Environmentai Hygiene, and ^Institute of Environmental Medicine, Karoiinska institutet, Stockhoim, Sweden
Bjorkman L, Lind B: Factors itifluencing mercury evaporation rate from dental amalgam fillings. Scand J Dent Res 1992: 100: 354-60. Factors influencing mercury evaporation from dental amalgam fillings were ' ^ studied in 11 volunteers. Air was drawn from the oral cavity for 1 min and continuously analyzed with a mercury detector. In six volunteers the median unstimulated evaporation rate was 0.1 ng Hg/s, range 0.09-1.3 ng Hg/s. After chewing gum for 5 min the highest evaporation rate was 2.7 ng Hg/s. Chewing paraffin wax gave only a small increase in evaporation rate. Changes in airflow rates between 1.5 and 2.5 1/min during the 1 min sampling did not change the amount of mercury drawn from the oral cavity. Sampling with different mouthpieces and closed mouth was compared to open mouth sampling with a thin plastic tube. It was found that the latter method could result in lower values for some volunteers due to simultaneous mouth breathing. After placing individual plastic teeth covers in the mouth, the intraoral evaporation of mercury decreased immediately by 89-100% of previous levels. This technique could be used to detect mercury evaporation from separate amalgam fillings or to reduce the intraoral mercury vapor concentration. Rinsing the mouth with heated water for 1 min increased the mean evaporation rate by a factor of 1.7 when the water temperature increased from 35'C to 45°C.
Dental amalgam is an alloy consisting of silver, tin, copper, zinc, and metallic mercury. The mercury content is about 50% by weight (1). Since dental amalgam was introduced in the middle of the 19th century, its use has been questioned on several occasions due to its mercury content and the potential hazard to health (2-5). Recent studies show that mercury evaporates from amalgam fillings in the mouth (6-14). Furthermore, correlation was found between the number of teeth surfaces filled with amalgam and mercury concentrations in urine (15, 16), and plasma (17) as well as in occipital cortex, pituitary gland and renal cortex from autopsy material (18-21). For the general population the mean daily uptake of mercury vapor from dental amalgam fillings has been estimated to 3.8-21 l^g/day (22). Mechanical factors (chewing, toothbrushing) may increase the evaporation rate severalfold (9, 10, 13). The methods used for determining mercury evaporation vary between different studies (23) and the detailed infiuence of different sampling techniques, including design of mouthpieces and different types of chewing stimulation, have not previously been evaluated. The aim of this work is to study some factors
Key words: amalgam; mercury vapor L. Bjorkman, Departmenf of Environmental Hygiene, Karoiinska Institutet, S-104 01 Stockhoim, Sweden Accepted for publication 20 October 1991
(fiow rate, temperature, and chewing stimulation) which may influence the mercury evaporation from dental amalgam fillings. The infiuence of different sampling techniques on the estimation of mercury evaporation rate is studied as well. iVIaterial and methods
Eleven volunteers, 5 men and 6 women, aged 25-49 yr with 6-19 teeth with amalgam fillings (6-53 teeth surfaces filled with dental amalgam) participated in the experiments (see Table 1). Four volunteers (number 4, 5, 7 and 8) were dentists and were also occupationally exposed to mercury. Before determining the basal values (the unstimulated evaporation rate, expressed as ng/s, of mercury vapor in the oral cavity) no food intake, drinking, chewing or smoking was allowed for 8 h before sampling. The volunteer was instructed to rinse the mouth for 10 s with tap water, measured to about pH 7, heated to a temperature of 36°C, and to swallow the saliva immediately before the sampling procedure. Intraoral air was drawn for 1 min with a fiow rate of 2.2 1/min using the Dtype mouthpiece (Fig. 1) while the volunteer was breathing through the nose, holding the tube plate
Mercury evaporation from dental amalgam
355
Table 1 Sex, age, number of amatgatn surfaces and number of teeth witti amalgatn jiltitigs jor the test subjects Teeth with amalgam No. of amalgam Subject Sex Age •No. surfaces fillings 1 2
3 4
5 6 7
8 9 10 11
F M M M F F M M F F F
31 31 49 29 34 44 37 34 30 25 38
31 11
53 13 50 43 15
13
8 16 10
,1ft . 15 , 9
6
6
11 12 10
'8
7
' ^
against the lips and keeping the lips tight around the tube in order to draw room air only via the nose into the mouth. The amount of mercury vapor drawn from the oral cavity was determined by a mercury monitor (model 1235, LDC Analytical, Riviera Beach, FL, USA) detecting the absorbance at 254 nm wavelength. A recorder (model 56, Perkin-Eliner Corp., Norwalk, CT, USA) and a digital A / D multimeter (model 197, Keithley Instruments, Inc., Cleveland, OH, USA) were connected to the mercury monitor. The equipment setup is shown in Fig. 2 and described elsewhere (13). The fiow rate was controlled by a rotameter (Rota model LSF 0.01., L 6.3/250 Titan, Kontram - S. Holm AB, Bandhagen, Sweden). One milliliter of mercury chloride (HgCU) standards of 15, 25, 50, 100, 150, 200 and 300 ng Hg/ ml and blanks were analyzed by reducing Hg-^ to Hg" with SnCli (cold vapor technique) in a reaction vessel and the generated mercury vapor was determined (24, 25). A diagram and equation for the amount of mercury in standards (x) versus area (mVs) under the signal curve (y), calculated by using a digital planimeter (Placom KP-90, Koi-
Fig. 2. Equipment setup for mercury vapor determination. Air from the reaction vessel (A) or mouthpiece (B) is drawn through two ice-cooled traps (C) and a drying filter (D) before passing through the mercury monitor (E), which is connected to a recorder (F) and a digital voltmeter (G). After analyses the mercury vapor is absorbed in a iodine carbon filter (H). A Oow meter (I) and a pump (J) are also connected. Arrows indicate How direction.
zumi), was established. The equation of the regression line was y = 1.62x + 2.56, n = 25, r = 0.999. The mercury level in room air was checked by connecting an iodine carbon filter, which adsorbs mercury vapor, in place of the mouthpiece. The signal read out level from the mercury monitor did not change when the filter was connected. When results are expressed in amount of mercury released from the oral cavity per time unit, the area under the signal curve from the intraoral sampling is related to the area under the signal curves of a mercury standard (20 ng Hg) decreased with the mean area under the signal curves of the blanks analyzed the same day. When results are given in "mean deflection, |iV" the area under the signal curve (cm-) is divided with recorder chart speed PEAK AREA mVs
40
A
SrANDARD
•
NO. 2
30-
20
10-
100 mm
Fig. 1. Cross-sectional view of mouthpieces. Left part is connected to a tube leading to the mercury monitor. Right part is placed intraorally. Length of intraoral part is 18 mm for A, B, and C, and 50 mm for D, and outer diameter is 10, 15, 40 and 15 mm for A, B, C, and D, respectively. For additional information see ref. 13.
I I 1 1.5 2 FLOW RATE, t/min
2.5
Fig. 3. Infiuence of different flow rates on recorder signal peak area (mVs). Analyses of standard solution of 5 ng Hg and intraoral sampling for volunteer No. 2 during 1 min, bars are I SD, n = 4. Signal from standard is total signal, hence not decreased with blank signal.
356
Bjorkman and Eind
(cm/min) and time (min) of sampling period and multiplied by the ratio of recorder full-scale voltage (in |iV), and the recorder chart paper span (25 cm). The following equation was used: Dwean = A X (V X t) " ' X U,^^ X 1" '
D^ean- mean defiection (|iV) A: area under signal curve (cm-) v: recorder chart speed (cm x min"') t: time of sampling period (min) Un,.,^: recorder full scale voltage (|iV) 1: recorder chart paper span (25 cm) With a fiow rate of 2.2 1/min a mean defiection of 100 i^V corresponds approximately to a mercury evaporation rate of 0.05 ng Hg/s. Flow rates
A standard of 5 ng Hg as well as intraoral sampling for one person (No. 2) was analyzed with fiow rates of 1.0, 1.5, 2.0, and 2.5 1/min. For each fiow rate four intraoral samplings were carried out, two with increasing fiow rate sequence and two with decreasing fiow rate sequence. ;. ,
The D-type was previously used by ARONSSON et al. (14). The sampling procedure was the same as in "basal value" determination, but for the plastic tube E the volunteer was instructed to keep the mouth open, and the tube was turned forward and backward in the mouth. In the sequence A-B-CD-E each mouthpiece was tested for 1 min, repeated three times. Between each sampling the volunteer rinsed the mouth for 10 s with tap water heated to a temperature of 36°C, and swallowed the saliva immediately before the test period. Temperature
The volunteers kept 30 ml tap water heated to 15-45°C in the mouth for 60 s without rinsing movements, spat out the water and swallowed the saliva immediately before start of the sampling, carried out the same way as in "basal value" determination. Each temperature experiment was performed three times. The water temperature was checked with a standard thermometer immediately before each experiment. The mercury concentration in the tap water was also determined. Chewing stimulation
Mouthpieces
Four different Plexiglas mouthpieces. A, B, C, and D, (Fig. 1) and a plastic tube, E, with a diameter of 8 mm (inside diameter 5 mm), were compared.
BASAL EVAPORATION RATE ng Hg/s 1.5-r
1.2-
• No. 1 • No. 2 A No. 3 O
.. ,
,-
•
No. 4
B
D No. 5
'
'
••
;
:
Two types of chewing stimulation were compared, sugar-free chewing gum ("V6", Fertin Laboratories A/S Vejle, Denmark) and 1 g piece of paraffin wax (Dentocult paraffin, Orion Diagnostica, Helsinki, Finland). First the basal value was determined, after which the volunteer was told to chew paraffin wax for 5 min at a frequency of 90 chewings/min, while breathing through the nose with the mouth closed. Immediately before sampling of intraoral air the volunteer removed the piece of paraffin wax, rinsed the mouth with 36°C tap water and swallowed the saliva. Before the stimulation with chewing gum started, it was checked that the evaporation rate had decreased to the basal value.
A No. 6 .9
Experiment with plastic covers
.6-
.3-
A
1 0
0 20 40 60 NUMBER OF AMALGAM SURFACES
Fig. 4. Basal evaporation rate (ng Hg/s) for six volunteers in relation to number of amalgam surfaces. Three 1-min samplings were carried out for eaeh test subject.
: ;
Individual plastic teeth covers (Bioplast, Scheu Dental, Iserlohn, Germany) were made for two of the volunteers (Nos. 2 and 6) on stone gypsum models constructed from alginate impressions of their dental arches. The teeth and about 2 mm of the marginal gingiva were covered. Evaporation rates were determined before and after the plastic covers were placed in the mouth in two separate experiments. Results When standard solutions containing 5 ng Hg were analyzed with different air fiows, the area under
Mercury evaporation from dental amalgam the signal curve increased in proportion to the inverse fiow rate (Fig. 3). For intraoral sampling the same relation was seen in fiow rates exceeding 1.5 1/min and it was indicated that in the interval 1.5-2.5 1/min, the amount of mercury vapor drawn from the oral cavity per minute was constant. If the sampling time was extended over a longer period of time (5 min), no decrease in intraoral mercury vapor evaporation was observed. The mercury concentration in the tap water was found to be less than 0.04 ng Hg/g. No significant difference in evaporation rate, expressed as mean defiection in HV, was found between rinsing with tap water (475 + 95 laV, n = 3) or deionized water (517 + 65 HV, n = 3). The median value for the basal evaporation rate for the six volunteers was 0.1 ng/s (range 0.09-1.3 ng/s) (Fig. 4). One volunteer (No. 5) had a remark-
357
ably high basal evaporation rate, 1.3 ng Hg/s, and 2.7 ng Hg/s after chewing gutn. An exatnple of recordings of a blank, a standard solution containing 20 ng Hg, and intraoral sampling is shown in Fig. 5. The test of different mouthpieces showed there was a slight variation in the amount of mercury vapor sampled from the oral cavity due to different tnouthpieces. In four of the volunteers mouthpiece C gave the highest value. For volunteers Nos. 6 and 7 the values were considerably lower for the plastic tube ("E") than for the other mouthpieces (Fig. 6). The relative standard deviations for each mouthpiece and volunteer were calculated. The mean of all volunteers' relative standard deviation (%) for each mouthpiece was for A: 18.4, B: 17.4, C: 15.2, D: 15.4, and for E: 40.4. The evaporation rate increased when the temper-
- --•
",: «i
B :
"
'
/ " : — : : : : • !
!
•
-
;
_
Fig. 5. Recordings of mercury vapor from intraoral sampling (A-C), standard of 20 ng Hg (D) and blank (E). Chart speed 60 mm/min, recorder full-scale voltage 2 mV (A-D) and 1 mV (E). Air How 2.2 1/min.
358 :'
Bjorkman and Lind MEAN DEFLECTION
400
ABCDEABCDEABCDEABCDEABCDE 2 6 7 89 VOLUNTEliR NUMBER
Fig. 6. Intraoral mercury sampling with different mouthpieces (A-D) and plastic tube (E) for test subjects Nos. 2, 6, 7, 8, and 9. Mercury evaporation expressed as mean recorder deflection ((JV) during the 1-min sainpling period, bars are 1 SD for n = 3.
ature of the rinsing tap water increased (Fig. 7). The evaporation increased with temperature from 15 to 45°C. When the temperature of the rinsing tap water was increased 10°C (from 35°C to 45°C) the mean intraoral mercury evaporation under these experimental conditions increased by a factor of 1.7. Chewing paraffin wax increased the evaporation rate for one of the volunteers (No. 3) by a factor of 1.9, while no detectable increase was found in the other two (Nos. 4 and 5). After chewing gum, the evaporation rates in volunteers 3, 4, and 5 increased by a factor of 6.7, 1.5, and 2.1, respectively (Fig. 8). In the first experiment with plastic covers the intraoral mercury evaporation, expressed as rnean deflection, for volunteer No. 6 decreased from 0.82
MEAN DEFLECTION
mV to 0.14 mV and in a second experimetit the evaporation decreased from 0.75 mV to 0.06 mV when the plastic covers were placed in the mouth. For volunteer No. 2 the mean deflection in the experiments decreased from 0.16 mV and 0.12 mV to not detectable deflections. Discussion In the present study determinations of intraoral evaporation rate of mercury from dental amalgam fillings are, in general, in agreement with other studies (8, 12, 14). In some studies (7, 9-11, 13) the evaporated intraoral mercury is expressed as amount of mercury per volume and therefore has to be recalculated with regard to sampling time in order to make a comparison possible (12, 23). Using direct monitoring of mercury evaporation without collecting the sampled mercury on, for example, gold foil or silver wool, makes it possible to record the mercury signal cotitinuously during the period of sampling. Mercury vapor released in the oral cavity has to be expressed as amount of mercury released per time unit, as also concluded by BERGLUND et al. (12). When sampling intraoral air the sampling period can be extended more than 60 s without decrease in the mercury vapor concetitration in the oral cavity. No change in the amount of mercury drawn from the oral cavity per titiie unit was observed, not even when different flow rates were used, in agreement with other studies (8, 9, 12). Flow rates between 1.5 and 2.5 1/rnin were found to be suitable, but generally 2.2 1/min was used. At 254 nm wavelength the absorption of light is also influenced by water vapor, SO,, HjS, and some aromatic hydrocarbons (e.g., benzene) (26). Such interferences seem most unlikely, since ati amalEVAPORATION RATE ng Hg/s ^ BASAL VALUE BH PARAFFIN • CHEWING GUM
0
10 20 30 40 50 TEMPERATURE OF WATER
Fig. 7. Mercury evaporation from dental amalgam fillings for volunteers No. 2 (age 31), No. 8 (age 34), No. 10 (age 25), and No. 11 (age 38) after rinsing with 30 ml water of different temperatures (15, 25, 35, and 45'C). Mean for three samplings for each test subject and temperature. Mercury evaporation expressed as mean recorder deflection (|.iV) during the 1-min sampling period.
3
4
5
VOLUNTEER NUMBER
Fig. 8. Comparison of mercury evaporation (ng Hg/s) from dental amalgam fillings for test subjects Nos. 3, 4, and 5 after no chewing stimulation, chewing paraffin wax or gum for 5
Mercury evaporation from dental amalgam gam-free volunteer did not give a detectable signal (13). Furthermore, sampling with plastic covers in the mouth decreased the signal considerably. Hence, this strongly indicates that all mercury vap o r sampled in the oral cavity originates directly from the amalgam fillings. To exclude that part of the sampled mercury vapor was generated before the sampling period started, the volunteer was told to rinse the mouth with water at mouth temperature (36°C) and swallow immediately before the sampling. When comparing rinsing with tap water or deionized water n o significant difference in evaporation rate was found. However, it is not excluded that a lower pH might influence the evaporation rate. The possibility that the ambient laboratory air would be mercury contaminated and infiuence the determinations is excluded as control of the signal level with the iodine carbon filter was performed and no decrease of the signal could be detected. In the experiment with different mouthpieces the mean deflection ()aV) was slightly higher for the Ctype with an outer diameter of 40 mm. Using this the volunteers had to open the mouth widely compared to when the other mouthpieces were used. Possibly, the teeth and the amalgam fillings were less covered by the intraoral mucous membranes when the C-type mouthpiece was used and, therefore, the intraoral mercury evaporation was higher. The considerable difference between mouthpiece C and the plastic tube E for volunteers Nos. 6 and 7 could be explained by simultaneous mouth breathing and inhalation of mercury vapor, which could not be controlled when the sampling was carried out with the mouth open. It is possible that erroneously low and irreproducible results could be obtained by sampling with the mouth open. A relationship was also found between temperature of the rinsing water and evaporation rate. However, in a recent study it was shown that intake of a cup of coffee did not increase the evaporation rate (14). This conflicting result could be explained by the differences between drinking hot liquid and rinsing the mouth with heated water. This experiment shows that there seems to be an age-related factor since younger volunteers released more mercury than the older ones (Fig. 7). One explanation for the age relationship could be that less mercury evaporates from old fillings than from new ones. According to the volunteers the fillings were made several years ago. However, the material is too small to draw conclusions. To compare the effect of different chewing stimulation a regular sugar-free chewing gum was compared with paraffin wax. When chewing paraffin wax the evaporation rate did not increase in two of the three volunteers. After chewing gum, all
359
evaporation rates increased. Differences in chewing resistance could be an explanation for this finding. One volunteer (No. 5) had a very high basal evaporation rate and after chewing gum the evaporation rate increased by a factor of 2.1, For another volunteer (No. 3) with even more amalgam fillings (53 filled surfaces) the basal evaporation rate was only 15% compared to No. 5, but after chewing gum the evaporation rate increased by a factor of 6.7. This corresponds to 48% of the chewing stimulated evaporation rate of No. 5. An explanation for the very high basal evaporation rate of volunteer No. 5 could be that occlusal stress (e.g., teeth grinding or clenching) before the intraoral sampling increased the basal evaporation rate. Infiuence of a possible age factor has also to be considered. The fact that this volunteer was a dentist and also occupationally exposed to mercury, however, cannot explain the high basal evaporation rate. The experiment with plastic covers showed that the intraoral evaporation of mercury decreased immediately by 89-100'^ of previous levels when they were placed in the mouth. This could possibly be used when detecting mercury evaporation from separate amalgam fillings or to reduce temporarily the intraoral mercury vapor concentration. In conclusion, it was found that mercury evaporation from dental amalgam fillings can be as high as 1.3 ng/s before chewing and 2.7 ng/s after chewing gum. Different designs of mouthpieces for intraoral sampling may result in a slight variation of the amount of mercury vapor sampled from the oral cavity, but if sampling is carried out with the mouth open it is possible that erroneously low and irreproducible results can be obtained. Heating the amalgam fillings by rinsing with heated water increases the evaporation rate, chewing paraffin wax does not increase the evaporation rate as much as chewing gum. The day-to-day variation in the levels of mercury evaporation in the oral cavity needs further investigation as well as estimating the uptake of Hg vapor from dental amalgam fillings in oral mucous membranes, saliva and gastric mucosa, A possible protective infiuence from the saliva as well as influence of age of the amalgam fillings also have to be investigated. Acknowledgment - Funds from the Karolinska Institutet are gratefully acknowledged.
References 1. JoRGENSEN KD. Dentale amalgamer. In: HOLST JJ, NYGAARD OsTBY B, OsvALD O, eds. Nordi.sk klini.sk odontologi.
Copenhagen; Forlaget for faglitteratur, 1968; 1; 7A-1, 1^7. 2. SOCIAL.STYRI:LSEN. Kvicksilver/amalgam, halsorisker. Stockholm; Socialstyrelsen redovisar, 1987; 10.
360
Bjorkman and Lind
3. STOCK A. Die chronische Quecksilber- und Amalgam-vergiftung. Zahnaerztl Rtttidsch 1939; 10: 371-7, 403-7. 4. ENWONWU CO. Potential health hazard of use of mercury in dentistry: critical review of the literature. Environ Res 1987; 42: 257-74. 5. WEINER J, NYLANDER M , BERGLUND F. Does mercury from
amalgam restorations constitute a health hazard? Sci Total Environ 1990; 99: 1-22. 6. GAY DD, Cox RD, REINHARDT JW. Chewing releases mercury from fillings. Lancet 1979; No. 8123: 985-6. 7. SvARE CW, PETERSON LC. REINHARDT JW, et al. The effect
17. MoLiN M, MARKLUND S, BERGMAN B, BERGMAN M , STEN-
MAN E. Plasma-selenium, glutathione peroxidase in erythrocytes and mercury in plasma in patients allegedly subject to oral galvanism. Seand f Dent Res 1987; 95: 328-34. 18. NYLANDER M , FRIBERG L, LIND B. Mercury concentrations
in the human brain in relation to exposure from dental amalgam fillings. Swed Dent J 1987; 11: 179-87. 19. ScHiELE R. Quecksilberabgabe aus Amalgam und Quecksilberablagerung in Organismus und Toxikologische Bewertung. In: KNOLLE G , ed. Amalgam - Pro und Contra. Cologne: Deutsche Artzte-Verlag, 1988; 123-31.
of dental amalgams on mercury levels in expired air. f Dent Res 1981; 60: 1668-71.
20. DRASCH G , SCHUPP I, GUNTHER G . Einfluss von Amalgam-
8. ABRAHAM JE, SVARE CW, FRANK CW. The effect of dental
rinde. In: ANKE M , BAUMAN W, BRAUNLICII H , BRUCKNER
amalgam restorations on blood mercury levels. J Dent Res 1984; 63: 71-3. 9. PATTERSON JE, WEISSBERG BG, DENNLSON PJ. Mercury in
human breath from dental amalgams. Bull Environ Contain To.xicol 1985; 34: 459-68. 10. ViMY MJ, LoRSCiiEiDER FL. Intra-oral air mercury released from dental amalgam. J Dent Res 1985; 64: 1069-71. 11. LANGWORTH S, KOLBECK K - G , AKESSON A. Mercury expo-
sure from dental fillings. II. Release and absorption. Swed Denl f 1988; 12: 71-2. 12. BERGLUND A, POHL L, OL.SSON S, BERGMAN M . Determina-
tion of the rate of release of intra-oral mercury vapor from amalgam. J Dent Res 1988; 67: 1235-42. 13. ARONSSON A M , LIND B, NYLANDER M , NORDBERG M .
Dental amalgam and mercury. Biol Metals 1989; 2: 25-30. 14. BERGLUND A. Estimation by a 24-hour study of the daily dose of intra-oral mercury vapor inhaled after release from dental amalgam. J Dent Res 1990; 69: 1646-51. 15. OLSTAD M L , HOLLAND RI, WANDEL N , HENSTEN PET-
TERSEN A. Correlation between amalgam restorations and mercury concentrations in urine. J Dent Res 1987; 66: 1179-82. 16. LANGWORTH S, ELINDER C-G, AKESSON A. Mercury expo-
sure from dental fillings. 1. Mercury concentrations in blood and urine. Swed Dent f 1988; 12: 69-70. . .: ..
fiillungen auf die Quecksilberkonzentration in der NierenC, GROPPEL B, GRUN M , eds. Proceedings of the 6th Inter-
national Trace Element Symposium in Leipzig. Jena: Der Friedrich Schiller Universitat, 1989; 1653-9. 21. NYLANDER M , FRIBERG L, EGGLESTON D, BJORKMAN L.
Mercury accumulation in tissues from dental staff and controls in relation to exposure. Swed Dent J 1989; 13: 235-43. 22. WHO. Environmental Health Criteria US: Inorganic mercury Geneva: World Health Organization, 1991. 23. 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 toxle metals. New York: Plenum Press, 1988; 247-64. 24. MAGOS L. Selective atomic-absorption determination of inorganic mercury and methylmercury in undigested biological samples. Analyst 1971; 96: 847-53. 25. MAGOS L, CLARKSON TW. Atomic absorption determination of total, inorganic, and organic mercury in blood. J Assoc Off Anal Chem 1972; 55: 966-71. 26. LDC Analytical, Riviera Beach FL. Instruction Manual, Mercury Monitor Model 1235, Sept. 1980.
Lars Bjorkman^ and Birger Lind^ 'Department of Environmentai Hygiene, and ^Institute of Environmental Medicine, Karoiinska institutet, Stockhoim, Sweden
Bjorkman L, Lind B: Factors itifluencing mercury evaporation rate from dental amalgam fillings. Scand J Dent Res 1992: 100: 354-60. Factors influencing mercury evaporation from dental amalgam fillings were ' ^ studied in 11 volunteers. Air was drawn from the oral cavity for 1 min and continuously analyzed with a mercury detector. In six volunteers the median unstimulated evaporation rate was 0.1 ng Hg/s, range 0.09-1.3 ng Hg/s. After chewing gum for 5 min the highest evaporation rate was 2.7 ng Hg/s. Chewing paraffin wax gave only a small increase in evaporation rate. Changes in airflow rates between 1.5 and 2.5 1/min during the 1 min sampling did not change the amount of mercury drawn from the oral cavity. Sampling with different mouthpieces and closed mouth was compared to open mouth sampling with a thin plastic tube. It was found that the latter method could result in lower values for some volunteers due to simultaneous mouth breathing. After placing individual plastic teeth covers in the mouth, the intraoral evaporation of mercury decreased immediately by 89-100% of previous levels. This technique could be used to detect mercury evaporation from separate amalgam fillings or to reduce the intraoral mercury vapor concentration. Rinsing the mouth with heated water for 1 min increased the mean evaporation rate by a factor of 1.7 when the water temperature increased from 35'C to 45°C.
Dental amalgam is an alloy consisting of silver, tin, copper, zinc, and metallic mercury. The mercury content is about 50% by weight (1). Since dental amalgam was introduced in the middle of the 19th century, its use has been questioned on several occasions due to its mercury content and the potential hazard to health (2-5). Recent studies show that mercury evaporates from amalgam fillings in the mouth (6-14). Furthermore, correlation was found between the number of teeth surfaces filled with amalgam and mercury concentrations in urine (15, 16), and plasma (17) as well as in occipital cortex, pituitary gland and renal cortex from autopsy material (18-21). For the general population the mean daily uptake of mercury vapor from dental amalgam fillings has been estimated to 3.8-21 l^g/day (22). Mechanical factors (chewing, toothbrushing) may increase the evaporation rate severalfold (9, 10, 13). The methods used for determining mercury evaporation vary between different studies (23) and the detailed infiuence of different sampling techniques, including design of mouthpieces and different types of chewing stimulation, have not previously been evaluated. The aim of this work is to study some factors
Key words: amalgam; mercury vapor L. Bjorkman, Departmenf of Environmental Hygiene, Karoiinska Institutet, S-104 01 Stockhoim, Sweden Accepted for publication 20 October 1991
(fiow rate, temperature, and chewing stimulation) which may influence the mercury evaporation from dental amalgam fillings. The infiuence of different sampling techniques on the estimation of mercury evaporation rate is studied as well. iVIaterial and methods
Eleven volunteers, 5 men and 6 women, aged 25-49 yr with 6-19 teeth with amalgam fillings (6-53 teeth surfaces filled with dental amalgam) participated in the experiments (see Table 1). Four volunteers (number 4, 5, 7 and 8) were dentists and were also occupationally exposed to mercury. Before determining the basal values (the unstimulated evaporation rate, expressed as ng/s, of mercury vapor in the oral cavity) no food intake, drinking, chewing or smoking was allowed for 8 h before sampling. The volunteer was instructed to rinse the mouth for 10 s with tap water, measured to about pH 7, heated to a temperature of 36°C, and to swallow the saliva immediately before the sampling procedure. Intraoral air was drawn for 1 min with a fiow rate of 2.2 1/min using the Dtype mouthpiece (Fig. 1) while the volunteer was breathing through the nose, holding the tube plate
Mercury evaporation from dental amalgam
355
Table 1 Sex, age, number of amatgatn surfaces and number of teeth witti amalgatn jiltitigs jor the test subjects Teeth with amalgam No. of amalgam Subject Sex Age •No. surfaces fillings 1 2
3 4
5 6 7
8 9 10 11
F M M M F F M M F F F
31 31 49 29 34 44 37 34 30 25 38
31 11
53 13 50 43 15
13
8 16 10
,1ft . 15 , 9
6
6
11 12 10
'8
7
' ^
against the lips and keeping the lips tight around the tube in order to draw room air only via the nose into the mouth. The amount of mercury vapor drawn from the oral cavity was determined by a mercury monitor (model 1235, LDC Analytical, Riviera Beach, FL, USA) detecting the absorbance at 254 nm wavelength. A recorder (model 56, Perkin-Eliner Corp., Norwalk, CT, USA) and a digital A / D multimeter (model 197, Keithley Instruments, Inc., Cleveland, OH, USA) were connected to the mercury monitor. The equipment setup is shown in Fig. 2 and described elsewhere (13). The fiow rate was controlled by a rotameter (Rota model LSF 0.01., L 6.3/250 Titan, Kontram - S. Holm AB, Bandhagen, Sweden). One milliliter of mercury chloride (HgCU) standards of 15, 25, 50, 100, 150, 200 and 300 ng Hg/ ml and blanks were analyzed by reducing Hg-^ to Hg" with SnCli (cold vapor technique) in a reaction vessel and the generated mercury vapor was determined (24, 25). A diagram and equation for the amount of mercury in standards (x) versus area (mVs) under the signal curve (y), calculated by using a digital planimeter (Placom KP-90, Koi-
Fig. 2. Equipment setup for mercury vapor determination. Air from the reaction vessel (A) or mouthpiece (B) is drawn through two ice-cooled traps (C) and a drying filter (D) before passing through the mercury monitor (E), which is connected to a recorder (F) and a digital voltmeter (G). After analyses the mercury vapor is absorbed in a iodine carbon filter (H). A Oow meter (I) and a pump (J) are also connected. Arrows indicate How direction.
zumi), was established. The equation of the regression line was y = 1.62x + 2.56, n = 25, r = 0.999. The mercury level in room air was checked by connecting an iodine carbon filter, which adsorbs mercury vapor, in place of the mouthpiece. The signal read out level from the mercury monitor did not change when the filter was connected. When results are expressed in amount of mercury released from the oral cavity per time unit, the area under the signal curve from the intraoral sampling is related to the area under the signal curves of a mercury standard (20 ng Hg) decreased with the mean area under the signal curves of the blanks analyzed the same day. When results are given in "mean deflection, |iV" the area under the signal curve (cm-) is divided with recorder chart speed PEAK AREA mVs
40
A
SrANDARD
•
NO. 2
30-
20
10-
100 mm
Fig. 1. Cross-sectional view of mouthpieces. Left part is connected to a tube leading to the mercury monitor. Right part is placed intraorally. Length of intraoral part is 18 mm for A, B, and C, and 50 mm for D, and outer diameter is 10, 15, 40 and 15 mm for A, B, C, and D, respectively. For additional information see ref. 13.
I I 1 1.5 2 FLOW RATE, t/min
2.5
Fig. 3. Infiuence of different flow rates on recorder signal peak area (mVs). Analyses of standard solution of 5 ng Hg and intraoral sampling for volunteer No. 2 during 1 min, bars are I SD, n = 4. Signal from standard is total signal, hence not decreased with blank signal.
356
Bjorkman and Eind
(cm/min) and time (min) of sampling period and multiplied by the ratio of recorder full-scale voltage (in |iV), and the recorder chart paper span (25 cm). The following equation was used: Dwean = A X (V X t) " ' X U,^^ X 1" '
D^ean- mean defiection (|iV) A: area under signal curve (cm-) v: recorder chart speed (cm x min"') t: time of sampling period (min) Un,.,^: recorder full scale voltage (|iV) 1: recorder chart paper span (25 cm) With a fiow rate of 2.2 1/min a mean defiection of 100 i^V corresponds approximately to a mercury evaporation rate of 0.05 ng Hg/s. Flow rates
A standard of 5 ng Hg as well as intraoral sampling for one person (No. 2) was analyzed with fiow rates of 1.0, 1.5, 2.0, and 2.5 1/min. For each fiow rate four intraoral samplings were carried out, two with increasing fiow rate sequence and two with decreasing fiow rate sequence. ;. ,
The D-type was previously used by ARONSSON et al. (14). The sampling procedure was the same as in "basal value" determination, but for the plastic tube E the volunteer was instructed to keep the mouth open, and the tube was turned forward and backward in the mouth. In the sequence A-B-CD-E each mouthpiece was tested for 1 min, repeated three times. Between each sampling the volunteer rinsed the mouth for 10 s with tap water heated to a temperature of 36°C, and swallowed the saliva immediately before the test period. Temperature
The volunteers kept 30 ml tap water heated to 15-45°C in the mouth for 60 s without rinsing movements, spat out the water and swallowed the saliva immediately before start of the sampling, carried out the same way as in "basal value" determination. Each temperature experiment was performed three times. The water temperature was checked with a standard thermometer immediately before each experiment. The mercury concentration in the tap water was also determined. Chewing stimulation
Mouthpieces
Four different Plexiglas mouthpieces. A, B, C, and D, (Fig. 1) and a plastic tube, E, with a diameter of 8 mm (inside diameter 5 mm), were compared.
BASAL EVAPORATION RATE ng Hg/s 1.5-r
1.2-
• No. 1 • No. 2 A No. 3 O
.. ,
,-
•
No. 4
B
D No. 5
'
'
••
;
:
Two types of chewing stimulation were compared, sugar-free chewing gum ("V6", Fertin Laboratories A/S Vejle, Denmark) and 1 g piece of paraffin wax (Dentocult paraffin, Orion Diagnostica, Helsinki, Finland). First the basal value was determined, after which the volunteer was told to chew paraffin wax for 5 min at a frequency of 90 chewings/min, while breathing through the nose with the mouth closed. Immediately before sampling of intraoral air the volunteer removed the piece of paraffin wax, rinsed the mouth with 36°C tap water and swallowed the saliva. Before the stimulation with chewing gum started, it was checked that the evaporation rate had decreased to the basal value.
A No. 6 .9
Experiment with plastic covers
.6-
.3-
A
1 0
0 20 40 60 NUMBER OF AMALGAM SURFACES
Fig. 4. Basal evaporation rate (ng Hg/s) for six volunteers in relation to number of amalgam surfaces. Three 1-min samplings were carried out for eaeh test subject.
: ;
Individual plastic teeth covers (Bioplast, Scheu Dental, Iserlohn, Germany) were made for two of the volunteers (Nos. 2 and 6) on stone gypsum models constructed from alginate impressions of their dental arches. The teeth and about 2 mm of the marginal gingiva were covered. Evaporation rates were determined before and after the plastic covers were placed in the mouth in two separate experiments. Results When standard solutions containing 5 ng Hg were analyzed with different air fiows, the area under
Mercury evaporation from dental amalgam the signal curve increased in proportion to the inverse fiow rate (Fig. 3). For intraoral sampling the same relation was seen in fiow rates exceeding 1.5 1/min and it was indicated that in the interval 1.5-2.5 1/min, the amount of mercury vapor drawn from the oral cavity per minute was constant. If the sampling time was extended over a longer period of time (5 min), no decrease in intraoral mercury vapor evaporation was observed. The mercury concentration in the tap water was found to be less than 0.04 ng Hg/g. No significant difference in evaporation rate, expressed as mean defiection in HV, was found between rinsing with tap water (475 + 95 laV, n = 3) or deionized water (517 + 65 HV, n = 3). The median value for the basal evaporation rate for the six volunteers was 0.1 ng/s (range 0.09-1.3 ng/s) (Fig. 4). One volunteer (No. 5) had a remark-
357
ably high basal evaporation rate, 1.3 ng Hg/s, and 2.7 ng Hg/s after chewing gutn. An exatnple of recordings of a blank, a standard solution containing 20 ng Hg, and intraoral sampling is shown in Fig. 5. The test of different mouthpieces showed there was a slight variation in the amount of mercury vapor sampled from the oral cavity due to different tnouthpieces. In four of the volunteers mouthpiece C gave the highest value. For volunteers Nos. 6 and 7 the values were considerably lower for the plastic tube ("E") than for the other mouthpieces (Fig. 6). The relative standard deviations for each mouthpiece and volunteer were calculated. The mean of all volunteers' relative standard deviation (%) for each mouthpiece was for A: 18.4, B: 17.4, C: 15.2, D: 15.4, and for E: 40.4. The evaporation rate increased when the temper-
- --•
",: «i
B :
"
'
/ " : — : : : : • !
!
•
-
;
_
Fig. 5. Recordings of mercury vapor from intraoral sampling (A-C), standard of 20 ng Hg (D) and blank (E). Chart speed 60 mm/min, recorder full-scale voltage 2 mV (A-D) and 1 mV (E). Air How 2.2 1/min.
358 :'
Bjorkman and Lind MEAN DEFLECTION
400
ABCDEABCDEABCDEABCDEABCDE 2 6 7 89 VOLUNTEliR NUMBER
Fig. 6. Intraoral mercury sampling with different mouthpieces (A-D) and plastic tube (E) for test subjects Nos. 2, 6, 7, 8, and 9. Mercury evaporation expressed as mean recorder deflection ((JV) during the 1-min sainpling period, bars are 1 SD for n = 3.
ature of the rinsing tap water increased (Fig. 7). The evaporation increased with temperature from 15 to 45°C. When the temperature of the rinsing tap water was increased 10°C (from 35°C to 45°C) the mean intraoral mercury evaporation under these experimental conditions increased by a factor of 1.7. Chewing paraffin wax increased the evaporation rate for one of the volunteers (No. 3) by a factor of 1.9, while no detectable increase was found in the other two (Nos. 4 and 5). After chewing gum, the evaporation rates in volunteers 3, 4, and 5 increased by a factor of 6.7, 1.5, and 2.1, respectively (Fig. 8). In the first experiment with plastic covers the intraoral mercury evaporation, expressed as rnean deflection, for volunteer No. 6 decreased from 0.82
MEAN DEFLECTION
mV to 0.14 mV and in a second experimetit the evaporation decreased from 0.75 mV to 0.06 mV when the plastic covers were placed in the mouth. For volunteer No. 2 the mean deflection in the experiments decreased from 0.16 mV and 0.12 mV to not detectable deflections. Discussion In the present study determinations of intraoral evaporation rate of mercury from dental amalgam fillings are, in general, in agreement with other studies (8, 12, 14). In some studies (7, 9-11, 13) the evaporated intraoral mercury is expressed as amount of mercury per volume and therefore has to be recalculated with regard to sampling time in order to make a comparison possible (12, 23). Using direct monitoring of mercury evaporation without collecting the sampled mercury on, for example, gold foil or silver wool, makes it possible to record the mercury signal cotitinuously during the period of sampling. Mercury vapor released in the oral cavity has to be expressed as amount of mercury released per time unit, as also concluded by BERGLUND et al. (12). When sampling intraoral air the sampling period can be extended more than 60 s without decrease in the mercury vapor concetitration in the oral cavity. No change in the amount of mercury drawn from the oral cavity per titiie unit was observed, not even when different flow rates were used, in agreement with other studies (8, 9, 12). Flow rates between 1.5 and 2.5 1/rnin were found to be suitable, but generally 2.2 1/min was used. At 254 nm wavelength the absorption of light is also influenced by water vapor, SO,, HjS, and some aromatic hydrocarbons (e.g., benzene) (26). Such interferences seem most unlikely, since ati amalEVAPORATION RATE ng Hg/s ^ BASAL VALUE BH PARAFFIN • CHEWING GUM
0
10 20 30 40 50 TEMPERATURE OF WATER
Fig. 7. Mercury evaporation from dental amalgam fillings for volunteers No. 2 (age 31), No. 8 (age 34), No. 10 (age 25), and No. 11 (age 38) after rinsing with 30 ml water of different temperatures (15, 25, 35, and 45'C). Mean for three samplings for each test subject and temperature. Mercury evaporation expressed as mean recorder deflection (|.iV) during the 1-min sampling period.
3
4
5
VOLUNTEER NUMBER
Fig. 8. Comparison of mercury evaporation (ng Hg/s) from dental amalgam fillings for test subjects Nos. 3, 4, and 5 after no chewing stimulation, chewing paraffin wax or gum for 5
Mercury evaporation from dental amalgam gam-free volunteer did not give a detectable signal (13). Furthermore, sampling with plastic covers in the mouth decreased the signal considerably. Hence, this strongly indicates that all mercury vap o r sampled in the oral cavity originates directly from the amalgam fillings. To exclude that part of the sampled mercury vapor was generated before the sampling period started, the volunteer was told to rinse the mouth with water at mouth temperature (36°C) and swallow immediately before the sampling. When comparing rinsing with tap water or deionized water n o significant difference in evaporation rate was found. However, it is not excluded that a lower pH might influence the evaporation rate. The possibility that the ambient laboratory air would be mercury contaminated and infiuence the determinations is excluded as control of the signal level with the iodine carbon filter was performed and no decrease of the signal could be detected. In the experiment with different mouthpieces the mean deflection ()aV) was slightly higher for the Ctype with an outer diameter of 40 mm. Using this the volunteers had to open the mouth widely compared to when the other mouthpieces were used. Possibly, the teeth and the amalgam fillings were less covered by the intraoral mucous membranes when the C-type mouthpiece was used and, therefore, the intraoral mercury evaporation was higher. The considerable difference between mouthpiece C and the plastic tube E for volunteers Nos. 6 and 7 could be explained by simultaneous mouth breathing and inhalation of mercury vapor, which could not be controlled when the sampling was carried out with the mouth open. It is possible that erroneously low and irreproducible results could be obtained by sampling with the mouth open. A relationship was also found between temperature of the rinsing water and evaporation rate. However, in a recent study it was shown that intake of a cup of coffee did not increase the evaporation rate (14). This conflicting result could be explained by the differences between drinking hot liquid and rinsing the mouth with heated water. This experiment shows that there seems to be an age-related factor since younger volunteers released more mercury than the older ones (Fig. 7). One explanation for the age relationship could be that less mercury evaporates from old fillings than from new ones. According to the volunteers the fillings were made several years ago. However, the material is too small to draw conclusions. To compare the effect of different chewing stimulation a regular sugar-free chewing gum was compared with paraffin wax. When chewing paraffin wax the evaporation rate did not increase in two of the three volunteers. After chewing gum, all
359
evaporation rates increased. Differences in chewing resistance could be an explanation for this finding. One volunteer (No. 5) had a very high basal evaporation rate and after chewing gum the evaporation rate increased by a factor of 2.1, For another volunteer (No. 3) with even more amalgam fillings (53 filled surfaces) the basal evaporation rate was only 15% compared to No. 5, but after chewing gum the evaporation rate increased by a factor of 6.7. This corresponds to 48% of the chewing stimulated evaporation rate of No. 5. An explanation for the very high basal evaporation rate of volunteer No. 5 could be that occlusal stress (e.g., teeth grinding or clenching) before the intraoral sampling increased the basal evaporation rate. Infiuence of a possible age factor has also to be considered. The fact that this volunteer was a dentist and also occupationally exposed to mercury, however, cannot explain the high basal evaporation rate. The experiment with plastic covers showed that the intraoral evaporation of mercury decreased immediately by 89-100'^ of previous levels when they were placed in the mouth. This could possibly be used when detecting mercury evaporation from separate amalgam fillings or to reduce temporarily the intraoral mercury vapor concentration. In conclusion, it was found that mercury evaporation from dental amalgam fillings can be as high as 1.3 ng/s before chewing and 2.7 ng/s after chewing gum. Different designs of mouthpieces for intraoral sampling may result in a slight variation of the amount of mercury vapor sampled from the oral cavity, but if sampling is carried out with the mouth open it is possible that erroneously low and irreproducible results can be obtained. Heating the amalgam fillings by rinsing with heated water increases the evaporation rate, chewing paraffin wax does not increase the evaporation rate as much as chewing gum. The day-to-day variation in the levels of mercury evaporation in the oral cavity needs further investigation as well as estimating the uptake of Hg vapor from dental amalgam fillings in oral mucous membranes, saliva and gastric mucosa, A possible protective infiuence from the saliva as well as influence of age of the amalgam fillings also have to be investigated. Acknowledgment - Funds from the Karolinska Institutet are gratefully acknowledged.
References 1. JoRGENSEN KD. Dentale amalgamer. In: HOLST JJ, NYGAARD OsTBY B, OsvALD O, eds. Nordi.sk klini.sk odontologi.
Copenhagen; Forlaget for faglitteratur, 1968; 1; 7A-1, 1^7. 2. SOCIAL.STYRI:LSEN. Kvicksilver/amalgam, halsorisker. Stockholm; Socialstyrelsen redovisar, 1987; 10.
360
Bjorkman and Lind
3. STOCK A. Die chronische Quecksilber- und Amalgam-vergiftung. Zahnaerztl Rtttidsch 1939; 10: 371-7, 403-7. 4. ENWONWU CO. Potential health hazard of use of mercury in dentistry: critical review of the literature. Environ Res 1987; 42: 257-74. 5. WEINER J, NYLANDER M , BERGLUND F. Does mercury from
amalgam restorations constitute a health hazard? Sci Total Environ 1990; 99: 1-22. 6. GAY DD, Cox RD, REINHARDT JW. Chewing releases mercury from fillings. Lancet 1979; No. 8123: 985-6. 7. SvARE CW, PETERSON LC. REINHARDT JW, et al. The effect
17. MoLiN M, MARKLUND S, BERGMAN B, BERGMAN M , STEN-
MAN E. Plasma-selenium, glutathione peroxidase in erythrocytes and mercury in plasma in patients allegedly subject to oral galvanism. Seand f Dent Res 1987; 95: 328-34. 18. NYLANDER M , FRIBERG L, LIND B. Mercury concentrations
in the human brain in relation to exposure from dental amalgam fillings. Swed Dent J 1987; 11: 179-87. 19. ScHiELE R. Quecksilberabgabe aus Amalgam und Quecksilberablagerung in Organismus und Toxikologische Bewertung. In: KNOLLE G , ed. Amalgam - Pro und Contra. Cologne: Deutsche Artzte-Verlag, 1988; 123-31.
of dental amalgams on mercury levels in expired air. f Dent Res 1981; 60: 1668-71.
20. DRASCH G , SCHUPP I, GUNTHER G . Einfluss von Amalgam-
8. ABRAHAM JE, SVARE CW, FRANK CW. The effect of dental
rinde. In: ANKE M , BAUMAN W, BRAUNLICII H , BRUCKNER
amalgam restorations on blood mercury levels. J Dent Res 1984; 63: 71-3. 9. PATTERSON JE, WEISSBERG BG, DENNLSON PJ. Mercury in
human breath from dental amalgams. Bull Environ Contain To.xicol 1985; 34: 459-68. 10. ViMY MJ, LoRSCiiEiDER FL. Intra-oral air mercury released from dental amalgam. J Dent Res 1985; 64: 1069-71. 11. LANGWORTH S, KOLBECK K - G , AKESSON A. Mercury expo-
sure from dental fillings. II. Release and absorption. Swed Denl f 1988; 12: 71-2. 12. BERGLUND A, POHL L, OL.SSON S, BERGMAN M . Determina-
tion of the rate of release of intra-oral mercury vapor from amalgam. J Dent Res 1988; 67: 1235-42. 13. ARONSSON A M , LIND B, NYLANDER M , NORDBERG M .
Dental amalgam and mercury. Biol Metals 1989; 2: 25-30. 14. BERGLUND A. Estimation by a 24-hour study of the daily dose of intra-oral mercury vapor inhaled after release from dental amalgam. J Dent Res 1990; 69: 1646-51. 15. OLSTAD M L , HOLLAND RI, WANDEL N , HENSTEN PET-
TERSEN A. Correlation between amalgam restorations and mercury concentrations in urine. J Dent Res 1987; 66: 1179-82. 16. LANGWORTH S, ELINDER C-G, AKESSON A. Mercury expo-
sure from dental fillings. 1. Mercury concentrations in blood and urine. Swed Dent f 1988; 12: 69-70. . .: ..
fiillungen auf die Quecksilberkonzentration in der NierenC, GROPPEL B, GRUN M , eds. Proceedings of the 6th Inter-
national Trace Element Symposium in Leipzig. Jena: Der Friedrich Schiller Universitat, 1989; 1653-9. 21. NYLANDER M , FRIBERG L, EGGLESTON D, BJORKMAN L.
Mercury accumulation in tissues from dental staff and controls in relation to exposure. Swed Dent J 1989; 13: 235-43. 22. WHO. Environmental Health Criteria US: Inorganic mercury Geneva: World Health Organization, 1991. 23. 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 toxle metals. New York: Plenum Press, 1988; 247-64. 24. MAGOS L. Selective atomic-absorption determination of inorganic mercury and methylmercury in undigested biological samples. Analyst 1971; 96: 847-53. 25. MAGOS L, CLARKSON TW. Atomic absorption determination of total, inorganic, and organic mercury in blood. J Assoc Off Anal Chem 1972; 55: 966-71. 26. LDC Analytical, Riviera Beach FL. Instruction Manual, Mercury Monitor Model 1235, Sept. 1980.
