- Email: [email protected]
Biological specimen bank for smelter workers
The Science of the Total Environment, 139/140 (1993) 157-173 Elsevier Science Publishers B.V., Amsterdam
157
Biological specimen bank for smelter workers L. G e r h a r d s s o n a'b, D. B r u n e c, N . - G . L u n d s t r 6 m a, G. N o r d b e r g a a n d P.O. W e s t e r d
~Department of Environmental Medicine, University of Ume~, S-901 87 Ume~, Sweden bDepartment of Occupational and Environmental Medicine, Lund University Hospital, S-221 85 Lund, Sweden 'D.B.C., Dag Brune Consulting, Bakkehaugveien 16, N-0873 Oslo 8, Norway aDepartment of Medicine, University Hospital of Ume~, S-901 87 Ume~, Sweden
ABSTRACT The biological specimen bank was initiated by the department of Environmental Medicine and the department of Medicine at the University of Ume~ in 1975. The aims of the bank are to collect information on trace elements in human organs. Special attention is focused on the influence of occupational exposure. Tissue samples are taken from deceased workers from a copper and lead smelter in northern Sweden. Control specimens have been collected from deceased normal individuals from four control areas. Lung, liver and kidney samples are collected with quartz instruments and stored in quartz ampoules to avoid contamination. Other samples, e.g. bone, brain, fat, hair, heart muscle, nails, skin and stomach are taken with common autopsy instruments and stored in acid-washed polyethene vessels. The samples are stored at -20°C. The elements Sb, As, Cd, Cr, Co, La and Se are analyzed by neutron activation analysis, Pb and Zn by atomic absorption spectrophotometry. The findings over time can be related to a number of factors: normal values in tissues, airborne exposure and causes of death. Special attention is paid variation over time, reevaluation of threshold limit values and risk assessment.
Key words." specimen bank; human samples; smelter workers; metals
INTRODUCTION T h e e s t a b l i s h m e n t o f the biological s p e c i m e n b a n k was m a d e possible t h r o u g h c o o p e r a t i o n b e t w e e n the U n i v e r s i t y o f Ume~t a n d the Swedish mining c o m p a n y , Boliden M i n e r a l . T h e w o r k e r s f r o m w h o m tissue s a m p l e s are collected are e m p l o y e d at a n o n - f e r r o u s c o p p e r smelter in n o r t h e r n Sweden, R 6 n n s k / i r s v e r k e n , the biggest in S c a n d i n a v i a . F r o m the very start o f the smelter's o p e r a t i o n s in 1930, o c c u p a t i o n a l health surveillance was i n t r o d u c ed a n d g r a d u a l l y increased. T h e first o c c u p a t i o n a l physician a n d safety engineer were b o t h e m p l o y e d in 1932.
158
L. G E R H A R D S S O N ET AL.
The exposure of non-ferrous smelter workers is of complicated nature. They are exposed to a number of heavy metals and other substances and agents. The R6nnsk~ir smelter is located near the city of Skellefte~ on the tip of a peninsula protruding into the sea of Bothnia. From the beginning, only ore from the Boliden mine was processed. This sulfide ore had a very complex composition and was characterized by an extremely high concentration of arsenic. New process techniques had to be developed. The plant now integrates smelter and refining processes for ores and other raw materials such as concentrates and scrap. The processes for the main products are shown in Fig. 1. A more detailed technical description of the processes has previously been reported [1,2]. From the late 1920s, thousands of workers have been exposed to different dust concentrations over many years up to their pension age. The exposure may have both acute and long-term health effects. The disease panorama of the initial years was dominated by dermatitis and symptoms from the upper and lower respiratory tract [3,4] which were the main causes of registered sick-leaves [5]. In an extensive medical survey of 1459 workers, Lundgren et al. [6] defined a disease at the smelter 'Morbus R6nnsk~ir'. This disease was characterized by a chronic rhino-pharyngeo-
copper smelting materials
I•
copper sme/tlng malerials • I converting anode c a s t t n g ~
.... l eSOJ
[esPJ
~ bllater copper electrolysis F
"t
) copper cathodes
½pr.iooa,.goid,
""copper
,
" I metets ptent
smelting I • materials [ I copper I Kaldo furnace J
[
~1 selenium plant
Sl,..
I'Paatlnumgr°up of metals ~-~ selenium °'°""o'o""
i4 ..................... "i crude arsenic
- .nicpla.l ] -----J a"e" me'el t*ars°ni°°ata' "t
,
[
"l
plent
'.
,rssn~ tdox~,
' arsenic acid [central t suiphur dioxide gas cleaner ............................. •-~.~.!SU..~.,. ......................
t
........... slag
hi slag fuming' "1 I I lead oxide dust
t lead concentrates
:41-......... ~P1
~ smelting
sulphur [ ~ liquid dioxide plant ~ " sulphur dioxide ~Iclinkerf . . . . . . ~ zinc clinker ~
I
slag
~ slag products
~ lead ~ crude Kaldo lead - [ Kaldo plant ' leSPI #, crude lead
I
..... rtngHreioog
I
.eo
.d
Fig. 1. R6nnskfir processes and products. ESP, electrostatic precipitators. Courtesy of Boliden Mineral AB.
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
] 59
tracheo-bronchitis with atrophy of the mucous membranes in the respiratory tract, emphysema and impaired pulmonary function [6-8]. Arsenic trioxide and sulphur dioxide were alleged to be the most important etiologic factors. The first lung cancer cases related to work were reported by the Company's occupational health physician, Holmqvist, in the 1960s. In a casereferent study from the smelter, Axelson et al. [9] found that the lung cancer mortality was increased about 5-fold and cardiovascular disease about 2-fold among smelter workers. A dose-response relationship to arsenic was observed. Excess figures for tumours, mainly lung cancer and circulatory diseases have later been reported by Wall [10]. In a case-referent study, Pershagen et al. [11] found a multiplicative effect between smoking and arsenic exposure for the development of lung cancer among the smelter workers. Dust measurements were introduced in the early 1940s and continuously extended. Even if the environmental monitoring was of the compliance type (checks against permissible levels) and not of the dose-response type, the total findings over decades display rather broad medical information of the relationship between different types of jobs and manifest diseases. The workers' occupational titles are well defined and the jobs can be classified by work area and followed from assignment. The turnover has been low, many workers have been employed from 20 to 30 years of age and up to retirement. A continuous employment stretch of 30-40 years is not uncommon. After retirement many workers remained living in the smelter surroundings in homes built during their employment period. The collection of samples from deceased smelter workers to the biological specimen bank was started in 1975 by the department of Environmental Medicine and the department of Medicine at the University of Ume~. The aims of the bank are to collect information on trace elements in human organs. Special attention has been focused on the influence of occupational exposure. MATERIALS AND METHODS
Tissue sampling Through close cooperation with the R6nnskfir Company Health Service and the local hospital in Skellefte], tissue samples have been collected in connection with routine necropsies. The number of samples from different organs of smelter workers in the biological specimen bank is presented in Table 1. Control specimens have been collected from deceased normal individuals from four control areas: the city of Stockholm about 800 km from the smelter, the city of Skellefte~, the area surrounding the smelter (Bure~t and
160
L. G E R H A R D S S O N ET AL.
Skelleftehamn) and from a rural area some 50 km from the smelter (Burtr/isk and J6rn). Lung, liver and kidney samples of about 2 g wet weight are collected with quartz instruments to avoid contamination. They are stored in quartz ampoules and freeze dried. Other samples of 20-50 g wet weight (Table 1) are taken with common autopsy instruments and stored in acidwashed polyethene vessels. All samples are stored at the department of Environmental Medicine at -20°C. The autopsies of the urban controls from the city of Stockholm were all performed at the State Institute of Forensic Medicine, Stockholm. All other autopsies have without selection from diagnoses been undertaken at the County Hospital, Skellefte~ by a single technician. Lung samples are taken from the lower part of the right upper lung lobe and from trachea. Liver specimens are taken about 1 cm below the
TABLE 1
Number of samples from different organs of smelter workers and referents in the biological specimen bank Number of samples
Smelter workers
Referents
Brain Bronchus (trachea) Fat Hair Heart muscle Kidney Liver Lung Nails Psoas muscle Renal pelvis Skin Stomach Testis Bones Femur Finger-bone Iliac crest
I 15 115 115 115 115 115 115 115 115 115 115 115 54 12
42 42 42 42 42 42 42 42 42 42 42 42 23 4
94 56 56 56 115 56 56
35 23 23 23 42 23 23
Rib •
Sternum Temporal bone Vertebrae with spinous process
BIOLOGICAL SPECIMEN B A N K FOR SMELTER WORKERS
161
diaphragmatic surface of the right liver lobe and kidney samples from the cortex of the upper part of the right kidney. Fat and skin specimens are collected from the abdominal area, hair from the back of the head and nails from the left hallux. Samples are also collected from the right renal pelvis, left heart ventricle, right psoas muscle, right testis, the lower part of the stomach and the right occipital brain lobe. Bone samples are taken from seven locations: upper part of sternum, right femur about 1 dm above the knee, right iliac crest, right temporal bone, first lumbar vertebra with spinous process, third rib on the right side about 1 dm from sternum and from the middle phalanx of the second finger on the left hand.
Job classification and smoking habits Working histories from the smelter were provided by the company, Boliden Mineral. Data about smoking habits were obtained by questionnaires answered by relatives of the deceased, supplemented with telephone interviews.
Analytical methods Antimony, arsenic, cadmium, chromium, cobalt, lanthanum and selenium were analyzed by neutron activation analysis (NAA). Zinc and lead were determined by atomic absorption spectrophotometry (AAS). From 1975 to 1978 the neutron activation analysis was performed at the Swedish atomic energy station, Studsvik. In 1978, the Technical X-ray Centre (TRC) in Stockholm took over the analyses after the laboratory cessation at Studsvik. In 1982, these laboratory facilities with personnel were transferred to the Swedish Environmental Research Institute (IVL), Stockholm. The NAA technique used by Studsvik was developed by Wester et al. [12] and further modified by Samsahl et al. [13]. TRC and IVL utilized a slightly different technique developed by Sj6strand [14]. The specimens analyzed at Studsvik were irradiated with a thermal neutron flux of 2 × 1013 n cm -2 S -1 for 3 days (R2 reactor, Studsvik). The analyses performed by TRC and IVL were carried out with a thermal flux of 5 × 1012 n cm -2 s -~ for 2 days (Kjeller reactor, Norway). The radionuclides were chemically separated with an automated ion exchange technique and gamma-spectrometric measurements were carried out on the separated fractions [2,13]. Lead (flameless AAS) and zinc (flame AAS) were from 1975 to 1979 analyzed at the department of Environmental Hygiene at the Karolinska Institute, Stockholm. From 1980, lead (flameless AAS) has been analyzed at the department of Analytical Chemistry, University of Ume~ [15]. Zinc (flame AAS) has since 1980 been analyzed at the department of Environmen-
162
L. G E R H A R D S S O N ET AL.
tal Medicine, University of Ume~. The analytical precision, accuracy and quality control of the NAA and AAS methods have previously been described [2]. Special attention was paid to the contamination of the specimens during sampling, storage and the subsequent neutron irradiation procedure. The specimens were collected in pure quartz containers, which had previously been tested for contamination. The release of various elements from the walls of the pure ampoules was, however, found to be negligible. The ampoules were also immersed in liquid nitrogen prior to opening in order to reduce the pressure arising in the sealed ampoules during irradiation at high neutron fluxes due to 'radiolysis' of the biological material. In certain cases samples were also irradiated during frozen conditions in a cryostate inserted in the reactor in order to reduce contamination, volatilization effects, etc. The radioanalytical assay has been described in further detail by Brune et al. 1984 [161.
Statistics Trace metals in human tissues usually follow a skew distribution [17,18]. Based on computer-aided skewness tests, nonparametric statistical methods were applied (Kruskal-Wallis one-way analysis of variance, Mann-Whitney's U-test and Spearman rank-order correlation coefficients). A difference between the studied groups (Kruskal-Wallis, P < 0.05) was evaluated further by Mann-Whitney's U-test (pairwise comparisons). P-values of <0.05 (Mann-Whitney; two-tailed tests) were considered statistically significant. RESULTS Metal concentrations of antimony, arsenic, cadmium, chromium, cobolt, lanthanum, lead, selenium and zinc have been analyzed in part of the material, mainly in lung, liver and kidney tissues [19-23]. The main findings are presented here.
Age, exposure-time and length of retirement The mean age, duration of exposure and length of retirement of smelter workers and controls are presented in Table 2. No significant differences were found in age between the smelter workers and the control groups. About 50% of the smelter workers died from cardiovascular diseases and about 1/3 from malignancies (8-9% from lung cancer and 13-14% from gastrointestinal cancer). Deaths from cardiovascular diseases dominated in the two control groups, 75-80% in the rural group (living some 50 km from the smelter) and 100% in the urban group from the city of Stockholm.
163
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
TABLE 2 Mean-age, exposure-timeand length of retirement in smelter workers and rural (Burtr~iskand J6rn) and urban controls (the city of Stockholm); N, number of workers, mean values 4- S.D.) Group
N
Mean-age (years)
Exposure time (years)
Length of retirement (years)
All smelter workers Rural controls Urban controls
85 15 10
68.3 + 9.1 67.7 4- 10.1 68.8 4- 6.1
31.2 4- 8.2 --
7.3 4- 5.9 --
Tissue concentrations
Median values of Sb, As, Cd, Cr, Co, La, Pb, Se and Zn in lung liver and kidney tissues of smelter workers and controls are summarized in Tables 3 and 4. With the exception of zinc, significantly higher concentrations of these elements were found in the workers' lungs (N = 85) than in non-exposed rural controls from Burtr/isk and J6rn (N = 15, P _< 0.027). Significantly higher lung tissue concentrations of antimony, arsenic and lead were noted among all workers compared to urban controls from the city of Stockholm (N = 10, P < 0.001). The highest lung accumulation was noted for antimony and arsenic; an l l-fold and 6-fold increase, respectively [2,19,20]. The lung tissue concentrations of some metals, e.g. antimony and cobalt, did not decline with time after exposure had ceased among the smelter workers, which indicates a long biological half-time. Metal concentrations related to diagnosis
In the further analysis of the material the workers were subdivided into groups according to diagnoses in autopsy protocols. Workers who died from lung cancer (N = 7) had the lowest lung selenium content relatively to other metals, both compared with other disease categories among the workers (GIcancer, other cancers, cardiovascular diseases, cerebrovascular diseases, other causes) and with the two control groups [22]. The highest lung tissue levels of cadmium were found in the lung cancer group in which, however, smokers and ex-smokers were overrepresented [23]. Several of the lung cancer cases also had from high up to the highest lung concentrations of antimony, arsenic, cadmium, lanthanum and lead. The metal concentrations in liver (metabolism) and kidney (excretion) illustrate the systemic distribution. The highest cadmium concentrations in liver were found in the lung cancer group [23].
164
L. GERHARDSSON ET AL.
TABLE 3 Median values of antimony, chromium, cobalt and lanthanum in lung, liver and kidney tissues in smelter workers and controls. N, number of workers. All values in ppb (/~g/kg wet weight). Qr, quotient of median values between smelter workers and rural controls. QU, quotient of median values between smelter workers and urban controls. From Gerhardsson et al. 1982, 1984, 1988. Element
Antimony Liver N Lung N Kidney N Chromium Liver N Lung N Kidney N Cobalt Liver N Lung N Kidney N Lanthanum Liver N Lung N Kidney N
Smelter workers
Rural controls
12 21 260 85 6 21
7 8 32 15 5 8
11 20 450 85 9 21
4 8 l0 15 3 8
11 20 i7 85 3 21
16 8 7 15
5.0 20 9.6 85 0.2 20
5.5 8 5.0 15 0.1 8
Urban controls
Qr
QU
1.7 19 l0
8.1
13.7
1.2
2.8 199 10
4.1
2.3
3.0
0.7 ll 10
2.4
1.5
3.0
1
8 0.9 10.0 10
1.9
1.0
2.0
DISCUSSION
A multifactorial genesis for the development of lung cancer is supported from these studies [20-23]. In the lung cancer group smoking has probably played a dominant role. High levels of antimony, arsenic, cadmium and several other metals may contribute with different potency. Low levels of
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
165
TABLE 4 Median values of arsenic, cadmium, lead, selenium and zinc in lung, liver and kidney tissues in smelter workers and controls. N, number of workers. Zinc values in ppm (mg/kg wet weight), all other values in ppb (t~g/kg wet weight). Qr, quotient of median values between smelter workers and rural controls. QU, quotient of median values between smelter workers and urban controls. From Wester et al. 1981, Gerhardsson et al. 1986, 1988. Element
Arsenic Liver N Lung N Kidney N Cadmium Liver N Lung N Kidney N Lead Liver N Lung N Kidney N Selenium Liver N Lung N Kidney N Zinc Liver N Lung N Kidney N
Smelter workers
Rural controls
7 21 35 85 9 21
4 8 7 15 5 8
965 21 166 85 17 300 21
350 8 39 15 3025 8
470 21 140 85 348 21
285 8 55 15 235 8
195 20 152 85 495 20
195 8 110 15 755 8
55.7 21 11.5 85 31.0 21
49.8 8 10.2 15 26.3 8
Urban controls
Qr
QU
1.8 5 I0
5.0
7.0
1.8
2.8 79 10
4.3
2.1
5.7
1.6 39 10
2.5
3.6
1.5
1.0 136 10
1.4
1.1
0.7
1.1 12.2 10
1.1 0.9 1.2
166
L. GERHARDSSON ET A L
selenium indicate reduced protection. The smelter workers are also exposed to irritating gases such as SO2. Some air pollutants are of organic origin, e.g. polyaromates such as benzo[a]pyrene. O f the interactions that may occur, some may increase the carcinogenic risks. The biological specimen bank also gives information about tissue concentrations and adverse health effects in nonoccupationally exposed individuals living at different distances from the smelter.
Ambient surveillance Emission considerations placed the plant fairly remote from agriculture and forestry locations, with favourable water and wind directions. The prevailing wind directions are north-south. From the end of the 1960s the ambient environment protection program w a s greatly expanded, as was the production volume. In the period 1970-1984 such environmental investments amounted to SEK 270 million compared to SEK 120 million for improvements of the working environment. The ambient emissions to air during this period were reduced by an average of 60% and those to water by about 90% (Tables 5 and 6). In comparison, air emissions of arsenic were of the magnitude 1-2 tons per day 1930-1949, 0.2-0.8 tons per day 1950-1972 and around 0.1 ton per day in 1984 [24]. In the late 1960s the ambient air levels of arsenic and lead were mostly low 0.5/zg/m 3 (monthly average values) up to 7 km from the smelter [24]. Holmqvist [25] reported lead concentrations in air ranging from 0.3-0.5 tzg/m 3 from dust measurements in domestic areas up to 5 km from the smelter. For other areas with this distance lead varied between 0.2-3.8 /~g/m 3 with a mean value of 0.85 t~g/m 3.
TABLE 5 Emissions to atmosphere from the R6nnsk~ir smelter (tons/year). 1995~ are estimated figures Time period
Dust
Cu
Pb
Zn
Cd
As
Hg
1965-69 1970-73 1978 1984 1986 1995~
3200 2400 1945 857 340 240
190 203 164 42 21 14
684 555 235 103 64 41
488 393 250 122 55 38
13.4 13.5 5.0 2.5 1.4 1.0
154 110 75 39 13 10
3.5 2.4 1.1 0.8 0.6 0.4
20 000 22 000 10100 6750 5400 3900
167
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
TABLE 6 Emissions to water from the R6nnsk/ir smelter (tons/year). 1995 e, figures are estimates. Time period
Cu
Pb
Zn
Cd
As
Hg
1965-69 1970-73 1978 1984 1986 1995 e
165 89 12.5 5.6 3.3 3.3
214 164 11 2.4 2.4 2.4
93 99 37 8 8 8
7.2 7.8 2.2 0.4 0.4 0.4
2078 1846 589 24 24 24
5.2 1.7 0.46 0.079 < 0.1 <0.1
Several studies have reported elevated levels of cadmium and lead in crops and vegetables in the surroundings of the smelter [26]. The main food products consumed, however, are brought in from other parts of the country.
Earlier studies of ambient health effects Kjellstr6m et al. [26] analyzed cadmium in urine from women living at different distances from the smelter (Skelleftehamn, Bure~ and J6rn). Minor differences were noted between the areas. The concentrations were of the same magnitude as from southern Sweden (0.65 tzg Cd/g creatinine). The protein excretion in urine showed no signs of kidney damage. Holmqvist [25] compared the concentrations of arsenic, cadmium, lead and mercury in urine from three groups of women living at different distances and directions from the plant. Only normal values were observed. Lead determinations in blood [25] gave slightly increased mean values for men living in the immediate vicinity of the plant (mean 13.6 #g/100 ml). The work-titles, however, had a greater impact on the blood lead levels than the distance to the smelter. Elinder et al. [27] analyzed the concentrations of cadmium and lead in blood and urine from men and women living in Bure~t and Skelleftehamn. No association was found between a high consumption of locally cultivated food products and elevated blood lead and cadmium levels. In 1979, Rehnlund et al. [28] compared the blood lead levels of children living near the smelter with children living in a distant urban area (Ume~, about 140 km from the smelter). Low concentrations were found in both groups (7.3 ~g/100 ml and 7.7 t~g/100 ml, respectively). In a case-control study, Pershagen [29] found an increased mortality for
168
L. G E R H A R D S S O N ET AL.
respiratory cancer among former employees at the smelter (risk ratio 2.6). The risk was particularly raised for workers earlier exposed to high levels of inorganic arsenic (risk ratio 7.3) and could not be explained by differences in smoking habits. Recently, Pershagen [24] in a case-referent study reported a relative risk of 2.0 (P < 0.05) for lung cancer among men (never employed at R6nnsk~ir) who lived within approximately 20 km from the smelter. The increased mortality could not be explained by smoking habits or occupational background. Emissions, particularly of arsenic from the smelter, may have played a role. No increased levels of cadmium in liver and kidney of cows were found in J6rn and Bure3., compared to controls from southern Sweden. Horses showed a negative correlation between blood and kidney levels of cadmium and increasing distances to the smelter [30].
Future perspectives In an ageing population like the Swedish one, the development of work related diseases may take place long time after retirement. International bodies (ILO, WHO) have discussed both needs and possible ways to prolong occupational health supervision [31]. Current occupational health programs do not include the evaluation of workers' health changes after they have left the plant [32]. Our studies exemplify the difficulties but also the inherent value of such a prolongation. Retrospective studies of this kind are highly promoted by a prospective framework of systematization of the companies' health programs [33]. Earlier efforts of occupational health were directed towards the control of specific and quite obvious diseases from single substances such as silica, lead and mercury. In prospective studies now on the move, the evaluation of the effect of the total multifactorial exposure are brought into sharp focus. Routine periodic medical examinations should as far as possible be made part of life time health promotion programs [34]. The resources needed are, if coordinated, within economically acceptable limits. The occupational health organization can on an individual basis follow the health changes using each employee as his own control [35]. Groups of workers with similar exposure can be continuously followed. Complete work histories can be utilized in occupational epidemiology [36]. The weak link is that the public health care system has no resources or appropriate strategies for a follow-up after employment ceases. The concept, however, behind the revised Swedish public sick and health care legislation is that more pronounced preventive efforts should be enforced. If this preventive-oriented concept is maintained, exposures at work, in home and in traffic, etc. can be evaluated and related to their long-term
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
169
health effects [37]. Great difficulties have of tradition met the evaluation of total diurnal exposure over a life-time. There is a great difference between exposure and individual dose [38]. Earlier calculations have included a wide range of approximations [39]. In classical occupational monitoring programs, air-pollutant measurements were 10-20 times more expensive than biological sampling for the same - - and as a rule - - very restricted information [40]. The occupational hygiene measurements referred to from this smelter were children of its time and dominated by analyses of total aerosol concentrations. Comprehensive occupational hygiene dose evaluations from this period easily reached a cost level of 5-10% of the workers' annual wages. Biological monitoring is, on the other hand, available for only a restricted number of air pollutants and must even in the future be considered as supplemental to environmental monitoring [41]. In the health surveillance and monitoring programs now emerging, an array of instruments and analytical techniques are available for detailed characterization and quantification of pollutants, their diffusion and uptake [42,43]. Use of carefully selected indicator substances characteristic for hygienically important emitting sources, use of cluster analysis and pairwise comparisons, enrichment factors etc. make it possible to differentiate the exposure using basic emission data [44]. Data base management systems for tracking occupational health use personal modules, medical/clinical modules, occupational hygiene/toxicology modules as system functions [45]. Microcomputer programs are available for the evaluation of predictable long-term exposure [34]. Aerodynamic properties of particles and particle agglomerates determine deposition and retention in the respiratory system. Detailed studies of particles, their size, crystal form and chemical composition and properties can be made by using a combination of electron and transmission microscopy, Xray fluorescence and particle induced X-ray emission analysis, (PIXE), electron spectroscopy for chemical analyses (ESCA), etc. Such discrimination techniques which have been used to change particles from the R6nnsk~ir smelter [46] make it possible to compare new combinations of industrial aerosols with older ones for which significant epidemiological data are available [47]. Total suspended particulate in air can now be transformed to size classes and the evaluation of exposure can be made size specific. Particle size selective personal monitors have opened roads to the development of particle size selective threshold limit values for inspirable particulate mass, thoracic particulate mass and respirable particulate mass [48,49]. Such monitors can be carried without disturbing work [50]. Simultaneous determination of multielements in biological tissues, col-
170
L. GERHARDSSONET AL
lected in e.g. a biological specimen bank, can be used to construct an elemental 'fingerprint' for the body or for a critical target organ. The capability exists for studying individual molecules in biological specimens that weigh one quadrillion of a gram. Such element concentration profiles in tissues can be related to corresponding environmental exposure 'fingerprints'. Biological and environmental monitoring overlap, to the benefit of both. Human hair is for example a recording filament which can reflect metabolic changes, changes of concentrations in the body and serum, changes in external exposure and interdependencies over time [51]. Several other types of successful measurements of trace element distributions and variations in tissues and body fluids, covering time-periods of years to decades, have been reported. The findings over time in a biological specimen bank can be related to a number of factors: normal values in tissues, air-borne exposure and causes of death. Special attention is paid to variation over time, reevaluation of threshold limit values and risk assessment. CONCLUSION
New strategies and techniques for multielemental quantification of environmental exposure from work, air, food, water, etc can now be much better related to the uptake of corresponding elements in the body compartments. In other words, available techniques including the use of a biological specimen bank, make prolonged health studies after retirement feasible. The need is expanding. Occupational and environmental medicine here share future responsibilities. ACKNOWLEDGEMENT
Financial support was given by the Swedish Work Environment Fund, Project No. 80/107. REFERENCES 1 L. Gerhardsson, Trace elements in tissues of deceased smelter workers - - relationship to mortality. Medical Dissertation, The University of Ume~, Ume~, Sweden, 1986. 2 L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester, Multielemental assay of tissues of deceased smelter workers and controls. Sci. Total. Environ., 74 (1988) 97-110. 3 G. Inghe and A. Bursell, The R6nnskfir Study 1937. Report of investigation, Stockholm, Sweden (in Swedish, unpublished). 4 T. Sj6strand, Changes in the respiratory organs of workmen at an ore smelting works. Acta Med. Scand., Suppl., 196 (1947) 687-699. 5 .~. Nygren, Historic working environment description - - The R6nnskfir smelter. Report
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
6 7 8 9
10 11
12 13
14 15
16 17 18 19 20
21
22 23
24 25 26
171
of investigation, Boliden Mineral AB, S-932 00 Skelleftehamn, Sweden, 1980 (in Swedish). K.D. Lundgren, N.G. Richtn6r and T. Sj6strand, Morbus R6nnsk/ir. Nord. Med., 46 (1951) 1556-1560 (in Swedish). K.D. Lundgren, Lesions in the respiratory organs of workers at a smelter. Nord. Hyg. Tidskr., 3 (1954) 66-82 (in Swedish with English summary). L.E. Warfvinge, Contribution to the diagnosis of the R6nnsk/ir bronchitis. L/ikartidningen, 65 (1968) 1698-1701 (in Swedish). O. Axelsson, E. Dahlgren, C.-D. Jansson and S.O. Rehnlund, Arsenic exposure and mortality: a case-referent study from a Swedish copper smelter. Br. J. Ind. Med., 35 (1978) 8-15. S. Wall, Survival and mortality pattern among Swedish smelter workers. Int. J. Epidemiol., 9 (1980) 73-87. G. Pershagen, S. Wall, A. Taube and L. Linnman, On the interaction between occupational arsenic exposure and smoking and its relationship to lung cancer. Scand. J. Work. Environ. Health, 7 (1981) 302-309. P.O. Wester, D. Brune and K. Samsahl, Radiochemical recovery studies of a separation scheme for 23 elements in biological material. Int. J. Appl. Radiat. lsot., 15 (1964) 59-67. K. Samsahl, P.O. Wester and O. Landstr6m, An automatic group separation system for the simultaneous determination of a great number of elements in biological material. Anal. Chem., 40 (1968) 181-187. B. Sjfstrand, Simultaneous determination of mercury and arsenic in biological and organic materials by activation analysis. Anal. Chem., 36 (1964) 814-819. E. Lundberg, W. Frech and I. Lindberg, Determination of lead in biological materials by constant-temperature electrothermal atomic absorption spectrometry. Anal. Chimica. Acta, 160 (1984) 205-215. D. Brune, B. Forkman and B. Persson, Nuclear Analytical Chemistry, Studentlitteratur, Chartwell-Bratt Ltd, Lund, (1984)446-453. I. Harding-Barlow, Studies on the trace element content of human tissues. Doctoral dissertation, University of Capetown, 1961. P.O. Wester, Trace elements in heart tissue. Acta Med. Scand., Suppl., 439 (1965) 1-48. P.O. Wester, D. Brune and G.F. Nordberg, Arsenic and selenium in lung, liver and kidney tissue from dead smelter workers. Br. J. Ind. Med., 38 (1981) 179-184. L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester, Antimony in lung, liver and kidney tissue from deceased smelter workers. Scand. J. Work Environ. Health, 8 (1982) 201-208. L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester~ Chromium, cobalt and lanthanum in lung, liver and kidney tissue from deceased smelter workers. Sci. Total. Environ., 37 (1984) 233-246. L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester, Protective effect of selenium on lung cancer in smelter workers. Br. J. Ind. Med., 42 (1985) 617-626. L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester, Distribution of cadmium, lead and zinc in lung, liver and kidney in long-term exposed smelter workers. Sci. Total. Environ., 50 (1986) 65-85. G. Pershagen, Lung cancer mortality among men living near an arsenic-emitting smelter. Am. J. Epidemiol., 122 (1985) 684-694. I. Holmqvist, An environmental-hygienic investigation of heavy metals in Vagterbotten. Nord. Hyg. Tidskr., Band LV, (1974) 12-22. T. Kjellstr6m, L. Linnman, H. Oigaard, I. Holmqvist, A. Holmstr6m, B. Lindquist, E.
172
27
28
29
30
31 32 33 34 35
36 37
38 39 40 41 42 43
44
L. G E R H A R D S S O N ET A L
Nilsson and O. Vesterberg, Epidemiological dose-effect study of cadmium. An investigation of the ambient and working environment at the R6nnsk/ir smelter. Report of investigation. Department of Environmental Hygiene, The National Environmental Protection Board, Stockholm, Sweden (in Swedish), 1973. C.G. Elinder, K. Millqvist, H. Andersson, B. Lind, B. Nilsson and B. Pettersson, Cadmium and lead in human blood and urine after high consumption of locally cultivated food products from the surroundings of the R6nnsk/ir smelter. Report of investigation. Department of Environmental Hygiene, The Karolinska Institute, Stockholm, Sweden, 1982a (in Swedish). S. Rehnlund, P.-G. Bergfors, K.-H. Gustavson and H. Kollberg, Lead concentrations in the blood of children in the vicinity of a sulfide-ore smelting plant. An evaluation of the influence of various sources of lead pollution in the children's environment. Ambio, 8 (1979) 118-120. G. Pershagen, Lung cancer mortality, occupational exposure and smoking habits in a region surrounding a smeltery, in Proceedings International Symposium on the Control of Air Pollution in the Working Environment. International Labour Office, Geneva, 1978, 370-390. C,G. Elinder, K. Lind6n, E. Hedman and O. Olofsson, Accumulation of cadmium in horses in the R6nnsk/ir area. Report of investigation. Department of Environmental Hygiene, The Karolinska Institute, Stockholm, Sweden, 1982b (in Swedish). H. Ogawa, Systematic science in health care. J. UOEH., 7 (1985) 51-60. T.L. Guidotti, Adaptation of the lifetime health monitoring concept to defined employee groups not at exceptional risk. J. Occup. Med., 25 (1983) 731-736. D.A. Fraser and E.T. Chanlett, The assessment of the total toxicant exposure of man. Am. Ind. Hyg. Assoc. J., 24 (1963) 417-422. M. Re, Microcomputer programs for the evaluation of predictable long-term exposure. Am. Ind. Hyg. Assoc. J., 46 (1985) 369-372. S.A. Conibear and B.W. Carnow, Assessment of exposure and risk through use of a personal, cumulative organ risk index. Presented at the XIX International Congress of Occupational Health, Dubrovnik, Yugoslavia, 1978. J. Gamble and R. Spirtas, Job classification and utilization of complete work histories in occupational epidemiology. J. Occup. Med., 18 (1976) 399-404. R.I. Mitchell, G.R. Smithson Jr, B.P. Price and E.F. Harris, A pilot study to obtain 24hour air pollution exposure profiles. Swedish Council for Building Research. Indoor Air, 4 (1984) 75-80. G. Atherley, A critical review of time-weighted average as an index of exposure and dose and of its key elements. Am. Ind. Hyg. Assoc. J., 46 (1985) 481-487. N.A. Leidel, K.A. Busch and J.R. Lynch, Occupational exposure sampling strategy manual. N1OSH, Washington DC., 1977. L.R. Birkner and L.S. Salzman, Assessing exposure control strategy cost-effectiveness. Am. Ind. Hyg. Assoc. J., 47 (1986) 50-54. L.T. Friberg, The rationale of biological monitoring of chemicals - - with special reference to metals. Am. Ind. Hyg. Assoc. J., 46 (1985) 633-642. R.M. Stern, Production and characterization of a Reference Standard Welding Fume. IIW Doc VIII-861-80, Int. Inst. Welding, Paris, 1980. M. Bohgard, Particle induced X-ray emission analysis and complementary techniques for examination of aerosols in the environment of industrial workers. Doctoral dissertation, Lund University, Sweden, 1983. S.M. McCarthy, S.D. Colome and J.D. Spengler, Indoor and outdoor aerosols: a
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
45
46
47
48 49 50 51
173
multivariate approach to source identification. Swedish Council for Building Research, Stockholm, Sweden. Indoor Air, 2 (1984) 195-200. R.J. Soto, D.A. Kalan, B. Crisostomo, L.L. Falbo, W.J. Groh, L.W. Smith and R.D. Stewart, Data base management system for tracking occupational health. Am. Ind. Hyg. Assoc. J., 44 (1983) 389-394. G. Nordberg, Metal Containing Industrial dust - - Chemico-physical Characterization and Testing of Accumulation Properties. Report to the Swedish Work Environment Fund. Pnr 80-1070, 1987 (in Swedish). L.J. Ungers, P.D. Moskowitz, T.W. Owens, A.D. Harmon and T.M. Briggs, Methodology for an occupational risk assessment: an evaluation of four processes for the fabrication of photovoltaic cells. Am. Ind. Hyg. Assoc. J., 43 (1982) 73-79. O.G. Raabe, Size-selective sampling criteria for thoracic and respirable mass fractions. Ann. Am. Conf. Ind. Hyg., 11 (1984) 53-65. S.C. Soderholm, Size-selective sampling criteria for inspirable mass fraction. Ann. Am. Conf. Ind. Hyg., 11 (1984)47-52. L.A. Wallace and W.R. Ott, Personal monitors: a state-of-the-art survey. J. Air Poll. Contr. Assoc., 32 (1982) 601-610. L. Hong-Kou and K.G. Malmqvist, Hair analysis using proton induced X-ray emission techniques. Sci. Total. Environ., 42 (1985) 213-217.
157
Biological specimen bank for smelter workers L. G e r h a r d s s o n a'b, D. B r u n e c, N . - G . L u n d s t r 6 m a, G. N o r d b e r g a a n d P.O. W e s t e r d
~Department of Environmental Medicine, University of Ume~, S-901 87 Ume~, Sweden bDepartment of Occupational and Environmental Medicine, Lund University Hospital, S-221 85 Lund, Sweden 'D.B.C., Dag Brune Consulting, Bakkehaugveien 16, N-0873 Oslo 8, Norway aDepartment of Medicine, University Hospital of Ume~, S-901 87 Ume~, Sweden
ABSTRACT The biological specimen bank was initiated by the department of Environmental Medicine and the department of Medicine at the University of Ume~ in 1975. The aims of the bank are to collect information on trace elements in human organs. Special attention is focused on the influence of occupational exposure. Tissue samples are taken from deceased workers from a copper and lead smelter in northern Sweden. Control specimens have been collected from deceased normal individuals from four control areas. Lung, liver and kidney samples are collected with quartz instruments and stored in quartz ampoules to avoid contamination. Other samples, e.g. bone, brain, fat, hair, heart muscle, nails, skin and stomach are taken with common autopsy instruments and stored in acid-washed polyethene vessels. The samples are stored at -20°C. The elements Sb, As, Cd, Cr, Co, La and Se are analyzed by neutron activation analysis, Pb and Zn by atomic absorption spectrophotometry. The findings over time can be related to a number of factors: normal values in tissues, airborne exposure and causes of death. Special attention is paid variation over time, reevaluation of threshold limit values and risk assessment.
Key words." specimen bank; human samples; smelter workers; metals
INTRODUCTION T h e e s t a b l i s h m e n t o f the biological s p e c i m e n b a n k was m a d e possible t h r o u g h c o o p e r a t i o n b e t w e e n the U n i v e r s i t y o f Ume~t a n d the Swedish mining c o m p a n y , Boliden M i n e r a l . T h e w o r k e r s f r o m w h o m tissue s a m p l e s are collected are e m p l o y e d at a n o n - f e r r o u s c o p p e r smelter in n o r t h e r n Sweden, R 6 n n s k / i r s v e r k e n , the biggest in S c a n d i n a v i a . F r o m the very start o f the smelter's o p e r a t i o n s in 1930, o c c u p a t i o n a l health surveillance was i n t r o d u c ed a n d g r a d u a l l y increased. T h e first o c c u p a t i o n a l physician a n d safety engineer were b o t h e m p l o y e d in 1932.
158
L. G E R H A R D S S O N ET AL.
The exposure of non-ferrous smelter workers is of complicated nature. They are exposed to a number of heavy metals and other substances and agents. The R6nnsk~ir smelter is located near the city of Skellefte~ on the tip of a peninsula protruding into the sea of Bothnia. From the beginning, only ore from the Boliden mine was processed. This sulfide ore had a very complex composition and was characterized by an extremely high concentration of arsenic. New process techniques had to be developed. The plant now integrates smelter and refining processes for ores and other raw materials such as concentrates and scrap. The processes for the main products are shown in Fig. 1. A more detailed technical description of the processes has previously been reported [1,2]. From the late 1920s, thousands of workers have been exposed to different dust concentrations over many years up to their pension age. The exposure may have both acute and long-term health effects. The disease panorama of the initial years was dominated by dermatitis and symptoms from the upper and lower respiratory tract [3,4] which were the main causes of registered sick-leaves [5]. In an extensive medical survey of 1459 workers, Lundgren et al. [6] defined a disease at the smelter 'Morbus R6nnsk~ir'. This disease was characterized by a chronic rhino-pharyngeo-
copper smelting materials
I•
copper sme/tlng malerials • I converting anode c a s t t n g ~
.... l eSOJ
[esPJ
~ bllater copper electrolysis F
"t
) copper cathodes
½pr.iooa,.goid,
""copper
,
" I metets ptent
smelting I • materials [ I copper I Kaldo furnace J
[
~1 selenium plant
Sl,..
I'Paatlnumgr°up of metals ~-~ selenium °'°""o'o""
i4 ..................... "i crude arsenic
- .nicpla.l ] -----J a"e" me'el t*ars°ni°°ata' "t
,
[
"l
plent
'.
,rssn~ tdox~,
' arsenic acid [central t suiphur dioxide gas cleaner ............................. •-~.~.!SU..~.,. ......................
t
........... slag
hi slag fuming' "1 I I lead oxide dust
t lead concentrates
:41-......... ~P1
~ smelting
sulphur [ ~ liquid dioxide plant ~ " sulphur dioxide ~Iclinkerf . . . . . . ~ zinc clinker ~
I
slag
~ slag products
~ lead ~ crude Kaldo lead - [ Kaldo plant ' leSPI #, crude lead
I
..... rtngHreioog
I
.eo
.d
Fig. 1. R6nnskfir processes and products. ESP, electrostatic precipitators. Courtesy of Boliden Mineral AB.
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
] 59
tracheo-bronchitis with atrophy of the mucous membranes in the respiratory tract, emphysema and impaired pulmonary function [6-8]. Arsenic trioxide and sulphur dioxide were alleged to be the most important etiologic factors. The first lung cancer cases related to work were reported by the Company's occupational health physician, Holmqvist, in the 1960s. In a casereferent study from the smelter, Axelson et al. [9] found that the lung cancer mortality was increased about 5-fold and cardiovascular disease about 2-fold among smelter workers. A dose-response relationship to arsenic was observed. Excess figures for tumours, mainly lung cancer and circulatory diseases have later been reported by Wall [10]. In a case-referent study, Pershagen et al. [11] found a multiplicative effect between smoking and arsenic exposure for the development of lung cancer among the smelter workers. Dust measurements were introduced in the early 1940s and continuously extended. Even if the environmental monitoring was of the compliance type (checks against permissible levels) and not of the dose-response type, the total findings over decades display rather broad medical information of the relationship between different types of jobs and manifest diseases. The workers' occupational titles are well defined and the jobs can be classified by work area and followed from assignment. The turnover has been low, many workers have been employed from 20 to 30 years of age and up to retirement. A continuous employment stretch of 30-40 years is not uncommon. After retirement many workers remained living in the smelter surroundings in homes built during their employment period. The collection of samples from deceased smelter workers to the biological specimen bank was started in 1975 by the department of Environmental Medicine and the department of Medicine at the University of Ume~. The aims of the bank are to collect information on trace elements in human organs. Special attention has been focused on the influence of occupational exposure. MATERIALS AND METHODS
Tissue sampling Through close cooperation with the R6nnskfir Company Health Service and the local hospital in Skellefte], tissue samples have been collected in connection with routine necropsies. The number of samples from different organs of smelter workers in the biological specimen bank is presented in Table 1. Control specimens have been collected from deceased normal individuals from four control areas: the city of Stockholm about 800 km from the smelter, the city of Skellefte~, the area surrounding the smelter (Bure~t and
160
L. G E R H A R D S S O N ET AL.
Skelleftehamn) and from a rural area some 50 km from the smelter (Burtr/isk and J6rn). Lung, liver and kidney samples of about 2 g wet weight are collected with quartz instruments to avoid contamination. They are stored in quartz ampoules and freeze dried. Other samples of 20-50 g wet weight (Table 1) are taken with common autopsy instruments and stored in acidwashed polyethene vessels. All samples are stored at the department of Environmental Medicine at -20°C. The autopsies of the urban controls from the city of Stockholm were all performed at the State Institute of Forensic Medicine, Stockholm. All other autopsies have without selection from diagnoses been undertaken at the County Hospital, Skellefte~ by a single technician. Lung samples are taken from the lower part of the right upper lung lobe and from trachea. Liver specimens are taken about 1 cm below the
TABLE 1
Number of samples from different organs of smelter workers and referents in the biological specimen bank Number of samples
Smelter workers
Referents
Brain Bronchus (trachea) Fat Hair Heart muscle Kidney Liver Lung Nails Psoas muscle Renal pelvis Skin Stomach Testis Bones Femur Finger-bone Iliac crest
I 15 115 115 115 115 115 115 115 115 115 115 115 54 12
42 42 42 42 42 42 42 42 42 42 42 42 23 4
94 56 56 56 115 56 56
35 23 23 23 42 23 23
Rib •
Sternum Temporal bone Vertebrae with spinous process
BIOLOGICAL SPECIMEN B A N K FOR SMELTER WORKERS
161
diaphragmatic surface of the right liver lobe and kidney samples from the cortex of the upper part of the right kidney. Fat and skin specimens are collected from the abdominal area, hair from the back of the head and nails from the left hallux. Samples are also collected from the right renal pelvis, left heart ventricle, right psoas muscle, right testis, the lower part of the stomach and the right occipital brain lobe. Bone samples are taken from seven locations: upper part of sternum, right femur about 1 dm above the knee, right iliac crest, right temporal bone, first lumbar vertebra with spinous process, third rib on the right side about 1 dm from sternum and from the middle phalanx of the second finger on the left hand.
Job classification and smoking habits Working histories from the smelter were provided by the company, Boliden Mineral. Data about smoking habits were obtained by questionnaires answered by relatives of the deceased, supplemented with telephone interviews.
Analytical methods Antimony, arsenic, cadmium, chromium, cobalt, lanthanum and selenium were analyzed by neutron activation analysis (NAA). Zinc and lead were determined by atomic absorption spectrophotometry (AAS). From 1975 to 1978 the neutron activation analysis was performed at the Swedish atomic energy station, Studsvik. In 1978, the Technical X-ray Centre (TRC) in Stockholm took over the analyses after the laboratory cessation at Studsvik. In 1982, these laboratory facilities with personnel were transferred to the Swedish Environmental Research Institute (IVL), Stockholm. The NAA technique used by Studsvik was developed by Wester et al. [12] and further modified by Samsahl et al. [13]. TRC and IVL utilized a slightly different technique developed by Sj6strand [14]. The specimens analyzed at Studsvik were irradiated with a thermal neutron flux of 2 × 1013 n cm -2 S -1 for 3 days (R2 reactor, Studsvik). The analyses performed by TRC and IVL were carried out with a thermal flux of 5 × 1012 n cm -2 s -~ for 2 days (Kjeller reactor, Norway). The radionuclides were chemically separated with an automated ion exchange technique and gamma-spectrometric measurements were carried out on the separated fractions [2,13]. Lead (flameless AAS) and zinc (flame AAS) were from 1975 to 1979 analyzed at the department of Environmental Hygiene at the Karolinska Institute, Stockholm. From 1980, lead (flameless AAS) has been analyzed at the department of Analytical Chemistry, University of Ume~ [15]. Zinc (flame AAS) has since 1980 been analyzed at the department of Environmen-
162
L. G E R H A R D S S O N ET AL.
tal Medicine, University of Ume~. The analytical precision, accuracy and quality control of the NAA and AAS methods have previously been described [2]. Special attention was paid to the contamination of the specimens during sampling, storage and the subsequent neutron irradiation procedure. The specimens were collected in pure quartz containers, which had previously been tested for contamination. The release of various elements from the walls of the pure ampoules was, however, found to be negligible. The ampoules were also immersed in liquid nitrogen prior to opening in order to reduce the pressure arising in the sealed ampoules during irradiation at high neutron fluxes due to 'radiolysis' of the biological material. In certain cases samples were also irradiated during frozen conditions in a cryostate inserted in the reactor in order to reduce contamination, volatilization effects, etc. The radioanalytical assay has been described in further detail by Brune et al. 1984 [161.
Statistics Trace metals in human tissues usually follow a skew distribution [17,18]. Based on computer-aided skewness tests, nonparametric statistical methods were applied (Kruskal-Wallis one-way analysis of variance, Mann-Whitney's U-test and Spearman rank-order correlation coefficients). A difference between the studied groups (Kruskal-Wallis, P < 0.05) was evaluated further by Mann-Whitney's U-test (pairwise comparisons). P-values of <0.05 (Mann-Whitney; two-tailed tests) were considered statistically significant. RESULTS Metal concentrations of antimony, arsenic, cadmium, chromium, cobolt, lanthanum, lead, selenium and zinc have been analyzed in part of the material, mainly in lung, liver and kidney tissues [19-23]. The main findings are presented here.
Age, exposure-time and length of retirement The mean age, duration of exposure and length of retirement of smelter workers and controls are presented in Table 2. No significant differences were found in age between the smelter workers and the control groups. About 50% of the smelter workers died from cardiovascular diseases and about 1/3 from malignancies (8-9% from lung cancer and 13-14% from gastrointestinal cancer). Deaths from cardiovascular diseases dominated in the two control groups, 75-80% in the rural group (living some 50 km from the smelter) and 100% in the urban group from the city of Stockholm.
163
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
TABLE 2 Mean-age, exposure-timeand length of retirement in smelter workers and rural (Burtr~iskand J6rn) and urban controls (the city of Stockholm); N, number of workers, mean values 4- S.D.) Group
N
Mean-age (years)
Exposure time (years)
Length of retirement (years)
All smelter workers Rural controls Urban controls
85 15 10
68.3 + 9.1 67.7 4- 10.1 68.8 4- 6.1
31.2 4- 8.2 --
7.3 4- 5.9 --
Tissue concentrations
Median values of Sb, As, Cd, Cr, Co, La, Pb, Se and Zn in lung liver and kidney tissues of smelter workers and controls are summarized in Tables 3 and 4. With the exception of zinc, significantly higher concentrations of these elements were found in the workers' lungs (N = 85) than in non-exposed rural controls from Burtr/isk and J6rn (N = 15, P _< 0.027). Significantly higher lung tissue concentrations of antimony, arsenic and lead were noted among all workers compared to urban controls from the city of Stockholm (N = 10, P < 0.001). The highest lung accumulation was noted for antimony and arsenic; an l l-fold and 6-fold increase, respectively [2,19,20]. The lung tissue concentrations of some metals, e.g. antimony and cobalt, did not decline with time after exposure had ceased among the smelter workers, which indicates a long biological half-time. Metal concentrations related to diagnosis
In the further analysis of the material the workers were subdivided into groups according to diagnoses in autopsy protocols. Workers who died from lung cancer (N = 7) had the lowest lung selenium content relatively to other metals, both compared with other disease categories among the workers (GIcancer, other cancers, cardiovascular diseases, cerebrovascular diseases, other causes) and with the two control groups [22]. The highest lung tissue levels of cadmium were found in the lung cancer group in which, however, smokers and ex-smokers were overrepresented [23]. Several of the lung cancer cases also had from high up to the highest lung concentrations of antimony, arsenic, cadmium, lanthanum and lead. The metal concentrations in liver (metabolism) and kidney (excretion) illustrate the systemic distribution. The highest cadmium concentrations in liver were found in the lung cancer group [23].
164
L. GERHARDSSON ET AL.
TABLE 3 Median values of antimony, chromium, cobalt and lanthanum in lung, liver and kidney tissues in smelter workers and controls. N, number of workers. All values in ppb (/~g/kg wet weight). Qr, quotient of median values between smelter workers and rural controls. QU, quotient of median values between smelter workers and urban controls. From Gerhardsson et al. 1982, 1984, 1988. Element
Antimony Liver N Lung N Kidney N Chromium Liver N Lung N Kidney N Cobalt Liver N Lung N Kidney N Lanthanum Liver N Lung N Kidney N
Smelter workers
Rural controls
12 21 260 85 6 21
7 8 32 15 5 8
11 20 450 85 9 21
4 8 l0 15 3 8
11 20 i7 85 3 21
16 8 7 15
5.0 20 9.6 85 0.2 20
5.5 8 5.0 15 0.1 8
Urban controls
Qr
QU
1.7 19 l0
8.1
13.7
1.2
2.8 199 10
4.1
2.3
3.0
0.7 ll 10
2.4
1.5
3.0
1
8 0.9 10.0 10
1.9
1.0
2.0
DISCUSSION
A multifactorial genesis for the development of lung cancer is supported from these studies [20-23]. In the lung cancer group smoking has probably played a dominant role. High levels of antimony, arsenic, cadmium and several other metals may contribute with different potency. Low levels of
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
165
TABLE 4 Median values of arsenic, cadmium, lead, selenium and zinc in lung, liver and kidney tissues in smelter workers and controls. N, number of workers. Zinc values in ppm (mg/kg wet weight), all other values in ppb (t~g/kg wet weight). Qr, quotient of median values between smelter workers and rural controls. QU, quotient of median values between smelter workers and urban controls. From Wester et al. 1981, Gerhardsson et al. 1986, 1988. Element
Arsenic Liver N Lung N Kidney N Cadmium Liver N Lung N Kidney N Lead Liver N Lung N Kidney N Selenium Liver N Lung N Kidney N Zinc Liver N Lung N Kidney N
Smelter workers
Rural controls
7 21 35 85 9 21
4 8 7 15 5 8
965 21 166 85 17 300 21
350 8 39 15 3025 8
470 21 140 85 348 21
285 8 55 15 235 8
195 20 152 85 495 20
195 8 110 15 755 8
55.7 21 11.5 85 31.0 21
49.8 8 10.2 15 26.3 8
Urban controls
Qr
QU
1.8 5 I0
5.0
7.0
1.8
2.8 79 10
4.3
2.1
5.7
1.6 39 10
2.5
3.6
1.5
1.0 136 10
1.4
1.1
0.7
1.1 12.2 10
1.1 0.9 1.2
166
L. GERHARDSSON ET A L
selenium indicate reduced protection. The smelter workers are also exposed to irritating gases such as SO2. Some air pollutants are of organic origin, e.g. polyaromates such as benzo[a]pyrene. O f the interactions that may occur, some may increase the carcinogenic risks. The biological specimen bank also gives information about tissue concentrations and adverse health effects in nonoccupationally exposed individuals living at different distances from the smelter.
Ambient surveillance Emission considerations placed the plant fairly remote from agriculture and forestry locations, with favourable water and wind directions. The prevailing wind directions are north-south. From the end of the 1960s the ambient environment protection program w a s greatly expanded, as was the production volume. In the period 1970-1984 such environmental investments amounted to SEK 270 million compared to SEK 120 million for improvements of the working environment. The ambient emissions to air during this period were reduced by an average of 60% and those to water by about 90% (Tables 5 and 6). In comparison, air emissions of arsenic were of the magnitude 1-2 tons per day 1930-1949, 0.2-0.8 tons per day 1950-1972 and around 0.1 ton per day in 1984 [24]. In the late 1960s the ambient air levels of arsenic and lead were mostly low 0.5/zg/m 3 (monthly average values) up to 7 km from the smelter [24]. Holmqvist [25] reported lead concentrations in air ranging from 0.3-0.5 tzg/m 3 from dust measurements in domestic areas up to 5 km from the smelter. For other areas with this distance lead varied between 0.2-3.8 /~g/m 3 with a mean value of 0.85 t~g/m 3.
TABLE 5 Emissions to atmosphere from the R6nnsk~ir smelter (tons/year). 1995~ are estimated figures Time period
Dust
Cu
Pb
Zn
Cd
As
Hg
1965-69 1970-73 1978 1984 1986 1995~
3200 2400 1945 857 340 240
190 203 164 42 21 14
684 555 235 103 64 41
488 393 250 122 55 38
13.4 13.5 5.0 2.5 1.4 1.0
154 110 75 39 13 10
3.5 2.4 1.1 0.8 0.6 0.4
20 000 22 000 10100 6750 5400 3900
167
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
TABLE 6 Emissions to water from the R6nnsk/ir smelter (tons/year). 1995 e, figures are estimates. Time period
Cu
Pb
Zn
Cd
As
Hg
1965-69 1970-73 1978 1984 1986 1995 e
165 89 12.5 5.6 3.3 3.3
214 164 11 2.4 2.4 2.4
93 99 37 8 8 8
7.2 7.8 2.2 0.4 0.4 0.4
2078 1846 589 24 24 24
5.2 1.7 0.46 0.079 < 0.1 <0.1
Several studies have reported elevated levels of cadmium and lead in crops and vegetables in the surroundings of the smelter [26]. The main food products consumed, however, are brought in from other parts of the country.
Earlier studies of ambient health effects Kjellstr6m et al. [26] analyzed cadmium in urine from women living at different distances from the smelter (Skelleftehamn, Bure~ and J6rn). Minor differences were noted between the areas. The concentrations were of the same magnitude as from southern Sweden (0.65 tzg Cd/g creatinine). The protein excretion in urine showed no signs of kidney damage. Holmqvist [25] compared the concentrations of arsenic, cadmium, lead and mercury in urine from three groups of women living at different distances and directions from the plant. Only normal values were observed. Lead determinations in blood [25] gave slightly increased mean values for men living in the immediate vicinity of the plant (mean 13.6 #g/100 ml). The work-titles, however, had a greater impact on the blood lead levels than the distance to the smelter. Elinder et al. [27] analyzed the concentrations of cadmium and lead in blood and urine from men and women living in Bure~t and Skelleftehamn. No association was found between a high consumption of locally cultivated food products and elevated blood lead and cadmium levels. In 1979, Rehnlund et al. [28] compared the blood lead levels of children living near the smelter with children living in a distant urban area (Ume~, about 140 km from the smelter). Low concentrations were found in both groups (7.3 ~g/100 ml and 7.7 t~g/100 ml, respectively). In a case-control study, Pershagen [29] found an increased mortality for
168
L. G E R H A R D S S O N ET AL.
respiratory cancer among former employees at the smelter (risk ratio 2.6). The risk was particularly raised for workers earlier exposed to high levels of inorganic arsenic (risk ratio 7.3) and could not be explained by differences in smoking habits. Recently, Pershagen [24] in a case-referent study reported a relative risk of 2.0 (P < 0.05) for lung cancer among men (never employed at R6nnsk~ir) who lived within approximately 20 km from the smelter. The increased mortality could not be explained by smoking habits or occupational background. Emissions, particularly of arsenic from the smelter, may have played a role. No increased levels of cadmium in liver and kidney of cows were found in J6rn and Bure3., compared to controls from southern Sweden. Horses showed a negative correlation between blood and kidney levels of cadmium and increasing distances to the smelter [30].
Future perspectives In an ageing population like the Swedish one, the development of work related diseases may take place long time after retirement. International bodies (ILO, WHO) have discussed both needs and possible ways to prolong occupational health supervision [31]. Current occupational health programs do not include the evaluation of workers' health changes after they have left the plant [32]. Our studies exemplify the difficulties but also the inherent value of such a prolongation. Retrospective studies of this kind are highly promoted by a prospective framework of systematization of the companies' health programs [33]. Earlier efforts of occupational health were directed towards the control of specific and quite obvious diseases from single substances such as silica, lead and mercury. In prospective studies now on the move, the evaluation of the effect of the total multifactorial exposure are brought into sharp focus. Routine periodic medical examinations should as far as possible be made part of life time health promotion programs [34]. The resources needed are, if coordinated, within economically acceptable limits. The occupational health organization can on an individual basis follow the health changes using each employee as his own control [35]. Groups of workers with similar exposure can be continuously followed. Complete work histories can be utilized in occupational epidemiology [36]. The weak link is that the public health care system has no resources or appropriate strategies for a follow-up after employment ceases. The concept, however, behind the revised Swedish public sick and health care legislation is that more pronounced preventive efforts should be enforced. If this preventive-oriented concept is maintained, exposures at work, in home and in traffic, etc. can be evaluated and related to their long-term
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
169
health effects [37]. Great difficulties have of tradition met the evaluation of total diurnal exposure over a life-time. There is a great difference between exposure and individual dose [38]. Earlier calculations have included a wide range of approximations [39]. In classical occupational monitoring programs, air-pollutant measurements were 10-20 times more expensive than biological sampling for the same - - and as a rule - - very restricted information [40]. The occupational hygiene measurements referred to from this smelter were children of its time and dominated by analyses of total aerosol concentrations. Comprehensive occupational hygiene dose evaluations from this period easily reached a cost level of 5-10% of the workers' annual wages. Biological monitoring is, on the other hand, available for only a restricted number of air pollutants and must even in the future be considered as supplemental to environmental monitoring [41]. In the health surveillance and monitoring programs now emerging, an array of instruments and analytical techniques are available for detailed characterization and quantification of pollutants, their diffusion and uptake [42,43]. Use of carefully selected indicator substances characteristic for hygienically important emitting sources, use of cluster analysis and pairwise comparisons, enrichment factors etc. make it possible to differentiate the exposure using basic emission data [44]. Data base management systems for tracking occupational health use personal modules, medical/clinical modules, occupational hygiene/toxicology modules as system functions [45]. Microcomputer programs are available for the evaluation of predictable long-term exposure [34]. Aerodynamic properties of particles and particle agglomerates determine deposition and retention in the respiratory system. Detailed studies of particles, their size, crystal form and chemical composition and properties can be made by using a combination of electron and transmission microscopy, Xray fluorescence and particle induced X-ray emission analysis, (PIXE), electron spectroscopy for chemical analyses (ESCA), etc. Such discrimination techniques which have been used to change particles from the R6nnsk~ir smelter [46] make it possible to compare new combinations of industrial aerosols with older ones for which significant epidemiological data are available [47]. Total suspended particulate in air can now be transformed to size classes and the evaluation of exposure can be made size specific. Particle size selective personal monitors have opened roads to the development of particle size selective threshold limit values for inspirable particulate mass, thoracic particulate mass and respirable particulate mass [48,49]. Such monitors can be carried without disturbing work [50]. Simultaneous determination of multielements in biological tissues, col-
170
L. GERHARDSSONET AL
lected in e.g. a biological specimen bank, can be used to construct an elemental 'fingerprint' for the body or for a critical target organ. The capability exists for studying individual molecules in biological specimens that weigh one quadrillion of a gram. Such element concentration profiles in tissues can be related to corresponding environmental exposure 'fingerprints'. Biological and environmental monitoring overlap, to the benefit of both. Human hair is for example a recording filament which can reflect metabolic changes, changes of concentrations in the body and serum, changes in external exposure and interdependencies over time [51]. Several other types of successful measurements of trace element distributions and variations in tissues and body fluids, covering time-periods of years to decades, have been reported. The findings over time in a biological specimen bank can be related to a number of factors: normal values in tissues, air-borne exposure and causes of death. Special attention is paid to variation over time, reevaluation of threshold limit values and risk assessment. CONCLUSION
New strategies and techniques for multielemental quantification of environmental exposure from work, air, food, water, etc can now be much better related to the uptake of corresponding elements in the body compartments. In other words, available techniques including the use of a biological specimen bank, make prolonged health studies after retirement feasible. The need is expanding. Occupational and environmental medicine here share future responsibilities. ACKNOWLEDGEMENT
Financial support was given by the Swedish Work Environment Fund, Project No. 80/107. REFERENCES 1 L. Gerhardsson, Trace elements in tissues of deceased smelter workers - - relationship to mortality. Medical Dissertation, The University of Ume~, Ume~, Sweden, 1986. 2 L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester, Multielemental assay of tissues of deceased smelter workers and controls. Sci. Total. Environ., 74 (1988) 97-110. 3 G. Inghe and A. Bursell, The R6nnskfir Study 1937. Report of investigation, Stockholm, Sweden (in Swedish, unpublished). 4 T. Sj6strand, Changes in the respiratory organs of workmen at an ore smelting works. Acta Med. Scand., Suppl., 196 (1947) 687-699. 5 .~. Nygren, Historic working environment description - - The R6nnskfir smelter. Report
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
6 7 8 9
10 11
12 13
14 15
16 17 18 19 20
21
22 23
24 25 26
171
of investigation, Boliden Mineral AB, S-932 00 Skelleftehamn, Sweden, 1980 (in Swedish). K.D. Lundgren, N.G. Richtn6r and T. Sj6strand, Morbus R6nnsk/ir. Nord. Med., 46 (1951) 1556-1560 (in Swedish). K.D. Lundgren, Lesions in the respiratory organs of workers at a smelter. Nord. Hyg. Tidskr., 3 (1954) 66-82 (in Swedish with English summary). L.E. Warfvinge, Contribution to the diagnosis of the R6nnsk/ir bronchitis. L/ikartidningen, 65 (1968) 1698-1701 (in Swedish). O. Axelsson, E. Dahlgren, C.-D. Jansson and S.O. Rehnlund, Arsenic exposure and mortality: a case-referent study from a Swedish copper smelter. Br. J. Ind. Med., 35 (1978) 8-15. S. Wall, Survival and mortality pattern among Swedish smelter workers. Int. J. Epidemiol., 9 (1980) 73-87. G. Pershagen, S. Wall, A. Taube and L. Linnman, On the interaction between occupational arsenic exposure and smoking and its relationship to lung cancer. Scand. J. Work. Environ. Health, 7 (1981) 302-309. P.O. Wester, D. Brune and K. Samsahl, Radiochemical recovery studies of a separation scheme for 23 elements in biological material. Int. J. Appl. Radiat. lsot., 15 (1964) 59-67. K. Samsahl, P.O. Wester and O. Landstr6m, An automatic group separation system for the simultaneous determination of a great number of elements in biological material. Anal. Chem., 40 (1968) 181-187. B. Sjfstrand, Simultaneous determination of mercury and arsenic in biological and organic materials by activation analysis. Anal. Chem., 36 (1964) 814-819. E. Lundberg, W. Frech and I. Lindberg, Determination of lead in biological materials by constant-temperature electrothermal atomic absorption spectrometry. Anal. Chimica. Acta, 160 (1984) 205-215. D. Brune, B. Forkman and B. Persson, Nuclear Analytical Chemistry, Studentlitteratur, Chartwell-Bratt Ltd, Lund, (1984)446-453. I. Harding-Barlow, Studies on the trace element content of human tissues. Doctoral dissertation, University of Capetown, 1961. P.O. Wester, Trace elements in heart tissue. Acta Med. Scand., Suppl., 439 (1965) 1-48. P.O. Wester, D. Brune and G.F. Nordberg, Arsenic and selenium in lung, liver and kidney tissue from dead smelter workers. Br. J. Ind. Med., 38 (1981) 179-184. L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester, Antimony in lung, liver and kidney tissue from deceased smelter workers. Scand. J. Work Environ. Health, 8 (1982) 201-208. L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester~ Chromium, cobalt and lanthanum in lung, liver and kidney tissue from deceased smelter workers. Sci. Total. Environ., 37 (1984) 233-246. L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester, Protective effect of selenium on lung cancer in smelter workers. Br. J. Ind. Med., 42 (1985) 617-626. L. Gerhardsson, D. Brune, G.F. Nordberg and P.O. Wester, Distribution of cadmium, lead and zinc in lung, liver and kidney in long-term exposed smelter workers. Sci. Total. Environ., 50 (1986) 65-85. G. Pershagen, Lung cancer mortality among men living near an arsenic-emitting smelter. Am. J. Epidemiol., 122 (1985) 684-694. I. Holmqvist, An environmental-hygienic investigation of heavy metals in Vagterbotten. Nord. Hyg. Tidskr., Band LV, (1974) 12-22. T. Kjellstr6m, L. Linnman, H. Oigaard, I. Holmqvist, A. Holmstr6m, B. Lindquist, E.
172
27
28
29
30
31 32 33 34 35
36 37
38 39 40 41 42 43
44
L. G E R H A R D S S O N ET A L
Nilsson and O. Vesterberg, Epidemiological dose-effect study of cadmium. An investigation of the ambient and working environment at the R6nnsk/ir smelter. Report of investigation. Department of Environmental Hygiene, The National Environmental Protection Board, Stockholm, Sweden (in Swedish), 1973. C.G. Elinder, K. Millqvist, H. Andersson, B. Lind, B. Nilsson and B. Pettersson, Cadmium and lead in human blood and urine after high consumption of locally cultivated food products from the surroundings of the R6nnsk/ir smelter. Report of investigation. Department of Environmental Hygiene, The Karolinska Institute, Stockholm, Sweden, 1982a (in Swedish). S. Rehnlund, P.-G. Bergfors, K.-H. Gustavson and H. Kollberg, Lead concentrations in the blood of children in the vicinity of a sulfide-ore smelting plant. An evaluation of the influence of various sources of lead pollution in the children's environment. Ambio, 8 (1979) 118-120. G. Pershagen, Lung cancer mortality, occupational exposure and smoking habits in a region surrounding a smeltery, in Proceedings International Symposium on the Control of Air Pollution in the Working Environment. International Labour Office, Geneva, 1978, 370-390. C,G. Elinder, K. Lind6n, E. Hedman and O. Olofsson, Accumulation of cadmium in horses in the R6nnsk/ir area. Report of investigation. Department of Environmental Hygiene, The Karolinska Institute, Stockholm, Sweden, 1982b (in Swedish). H. Ogawa, Systematic science in health care. J. UOEH., 7 (1985) 51-60. T.L. Guidotti, Adaptation of the lifetime health monitoring concept to defined employee groups not at exceptional risk. J. Occup. Med., 25 (1983) 731-736. D.A. Fraser and E.T. Chanlett, The assessment of the total toxicant exposure of man. Am. Ind. Hyg. Assoc. J., 24 (1963) 417-422. M. Re, Microcomputer programs for the evaluation of predictable long-term exposure. Am. Ind. Hyg. Assoc. J., 46 (1985) 369-372. S.A. Conibear and B.W. Carnow, Assessment of exposure and risk through use of a personal, cumulative organ risk index. Presented at the XIX International Congress of Occupational Health, Dubrovnik, Yugoslavia, 1978. J. Gamble and R. Spirtas, Job classification and utilization of complete work histories in occupational epidemiology. J. Occup. Med., 18 (1976) 399-404. R.I. Mitchell, G.R. Smithson Jr, B.P. Price and E.F. Harris, A pilot study to obtain 24hour air pollution exposure profiles. Swedish Council for Building Research. Indoor Air, 4 (1984) 75-80. G. Atherley, A critical review of time-weighted average as an index of exposure and dose and of its key elements. Am. Ind. Hyg. Assoc. J., 46 (1985) 481-487. N.A. Leidel, K.A. Busch and J.R. Lynch, Occupational exposure sampling strategy manual. N1OSH, Washington DC., 1977. L.R. Birkner and L.S. Salzman, Assessing exposure control strategy cost-effectiveness. Am. Ind. Hyg. Assoc. J., 47 (1986) 50-54. L.T. Friberg, The rationale of biological monitoring of chemicals - - with special reference to metals. Am. Ind. Hyg. Assoc. J., 46 (1985) 633-642. R.M. Stern, Production and characterization of a Reference Standard Welding Fume. IIW Doc VIII-861-80, Int. Inst. Welding, Paris, 1980. M. Bohgard, Particle induced X-ray emission analysis and complementary techniques for examination of aerosols in the environment of industrial workers. Doctoral dissertation, Lund University, Sweden, 1983. S.M. McCarthy, S.D. Colome and J.D. Spengler, Indoor and outdoor aerosols: a
BIOLOGICAL SPECIMEN BANK FOR SMELTER WORKERS
45
46
47
48 49 50 51
173
multivariate approach to source identification. Swedish Council for Building Research, Stockholm, Sweden. Indoor Air, 2 (1984) 195-200. R.J. Soto, D.A. Kalan, B. Crisostomo, L.L. Falbo, W.J. Groh, L.W. Smith and R.D. Stewart, Data base management system for tracking occupational health. Am. Ind. Hyg. Assoc. J., 44 (1983) 389-394. G. Nordberg, Metal Containing Industrial dust - - Chemico-physical Characterization and Testing of Accumulation Properties. Report to the Swedish Work Environment Fund. Pnr 80-1070, 1987 (in Swedish). L.J. Ungers, P.D. Moskowitz, T.W. Owens, A.D. Harmon and T.M. Briggs, Methodology for an occupational risk assessment: an evaluation of four processes for the fabrication of photovoltaic cells. Am. Ind. Hyg. Assoc. J., 43 (1982) 73-79. O.G. Raabe, Size-selective sampling criteria for thoracic and respirable mass fractions. Ann. Am. Conf. Ind. Hyg., 11 (1984) 53-65. S.C. Soderholm, Size-selective sampling criteria for inspirable mass fraction. Ann. Am. Conf. Ind. Hyg., 11 (1984)47-52. L.A. Wallace and W.R. Ott, Personal monitors: a state-of-the-art survey. J. Air Poll. Contr. Assoc., 32 (1982) 601-610. L. Hong-Kou and K.G. Malmqvist, Hair analysis using proton induced X-ray emission techniques. Sci. Total. Environ., 42 (1985) 213-217.