Mercury Determination in Hair of Malaysian Fishermen by Neutron

Activation Analysis S. B. SARMANI,* A. Z. KIPRAWl, AND R. B. ISMAJL

Department of Nuclear Science, National University of Malaysia, 43600 Bangi, Malaysia

ABSTRACT

Fish has been known as a source of nonoccupational mercury exposure to fish-consuming population groups. In this study, hair samples collected from fishermen and their families residing in an industrialized area in Penang and a nonindustrialized area in Tereng- ganu were analyzed for mercury by neutron activation. The range, arithmetic mean, geometric mean, and median of the mercury con- cent-rations for the groups in Penang and in Terengganu were 0.45-16.68, 3.61, 3.49, and 2.96 and 6.79-18.31, 12.08, 11.69, and 12.05 mg/kg, respectively. Somewhat lower values than from the Penang group were fotmd in a group from Selangor consisting mainly of office workers. The group in Penang took about 40-100 g of f ish/d, whereas the group in Terengganu consumed twice as much. This shows that hair mercury levels depend on a fish consumption pat- tern, and not on the location of the population. The levels of mercury found in this study were similar to those reported bv other workers for fish-consuming population groups worldwide.

Index Entries: Neutron activation analysis; mercury; hair.

INTRODUCTION

Hair has been known as one of the deposition sites for the excretion of metals from the body. Since mercury is excreted from the body partly by deposition in hair, the analysis of hair provides a convenient tool for estimating body burdens of mercury, and thereby also provides an indi- rect measure of dietary intake. A weekly mercury, intake of 0.2 rag, the

*Author to whom all correspondence and reprint requests should be addressed.

Biological Trace Element Research Editor: G. N. Schrauzer �9 1994 by Humana Press Inc.

435

436 Sarrnani, Kiprawi, and lsmail

Provisional Tolerance Weekly Intake (PTWI) set by WHO and FAO, is expected to result in the hair mercury contents of 5-6 m g / k g (1). An excellent review of the applications of hair analysis in biomedical and environmental studies has been presented by Chatt and Katz (2). Airey (3), who analyzed hair samples from 32 locations in 13 countries for mer- cury, found that mean hair concentrations were sig-nificantly different for the group that ate fish once a month or less (1.4 mg/kg) , once a fortnight (1.9 mg/kg) , once a week (2.5 mg/kg) , and once or more daily (11.6 mg/kg) . Several other reports have also shown the positive relationships between fish consumption and the concentration of mercury in hair (4-6).

Because Malaysia is peninsular state, fishing is an important indus- try, providing people with employment as well as supplying them with protein from the fish. The consumption of fish comprises about 70% of the protein intake of the population, and the daily per capita fish con- sumption is 100 g. The average mercury levels in common commercial sea fish markets in Malaysia range from 0.06 to 0.42 m g / k g fresh weight (7). With such a high intake of fish, it is apparent that fish is a possible source of nonoccupational mercury exposure to the general population. Earlier work by Silvalingam and Sani (8) showed that the mean mercury level of Malaysian fishermen was 8 mg/kg . This suggested that their weekly intake of mercury exceeded the PTWI level.

In this study, hair samples were collected from fishermen and their families residing in two fishing villages in Penang and Terengganu. For comparison purposes, hair samples were also obtained from a group of nonfishermen in Selangor. The aim of this study was to identify certain population groups that may have ingested mercury higher than the PTWI as set by the WHO. This s tudy is part of a coordinated research program on assessment of environmental exposure to mercury in selected h u m a n populations as studied by nuclear and other techniques organized by the IAEA. The aim of the project was to evaluate the expo- sure of selected population groups to mercury and methyl-mercury, and to estimate potential health risks in these groups.

METHODS

Sample Preparation Hair samples from fishermen's families were collected from resi-

dents of the Kuala Juru Village in the state of Penang on the West Coast, and the Chender ing Village in the state of Terengganu on the East Coast of Peninsular Malaysia. Penang is one of the most industrialized states in Malaysia, and the Kuala Juru Village is surrounded by a large industrial estate. There are about 100 households in the village with a total popu- lation of about 500. The Chendering Village, on the other hand, is located on a prestigious, idyllic coastline free from any industrial activity. Most of the adult male residents in the two villages are fishermen. Hair sam- ples were also collected from residents in Kajang, a small urbanized area with light industrial activities in the state of Selangor. This population

Mercury in Hair of Malaysian Fishermen 437

group represents the nonfishermen's community. Partict~tars of each donor, such as name, sex, age, occupation, and dietary habits, were gath- ered and compiled in a questionnaire form.

Hair samples were washed according to the IAEA washing proce- dures: Wash hair successively in acetone, thrice ha water, and once more in acetone. Add sufficient amounts of the solvents to cover the sample entirely. At each wash, allow the sample to stand at room temperature for 10 rain with constant stirring. After each wash, decant the liquid, and add flesh solvent. Carry out the washing in the dust-free environment.

Sample Irradiation

About 150 mg of each sample were packed in a precleaned polyeth- ylene vial. Triplicate analysis was carried out for each sample. Standard was prepared from standard chemical solution pipeted onto ashless filter paper and packed in a polyethylene bag after careful drying. About 50 gL of 10% solution of thioacetamide was pipeted onto the filter paper to prevent mercury losses. Samples, standards, and blanks were irradiated in a PUSPATI TRIGA Reactor for 6 hs in a neutron flux of 2.3 x 1012 n/cm-2/s-1

Measurement of Activities

Mercury was determined using both 197Hg and 203Hg radionuclides. The determination via 77 keV of 197Hg was carried out using an LEPD. The irradiated samples were cooled for 3 d and counted for 30 min. The 279.6 keV of 75Se interfered with the determination of mercury via the 279. 1-keV photopeak of 203Hg, thus a correction for the selenium con- tribution was necessary. The samples were cooled for at least 2 w k prior to counting.

The 7-ray activities of the samples and standards were measured by a high-resolution 7-spectroscopy system comprising a high-purity coaxial N-type germanium detector with a thin Be window (Tenelec), micro- computer-based MCA system with 80386 SX data processor, 8192 chan- nel ADC and MCA Card (Tenelec), and Tc-244 spectroscopy amplifier (Tenelec).

Accuracy and precision of the INAA technique were evaluated by analyzing IAEA standard reference materials MAA-1 (Copepoda), MAA- 2 (fish homogenate), and MAB-3 (fish tissue).

RESULTS

Accuracy of the mercury determination in the standard reference materials is shown in Table 1. Results of the mercury determinat ion in the three population groups are shown in Table 2. Frequency distribution of the mercury concentrations in hair samples of the Kuala Juru fisher- men is shown ha Fig. 1, whereas the distribution according to age groups is shown in Table 3.

438 Sarmani, Kiprawi, and lsrnail

~Iable 1 Results of Mercury Determination in IAEA Reference

Materials (mg/kg)

Reference Material This work Certified value

MAA-1 0.33 + 0.05 0.28 __ 0.01

MAA-2 0.49 _+ 0.14 0.47 4- 0.02

MAB-3 0.54 _ 0.19 0.51

Fable 2 Total Mercury in Hair (mg/kg)

Pulau Pinang Selangor Terengganu (non-fishermen)

No. of sample 106 45 33

Range 0.45-16.87 0.66-6.90 6.79-18.31

Arith. mean 3.6l 3.01 12.08

Median 2.96 2.69 12.05

Geom. mean 3.49 2.95 1 t.69

DISCUSSION

Accuracy and precision of the analytical me thod used for the deter- minat ion of mercury in the IAEA Standard Reference Materials were good, as shown ha Table 1. MercuD, in most marine organisms and hair samples can be analyzed quite easily and conveniently by ins t rumenta l neu t ron activation, so that tedious radiochemical separation steps can be avoided. By employing a low-energy pho ton detector, mercury can be de te rmined m u c h faster through the 69- and 77-keV photopeaks of 197Hg. However , the spectrum in this region is quite complex, and the analysis of spec t rum becomes complicated. Very convenient condi t ions can be aclMeved by analyzing the 279-keV photopeak of 203Hg. The 7 spectra of the irradiated hair samples are less complex than other bio- logical tissues, so that favorable detection limits of mercury in hair can

Mercury in Hair of Malaysian Fishermen

30~-

t

2~

t 204-

i ,

g

T t

tO T

t

T

L I

1 1 i" ~ [ i 5 6 7 8 9 ~0 tl t2

Hair ,mercury concen~raTwn ( mg/kg}

3 4 13 i4 ~@ :6

Fig. 1. The frequency distribution of hair mercury con- centrations for Penang.

439

be achieved. The detection limit of the technique used amounted to 0.01 mg/kg .

'The mercury levels in the three population groups studied were sig- nificantly different (Table 2). However, the narrow range of mercury con- centrations for the population in Selangor lies within the range for the group in Penang. The group in Selangor consists mainly of office work- ers, but their fish-consumption pattern is quite similar to the Penang group. Generally, they consumed fish once a day, but quite different types of fish. Different protein sources were also available to the group in Selangor, since they were economically more soLmd. Other protein sources, such as meat and poultry products, were generally low in mer- cury concentrations.

The frequency distribution of mercury concentrations for the Penang group is shown in Fig. 1. About 30% of the samples contained mercury in higher levels than 5 mg/kg . This suggests that a part of the popula- tion studied might have exceeded the PTWI of mercury. From the infor- mation gathered, some of the donors, especially adult males, consumed as much as 2 kg of fish weekly. This is also reflected in the distribution of mercury, concentrations in relation to age of the donors as shown in Table 3, where generally, mercury hair concentrations in adults were higher as compared to children. Based on the concentrations of mercury

440 Sarmani, Kiprawi, and Ismail

Table 3 Distribution of Mercury According to Age of Donor in

mg /kg

Age Range Arithmetic Mean

1-9 0.45-8.47 3.27

10-19 0.58-10.43 3.16

2 0 - 2 9 0.87-16.87 3.83

30-39 1.04-6.75 3.09

4 0 - 4 9 1.91-12.55 5.22

50-59 1.94-8.33 4.05

>60 1.29-16.65 6.25

in fish marke ted in Malaysia, the PTWI limit may be achieved by con- suming about 500 g of fish weekly (7).

The hair mercury levels of f ishermen and their families in Tereng- ganu were the highest among the groups studied. This is a selected pop- ulation group that consumed be tween 200 and 300 g of fish daily. This group preferred tuna to other types of fish. The Malaysian tunas, such as Euthynnus affinis, Auxis thazarg, and Thunnus tonggol, generally contain higher levels of mercury than other fish. Our s tudy showed that the mer- cury levels ranged from 0.81 to 3.35 m g / k g fresh weight. The lowest con- centrations were found in the white flesh, whereas the highest were found in the liver. Even though this group lives in a nonindustr ia l ized area where mercury pollut ion is expected to be absent, the mercury body burden, as reflected by the hair mercury, was high. This showed that the a m o u n t and type of fish consumed wou ld affect the hair mercury con- centrations.

The concentrat ions of mercury in the hair samples analyzed showed positive correlation with the amoun t of fish consumed, and is within that established by Airey (6). Based on the national per capita consumpt ion of fish at 100 g / p e r s o n / d , the expected hair mercury concentration of the general popula t ion is 6 m g / k g . This s tudy showed that about 30% of the hair samples from Penang, all from the group in Terengganu, and 15% of the samples f rom Selangor contained mercury at more than 6 m g / k g . This also indicated the annual per capita consumpt ion of fish of these subject was 35 kg. Our previous work shoved that the mean concentra-

Mercury in Hair o f Malays ian F i s h e r m e n 441

tion of hair mercury concentrat ion of the general populat ion in the cap- ital city was 6.6 m g / k g (9), suggest ing that the more affluent city, popu- lation consumed a higher amoun t of fish than expected. The hair mercury levels found in this s tudy were within the range of mercury concentrations in the head hair of the fish-eating popula t ion wor ldwide as reported by Airey (3).

SUMMARY

The hair mercury levels found in the Malaysian f ishermen d e p e n d on the amoLmt and type of fish consumed. Industrial activities do not affect the mercury levels in the hair of the popula t ion studied. This sug- gests that industrial activities in Malaysia do not discharge mercury into the environment . The mercury concentrat ion range is within those reported by other researchers for f ish-consuming popula t ion groups worldwide.

ACKNOWLEDGMENT

The research grant p rov ided by the IAEA through Contract N u m b e r MAL 6335/R1/RB is acknowledged. Thanks are due to the Nuclear Energy Unit for providing irradiation facilities, and to Idris Fakih Mokhar, Aziz Ramzan, Hamzah Pardi, Roslan Din, and A h m a d Takim Saring for their competent technical assistance.

REFERENCES

1. International Atomic Energy Agency, NAHRES-13, Vienna, 1992, pp. 1-3. 2. A. Chatt and S. A. Katz, Hair Analysis, VCH Publishers, New York, 1988,

p. 134. 3. D. Airey, Sci. Total Environ. 31, 157 (1983). 4. J. tt. Kyle and N. Ghani, Arch. Environ. Health 37, 266 (1988). 5. J. C. Hansen, H. C. Wulf, N. Kromann, and K. Aloge, Sci. Total Environ. 26,

233 (1983). 6. D. Airey, Environ. Health Prospec. 52, 303 (1983). 7. S. Sarmani, A. K. Wood, Z. Hamzah, and A. A. Majid, f. Radioanal. Nucl.

Chem. 169, 255 (1993). 8. P. M. Sitvalingam and A. B. Sani, Marine Pollut. Ball. 11, 188 (1980). 9. S. Sarmani, J. Radioanal. Nucl. Chem. 110, 627 (1987).