- Email: [email protected]
Lead in sweat and its relationship to salivary and urinary levels in normal healthy subjects
The Science of the Total Environnlent, 103 ( 1991 ) i 13-122 Elsevier Science Publishers B.V., Amsterdam
!!3
Lead in sweat and its relationship to salivary and urinary levels in normal healthy subjects* F.O. Omokhodion and G.W. Crockford Unit of Occupational Health, London School t!l"Hygibne and Tropical Medic#le, Keppel Street, London WCIE 7HT. United King&mr (Received January 24th, 1990; accepted March 6th, 1990)
ABSTRACT
Sweat was collected fl'om the arms of 24 normal healthy subjects while they sat in a hot chamber. Blood, urine and saliva samples were also collected. These were analyzed for lead by atomic absorption spectrophotometry. Sweat lead levels recorded in this study are lower than those previously reported. Subjects with mean blood lead levels of 8.6211gdl t (range 6-13.6) had mean sweat levels of 5.21~gl ~ (range !.5-13.0), .-. 25% of their urinary levels. Although salivary lead levels with a mean of 4.8 ltg I t (range 2.5-10) are comparable to sweat levels, their relationship to blood lead levels is poor (r = - 0 . 1 8 6 compared with r values of 0.7208 and 0.234 for sweat and urinary levels, respectively). INTRODUCTION
Indications from the result:, of nutritional balance studies suggest that nutrients are lost in sweat in such amounts that allowances have to be made for these losses in the calculation of daily requirements [1,2]. It is therel'ore possible thai sweating may relieve the body of toxic materials [3]. One of the earliest reports was on Spanish miners of cinnabar ore who had chronic signs of mercurialism and were put in a hot environment where forced sweating occurred; this resulted in a reduction of their symptoms and hence constituted a form of treatment [4]. More recently, Szadkowski et,al. [5] found lower blood copper levels in workmen chronically exposed to heat in a steel plant compared with controls who were not thus exposed. They attributed the thermally induced hypocupraemia to the excretion of copper in sweat. although sweat copper levels were not measured. The excretion of lead in the sweat of lead workers was demonstrated by Sheils in 1954 [6] and to ascertain whether or not contamination may be responsible for the levels recorded, he demonstrated measur~:ble amounts of *Based on a paper presented at tile International Symposiun'~ on Work in a Hot Environment and Hcat Related Disorders. Khartoum. 27-31 January 1988. 0048-9697/91/$03.50
~" 1991
)
,
t
- Elsevier Science ! ubhsners B.V.
ii4
lead in the sweat of a non-exposed subject which increased substantially with increasing oral doses of lead acetate. Several workers have since either refuted or confirmed the presence of lead in sweat, but the wide variation in reported values and incomplete data on blood and urinary lead levels in the subjects tested have done little to settle the question of the significance of sweating as an avenue for the loss of lead from the body. In this study, experiments were designed to quantify ehe io~e~ r~f lead in uncontaminated sweat of no~-occupationally exposed persons with known blood lead levels and to compare these losses with urinary losses. METHOD
In the majority of previous investigations, samples of sweat were collected from the skin by absorbing it on filter paper or gauze pads [6], scraping it from the skin or allowing it to drop freely into beakers or test tubes [7], using towels previously washed in deionised water to wipe sweat from the body surface [8], collecting it as it drops from the body into plastic sheets [l] or into plastic bags strapped around the arms [3]. There is a risk of contamination with each of these methods, from the skin and from the environment, including the materials used for sweat collection. The arm bag method was chosen for this study as the arm is easy to isolate for thorough cleaning and can be completely enclosed, thereby eliminating environmental contamination. Sweat was induced naturally by sitting subjects in a hot chamber at 40-45°C and 35% relative humidity for 1 h.
Subyects Twenty-four healthy men participated in the study. They were volunteers from among staff and students of the ~chool. Volunteers who had skin disease or a history of adverse reaction to heat were excluded from this study.
Sample collection Sweat, urine and saliva samples were collected in polythene bottles which had been soaked in 10% nitric acid for 48 h and then rinsed with deionised water.
Blood Blood samples (5 ml) were taken from the antecubital fossa with lithium heparin syringes. Urine Single voidings of urine were collected in polythene bottles immediately before the heat exposure.
!15
Saliva The mouth was rinsed thrice with 100ml of 3% citric acid and thrice with i 00 mi of deionised water. Saliva collected in the mouth was emptied into the polythene bottles provided. The procedure was repeated when necessary until 10 ml of saliva had been collected.
Sweat The subjects changed from day clothes into disposable overalls (Tyvek suits). Both sleeves were rolled up to the deltoid region and using a brush and lead-free soap (Phisohex), subjects throughly scrubbed nails, palms, hands and the arms. Both arms were rinsed with tap water and scrubbed again as before. They were then thoroughly rinsed with deionised distilled water. Excess water was shaken off the arms, subjects being careful not to touch any object until the arms were placed inside polythene bags made from a roll of heavy weight polythene tubing. The first set of arm bags were sealed at the distal ends and -- 50 ml of deionised water was poured into each bag and their upper ends were secured with tourniquets while each arm was rinsed inside the bag for --, 2 rain. These rinses were poured into polythene bottles and stored for analysis Excess water was shaken off the arms and each arm was placed inside an open-ended piece of polythene tubing which was then sealed at the wrists and the upper end of the arms with adhesive tape (Sleek). The subjects sat in the hot chamber for I h while body temperatures and heart rates were continuously monitored and recorded at 5-min intervals. At the end of the exposure period, sweat collected in the bags was poured into separate polythene bottles and stored for analy~s.
Sample analysis Storage apd handling Blood and urine samples were stored in a refrigerator until analysis. Sweat and, saliva sam01es were acidified to 1% with Aristar grade nitric acid to prevent bacterial action and then left to stand at room temperature until analysis, usually by the third day. In the handling of samples, from the collection stage to analysis, great care was taken to prevent contamination. All containers used for collection and storage of samples were tested and found to be free from lead. All glassware used for analysis was washed thoroughly, rinsed with 10% nitric acid and then rinsed with deionised water. Sample cups and caps were soaked in 10% nitric acid and rinsed in deionised water prior to use.
Analytical method All samples were analysed for lead using an atomic adsorption spectro-
116
photometer (Perkin-Elmer Model 460) equipped with a graphite furnace, an HG;~ 500 atomiser unit and deuterium background correction.
Blo,~d. Blood samples were rolled for 1 h to ensure an even mix. Samples were diluted 1 in 10 into 2ml sample cups using 0.1% TritonX and 0.11% NH,H_,HPO4 as diluent. The cups were covered and rolled for a further i h before analysis. ";;veat, saliva and urine samples. ,~n attempt to calibrate these samples against aqueous standards produced matrix effects which varied in magnitude among different individuals. Each sample was therefore analysed by the standard additions method (calibration by the addition of known amounts of lead to samples). Urine samples were analysed for creatinine. The temperature programme for lead determination was as follows: Step
Temp. (°C)
Ramp (s)
Hold Is)
I 2 3 4 5 6
100 130 140 650 1850 2200
10 5 5 20
5 5 10 25
I
5
3
I
Instrument conditions Wavelength Spectral bandwidth Lamp current Gas pressure Integration period Injection volurne Replicates
283.3 nm 0.7 nm 8 MA 45 psi (I psi = 6.9 kPa) 6s 20 ltl Three
Precision and accuracy At the time of this invest~$at~o,1, ,:he analytical qu~:lity control of the laboratory was monitored by participation in a national quality assurance scheme (UKEQAS). The analysis of internal quality assurance samples of sweat at the top range of values, i,e. 12.5~tg l -t, showed between-batch precision of 6.9% and within-batch precision of 3.0%. A within-batch precision of 18% was recorded for internal quality assurance samples containing 1.5 z~g l -i lead. Quality assurance analysis for urine did not yield consistent results after 2 months of storage and this was therefore discontinued. However, the precision of lead determinations in urine and saliva was demonstrated by regression coefficients of the standard additions curve varying from 0.•91 to 1.0. An-' sample with a coefficient below this value was reanalysed.
!!7
The limit of detection, as defined by the concentration of lead corresponding to three standard deviations of the response of 10 samples of the field blank [9] (1% nitric acid), was determined to be 1.1 p g l RESULTS
Although most subjects indicated a feeling of discomfort at the beginning of the exposure period, readings of heart rate and body temperature as well as subjective feelings of undue stress did not necessitate the termination of the experiments on any of the subjects. A mean total body sweat loss of 490 ml was recorded during the I-h exposure period, while a mean of 12.6 and 10.3 ml of sweat were collected from the left and right arms, respectively. The results of lead determination in left and right arm sweat samples were
TABLE
I
Lead concentrations in body lluids of non-occupationally cxposed persons Subject
I 2 3 4 5 t~ 7 8 9 10 !1 !2 13 14 15 16 17 18 19 20 21 22 23 24 Mean SD N
Blood
Swear'
Saliva
Urine
(/lgdl -I)
(/igl i)
(/~gl i)
(/:gl i)
(l~g/g creatinine)
14 26 44
7.7 ! 0. I 8.4
38 16 18 18 44 20 14 !6 16 26 18 30 20 16 10 26 18 20 12 16 21.56 9.43 23
13.9 10.2 3,4,4 11.8 23.6 24.q 8.4 28.3 il.2 19.6 13.3 31.6 '3'3 _..7 28.9 11.2 10.9 7.9 44.2 !q.3 9.7 17.92 10.16 23
9.0 6.0 9.8 7.9 8.0 6.8 9.0 111.5 13.6 6.0 8.3 9.3 8.3 6.0 12.0 8.0 7.0 10.0 8.0 7.0 9.8 10.4 10.2 6.0 8.62 1.96 24
5.5 3.0 6.0 6.0 3.25 4.75 6.0 13.0 4.0 7.25 5.0 2.25 1.5 4.75 5.75 2.5 9.5 4.25 9.25 3.5 3.0 5.23 2.74 2i
~Mean concentration of left and right arm sweat.
6.0 3.0 4.1) 2.5 4.0 5.9 5.5 4.5 2.5 6.0 4.0 3.5 2.5 2,5 6,5 10.0 6.0 8.5 7.5 4.5 2.5 4.80 2. ! I 21
118
Sweat Lead (pg/I)
14r
i I
i
/ .
L
/ •
I
t
/
/
t
//
.../..
•
•
•
•
4 " •
•
w
Z
i ! !
O k
0
......
I
20
40
60 80 Blood Lead (pg/I)
1. . . .
100
I
......
120
J
140
Fig. !. Lead in non-occupationally exposed persons: blood vs sweat.
similar. Sweat samples that were centrifuged before analysis yielded the same results as uncentrifuged samples. Samples analysed on the day of collection showed no difference in lead determinations between acidified and non-acidified sweat. Table ! shows the blood, sweat, saliva and urinary lead levels of 24 subjects. Sweat lead determinations were not recorded for subjects with arm rinses which showed traces of lead. Salivary samples with streaks of blood were discarded. Blood lead levels ranged from 6 to 13.6 #gdl- t, while corresponding sweat levels ranged from 1.5 to 13.0 #g l-t. Mean sweat lead levels were 6% of blood levels, while salivary levels were 5.5%. However, there is a stronger correlation between blood and sweat lead (Fig. 1) than between blood and saliva lead (Fig. 2); r = 0.7208 and r = -0.1864, respectively. The correlation between blood and urinary levels is poor (r = 0.234). The concentration of lead in sweat is ~ 25% of that in urine. DISCUSSION
Mean sweat lead levels in non-occupationally exposed subjects have been
!i9 Salivary Lead (pg/I) 12
10
~
,
~
ia
om
a
....1
I
• o •
0 0
1
1
20
40
n
I
. . . . .
•
I .
.
.
.
60 80 Blood Lead (ug/I)
.
I
I00
--
•
I .
120
.
.
.
140
Fig. 2. Lead in non-occupationally exposed persons: blood vs saliva.
reported a~ varying from 0 to 150ttgl -~. Mean arm bag sweat values of 83 #g 1-I [1] and 51 gg 1-I [3] have indicated that sweat and urine could contain equal amounts of lead, whereas values such as 1501~gl-~ [10] and estimates made by Shiels [6] from axillary pad sweat indicate that lead losses in sweat may be greater than those in the urine, lontophoresis generated sweat has also been reported to yield mean values of 24, 36 and 41/tg 1- ~[ 11], values around the range of urinary lead levels. Grant et al. [12] also reported a range of 10.9-42.0/~gl -~ (in 26 control children). Vellar [13], however, reported that lead, together with other heavy metals such as mercury, nickel and cobalt, were not present in measureable quantities in sweat. His finding was a good indication that these elements, and lead in this case, may be present in sweat in amounts much less than had previously been reported. The difficulty in collecting uncontaminated sweat is wel! recognised by investigators. The washing of arms with soap and rinsing with deionised water does not guarantee that/he skin is free frorr~ trace ciements. This was demonstrated by the Hohnadel et al study [3], in ~¢hich arm rinses after the washing procedure yielded trace amounts of copper, nickel, lead and zinc. In the present study, skin contar,:mmtion was monitored by the analysis of arm
120 TABLE 2 A comparison of sodium, potassium, calcium and lead levels in sweat in studies reporting sweat lead values Ref.
Method ~
Concentration (mmol 1- t ) Na ÷
I 3 10 13 Present study
A WB A A A WB A
Concentration (#g !- ~) K+
Ca: +
5.35 6.1 0.4 6.94
83 60 51 0.51 0.55 0.41 0.73
54 17 75 55.9 55 40.2
Pb
150 0 5.2
"A, ann bag sweat; WB, whole-body sweat.
rinses collected prior to sweating. Three subjects had traces of lead in arm rinses ranging from 4 to l0 pg i -~. Their sweat lead levels ranged from 10.5 to 30 pg ! -~. These values were excluded from the data. A mean sweat lead value of 5.2 #g l-~ (range 1.5- ! 3.0 ltg l-~) was recorded in this study for subjects with blood lead levels of between 6 and 13.6 pg dl-~. This is in agreement with the value of 7pgl -~ reported by Rabinowitz et al. [14] using radioisotopes. A comparison of the electrolyte content of sweat in studies reporting lead values (Table 2) indicates that sweat samples collected in this study are comparable to those in previous investigations. Urinary lead levels recorded in the present study are comparable to those reported by other workers [15,16], and these values indicate that the lead concentration in sweat is -~ 25% that in urine, a value close to the 20% re~orted by Rabinowitz et al. [14]. The lead metabolism model postulated by Rabinowitz and co-workers indicates that sweat and salivary lead originate frora the same pool. Saliva samples were collected in this study to determine wt, ether or not their lead content bore any relationship to those of sweat. However, the poor correlation betwen saliva and sweat lead (r = -0.1083) was only improved to r = 0.3669 with the exclusion of subjects 9 and 17, who had the highest sweat and saliva values, respectively. It is possible that with a larger number of subjects a stronger relationship could be demonstrated. The mean salivary lead level of 4.8 pgl -I (range 2.5-10 pgl -j ) found in this study is lower than some previously reported values of 54pgl -~ [17] and 31 t~g 1-~ [! 8], but is somewhat higher, though mbre comparable to, the mean value of 1.2/agl -~ reported by Rabinowitz et al. [i4]. Current methods of collection of saliva samples are so readily prone to contamination that the reliability of salivary lead levels as an indicator of body burden is doubtful. Sweat samples may be of some importance in the estimation of body burdens
121
of lead, but the risk of contamination of samples is such th~.~tthey may yield misleading results, more so in occupationally exposed persons. Within the limitation of the use of a l h sweat collection period in estimating daily lead losses in sweat, the results of this study indicate that sweat losses of lead may be of some significance in persons losing large amounts of sweat, as may occur in hot industries and hot climates. However, a study of long-term sweat collections of persons working or living in hot conditions is required to determine if such losses do occur and it"they influence the physiological significance of environmental exposure to lead. ACKNOWLEDGEMENTS
We would like to express our appreciation to Mr Chris Wells for assistance with the development of the method for sweat lead analysis, and to Mr John Mohiuddin for technical assistance. REFERENCES I J.R. Cohn and E.A. Emmett, The excretion of trace metals in human sweat, Ann. Clin. Lab. Sci., 8 (1978) 270-275. 2 S.J. Ritchey, M.K. Korslund, L.M. Gilbert, D.C. Fay and M.F. Robinson, Zinc retention and losses of zinc in sweat by pre-adolescent girls, Am. J. Ciin. Nutr., 32 {1979) 799-803. 3 D.C. Hohnadel, F.W. Sunderman Jr, M.W. Nechay and M.D. McNeel, Atomic absorption spectrometry of nickel, copper, zinc, and lead in sweat collected from healthy subjects during sauna bathing, Clin. Chem., 99 (1973) 1288-1292. 4 J.J. Putman and P.W. Madden, Quicksilver and slow death, Natl. Geographic, 142 {1972) 507-527. 5 D. Szadkowski, H.P. Ehrhardt, K.H. Schaller and G. Lehnert, Der zinc und kupferspiegcl im serum hitzeadaptierter schwerabeiter, Int. Arch. Gewerbepathol. Gewerbehyg., 25 {1969) 2{)2-214. 6 D.O. Sheils, Tlae elimination of lead in sweat, Australar. Ann. Med., 3 (1954) 225-229. 7 J.W. Robinson and S. Weiss, The direct determination of cadmium in urine and perspiration using a carbon bed atomizer for atomic adsorbtion spectroscopy, J. Environ. Sci. Health, A ! 5(6) (1980) 635-662. 8 K.J. Collins, T.P. Eddy, A. Hibbs, A.L. Stock and E.F. Wheeler. Nutritional and environmental studies on an ocean going oil tanker. 2. Heat acclimatization and nutrient balances, Br. J. had. Med., 28 (1971)246-257. 9 Analytical Methods Committee, Recommendations for the definition, estimation and use of the detection limit, Analyst, 112 (1987) 199-204. 10 M. Asayama, T. Ogawa, T. Morimoto, Y. Futiki and K. Naito, Excretion of heavy metals in sweat (Japanese), J. Aichi Med. Univ. Assoc., 3 (1975) 230-235. !! J.L. Stauber and T.M. Florence, A comparative study ofcopper, lead, cadmium and zinc in human sweat and blood, Sci. Total Environ., 74 (1988) 235-247. 12 E.C. Grant, J.M. Howard, S. Davies, H. Chasty, B. Hornsby and J. Galbriath, Zinc deficiency in children with dyslexia: concentrations of zinc and other minerals in sweat and hair, Br. Med. J. (Clin. Res.), 296 (1988) 607-609. 13 O.D. Vellar, Studies on Sw:at Losses of Nutrients {Dissertation) University ofOslo, Oslo, Norway, 1969.
122 14 M.B. Rabinowitz, G.W. Wetherill and J.D. Kopple, Kinetic analysis of lead metabolism in healthy humans, J. Ciin. Invest., 58 (1976) 260-270. 15 A. Bernard, J.P. Buchet, H. Roels, P. Mason and R. Lauwerys, Renal excretion ofproteins and enzymes in workers exposed to cadmium, Eur. J. Ciin. Invest., 9 (1979) I 1-12. 16 S. Selander and K. Cramer, Interrelationships between lead in blood, lead in urine and ALA in urine during lead work, Br. J. Ind. Med., 27 (1970) 28-39. !7 A.Y. P'an, Lead levels in saliva and in blood, J. Toxicol. Environ. Health, 7 (1981) 273-280. 18 G.J. DiGregorio, A.P. Fuko, R.G. Sample, E. Bobyock, R. McMichael and W.S. Chernick, Lead and 6-aminolevulinic acid concentrations in human parotid saliva, Toxicol. Appl. Pharmacol., 27 (1974) 491-493.
!!3
Lead in sweat and its relationship to salivary and urinary levels in normal healthy subjects* F.O. Omokhodion and G.W. Crockford Unit of Occupational Health, London School t!l"Hygibne and Tropical Medic#le, Keppel Street, London WCIE 7HT. United King&mr (Received January 24th, 1990; accepted March 6th, 1990)
ABSTRACT
Sweat was collected fl'om the arms of 24 normal healthy subjects while they sat in a hot chamber. Blood, urine and saliva samples were also collected. These were analyzed for lead by atomic absorption spectrophotometry. Sweat lead levels recorded in this study are lower than those previously reported. Subjects with mean blood lead levels of 8.6211gdl t (range 6-13.6) had mean sweat levels of 5.21~gl ~ (range !.5-13.0), .-. 25% of their urinary levels. Although salivary lead levels with a mean of 4.8 ltg I t (range 2.5-10) are comparable to sweat levels, their relationship to blood lead levels is poor (r = - 0 . 1 8 6 compared with r values of 0.7208 and 0.234 for sweat and urinary levels, respectively). INTRODUCTION
Indications from the result:, of nutritional balance studies suggest that nutrients are lost in sweat in such amounts that allowances have to be made for these losses in the calculation of daily requirements [1,2]. It is therel'ore possible thai sweating may relieve the body of toxic materials [3]. One of the earliest reports was on Spanish miners of cinnabar ore who had chronic signs of mercurialism and were put in a hot environment where forced sweating occurred; this resulted in a reduction of their symptoms and hence constituted a form of treatment [4]. More recently, Szadkowski et,al. [5] found lower blood copper levels in workmen chronically exposed to heat in a steel plant compared with controls who were not thus exposed. They attributed the thermally induced hypocupraemia to the excretion of copper in sweat. although sweat copper levels were not measured. The excretion of lead in the sweat of lead workers was demonstrated by Sheils in 1954 [6] and to ascertain whether or not contamination may be responsible for the levels recorded, he demonstrated measur~:ble amounts of *Based on a paper presented at tile International Symposiun'~ on Work in a Hot Environment and Hcat Related Disorders. Khartoum. 27-31 January 1988. 0048-9697/91/$03.50
~" 1991
)
,
t
- Elsevier Science ! ubhsners B.V.
ii4
lead in the sweat of a non-exposed subject which increased substantially with increasing oral doses of lead acetate. Several workers have since either refuted or confirmed the presence of lead in sweat, but the wide variation in reported values and incomplete data on blood and urinary lead levels in the subjects tested have done little to settle the question of the significance of sweating as an avenue for the loss of lead from the body. In this study, experiments were designed to quantify ehe io~e~ r~f lead in uncontaminated sweat of no~-occupationally exposed persons with known blood lead levels and to compare these losses with urinary losses. METHOD
In the majority of previous investigations, samples of sweat were collected from the skin by absorbing it on filter paper or gauze pads [6], scraping it from the skin or allowing it to drop freely into beakers or test tubes [7], using towels previously washed in deionised water to wipe sweat from the body surface [8], collecting it as it drops from the body into plastic sheets [l] or into plastic bags strapped around the arms [3]. There is a risk of contamination with each of these methods, from the skin and from the environment, including the materials used for sweat collection. The arm bag method was chosen for this study as the arm is easy to isolate for thorough cleaning and can be completely enclosed, thereby eliminating environmental contamination. Sweat was induced naturally by sitting subjects in a hot chamber at 40-45°C and 35% relative humidity for 1 h.
Subyects Twenty-four healthy men participated in the study. They were volunteers from among staff and students of the ~chool. Volunteers who had skin disease or a history of adverse reaction to heat were excluded from this study.
Sample collection Sweat, urine and saliva samples were collected in polythene bottles which had been soaked in 10% nitric acid for 48 h and then rinsed with deionised water.
Blood Blood samples (5 ml) were taken from the antecubital fossa with lithium heparin syringes. Urine Single voidings of urine were collected in polythene bottles immediately before the heat exposure.
!15
Saliva The mouth was rinsed thrice with 100ml of 3% citric acid and thrice with i 00 mi of deionised water. Saliva collected in the mouth was emptied into the polythene bottles provided. The procedure was repeated when necessary until 10 ml of saliva had been collected.
Sweat The subjects changed from day clothes into disposable overalls (Tyvek suits). Both sleeves were rolled up to the deltoid region and using a brush and lead-free soap (Phisohex), subjects throughly scrubbed nails, palms, hands and the arms. Both arms were rinsed with tap water and scrubbed again as before. They were then thoroughly rinsed with deionised distilled water. Excess water was shaken off the arms, subjects being careful not to touch any object until the arms were placed inside polythene bags made from a roll of heavy weight polythene tubing. The first set of arm bags were sealed at the distal ends and -- 50 ml of deionised water was poured into each bag and their upper ends were secured with tourniquets while each arm was rinsed inside the bag for --, 2 rain. These rinses were poured into polythene bottles and stored for analysis Excess water was shaken off the arms and each arm was placed inside an open-ended piece of polythene tubing which was then sealed at the wrists and the upper end of the arms with adhesive tape (Sleek). The subjects sat in the hot chamber for I h while body temperatures and heart rates were continuously monitored and recorded at 5-min intervals. At the end of the exposure period, sweat collected in the bags was poured into separate polythene bottles and stored for analy~s.
Sample analysis Storage apd handling Blood and urine samples were stored in a refrigerator until analysis. Sweat and, saliva sam01es were acidified to 1% with Aristar grade nitric acid to prevent bacterial action and then left to stand at room temperature until analysis, usually by the third day. In the handling of samples, from the collection stage to analysis, great care was taken to prevent contamination. All containers used for collection and storage of samples were tested and found to be free from lead. All glassware used for analysis was washed thoroughly, rinsed with 10% nitric acid and then rinsed with deionised water. Sample cups and caps were soaked in 10% nitric acid and rinsed in deionised water prior to use.
Analytical method All samples were analysed for lead using an atomic adsorption spectro-
116
photometer (Perkin-Elmer Model 460) equipped with a graphite furnace, an HG;~ 500 atomiser unit and deuterium background correction.
Blo,~d. Blood samples were rolled for 1 h to ensure an even mix. Samples were diluted 1 in 10 into 2ml sample cups using 0.1% TritonX and 0.11% NH,H_,HPO4 as diluent. The cups were covered and rolled for a further i h before analysis. ";;veat, saliva and urine samples. ,~n attempt to calibrate these samples against aqueous standards produced matrix effects which varied in magnitude among different individuals. Each sample was therefore analysed by the standard additions method (calibration by the addition of known amounts of lead to samples). Urine samples were analysed for creatinine. The temperature programme for lead determination was as follows: Step
Temp. (°C)
Ramp (s)
Hold Is)
I 2 3 4 5 6
100 130 140 650 1850 2200
10 5 5 20
5 5 10 25
I
5
3
I
Instrument conditions Wavelength Spectral bandwidth Lamp current Gas pressure Integration period Injection volurne Replicates
283.3 nm 0.7 nm 8 MA 45 psi (I psi = 6.9 kPa) 6s 20 ltl Three
Precision and accuracy At the time of this invest~$at~o,1, ,:he analytical qu~:lity control of the laboratory was monitored by participation in a national quality assurance scheme (UKEQAS). The analysis of internal quality assurance samples of sweat at the top range of values, i,e. 12.5~tg l -t, showed between-batch precision of 6.9% and within-batch precision of 3.0%. A within-batch precision of 18% was recorded for internal quality assurance samples containing 1.5 z~g l -i lead. Quality assurance analysis for urine did not yield consistent results after 2 months of storage and this was therefore discontinued. However, the precision of lead determinations in urine and saliva was demonstrated by regression coefficients of the standard additions curve varying from 0.•91 to 1.0. An-' sample with a coefficient below this value was reanalysed.
!!7
The limit of detection, as defined by the concentration of lead corresponding to three standard deviations of the response of 10 samples of the field blank [9] (1% nitric acid), was determined to be 1.1 p g l RESULTS
Although most subjects indicated a feeling of discomfort at the beginning of the exposure period, readings of heart rate and body temperature as well as subjective feelings of undue stress did not necessitate the termination of the experiments on any of the subjects. A mean total body sweat loss of 490 ml was recorded during the I-h exposure period, while a mean of 12.6 and 10.3 ml of sweat were collected from the left and right arms, respectively. The results of lead determination in left and right arm sweat samples were
TABLE
I
Lead concentrations in body lluids of non-occupationally cxposed persons Subject
I 2 3 4 5 t~ 7 8 9 10 !1 !2 13 14 15 16 17 18 19 20 21 22 23 24 Mean SD N
Blood
Swear'
Saliva
Urine
(/lgdl -I)
(/igl i)
(/~gl i)
(/:gl i)
(l~g/g creatinine)
14 26 44
7.7 ! 0. I 8.4
38 16 18 18 44 20 14 !6 16 26 18 30 20 16 10 26 18 20 12 16 21.56 9.43 23
13.9 10.2 3,4,4 11.8 23.6 24.q 8.4 28.3 il.2 19.6 13.3 31.6 '3'3 _..7 28.9 11.2 10.9 7.9 44.2 !q.3 9.7 17.92 10.16 23
9.0 6.0 9.8 7.9 8.0 6.8 9.0 111.5 13.6 6.0 8.3 9.3 8.3 6.0 12.0 8.0 7.0 10.0 8.0 7.0 9.8 10.4 10.2 6.0 8.62 1.96 24
5.5 3.0 6.0 6.0 3.25 4.75 6.0 13.0 4.0 7.25 5.0 2.25 1.5 4.75 5.75 2.5 9.5 4.25 9.25 3.5 3.0 5.23 2.74 2i
~Mean concentration of left and right arm sweat.
6.0 3.0 4.1) 2.5 4.0 5.9 5.5 4.5 2.5 6.0 4.0 3.5 2.5 2,5 6,5 10.0 6.0 8.5 7.5 4.5 2.5 4.80 2. ! I 21
118
Sweat Lead (pg/I)
14r
i I
i
/ .
L
/ •
I
t
/
/
t
//
.../..
•
•
•
•
4 " •
•
w
Z
i ! !
O k
0
......
I
20
40
60 80 Blood Lead (pg/I)
1. . . .
100
I
......
120
J
140
Fig. !. Lead in non-occupationally exposed persons: blood vs sweat.
similar. Sweat samples that were centrifuged before analysis yielded the same results as uncentrifuged samples. Samples analysed on the day of collection showed no difference in lead determinations between acidified and non-acidified sweat. Table ! shows the blood, sweat, saliva and urinary lead levels of 24 subjects. Sweat lead determinations were not recorded for subjects with arm rinses which showed traces of lead. Salivary samples with streaks of blood were discarded. Blood lead levels ranged from 6 to 13.6 #gdl- t, while corresponding sweat levels ranged from 1.5 to 13.0 #g l-t. Mean sweat lead levels were 6% of blood levels, while salivary levels were 5.5%. However, there is a stronger correlation between blood and sweat lead (Fig. 1) than between blood and saliva lead (Fig. 2); r = 0.7208 and r = -0.1864, respectively. The correlation between blood and urinary levels is poor (r = 0.234). The concentration of lead in sweat is ~ 25% of that in urine. DISCUSSION
Mean sweat lead levels in non-occupationally exposed subjects have been
!i9 Salivary Lead (pg/I) 12
10
~
,
~
ia
om
a
....1
I
• o •
0 0
1
1
20
40
n
I
. . . . .
•
I .
.
.
.
60 80 Blood Lead (ug/I)
.
I
I00
--
•
I .
120
.
.
.
140
Fig. 2. Lead in non-occupationally exposed persons: blood vs saliva.
reported a~ varying from 0 to 150ttgl -~. Mean arm bag sweat values of 83 #g 1-I [1] and 51 gg 1-I [3] have indicated that sweat and urine could contain equal amounts of lead, whereas values such as 1501~gl-~ [10] and estimates made by Shiels [6] from axillary pad sweat indicate that lead losses in sweat may be greater than those in the urine, lontophoresis generated sweat has also been reported to yield mean values of 24, 36 and 41/tg 1- ~[ 11], values around the range of urinary lead levels. Grant et al. [12] also reported a range of 10.9-42.0/~gl -~ (in 26 control children). Vellar [13], however, reported that lead, together with other heavy metals such as mercury, nickel and cobalt, were not present in measureable quantities in sweat. His finding was a good indication that these elements, and lead in this case, may be present in sweat in amounts much less than had previously been reported. The difficulty in collecting uncontaminated sweat is wel! recognised by investigators. The washing of arms with soap and rinsing with deionised water does not guarantee that/he skin is free frorr~ trace ciements. This was demonstrated by the Hohnadel et al study [3], in ~¢hich arm rinses after the washing procedure yielded trace amounts of copper, nickel, lead and zinc. In the present study, skin contar,:mmtion was monitored by the analysis of arm
120 TABLE 2 A comparison of sodium, potassium, calcium and lead levels in sweat in studies reporting sweat lead values Ref.
Method ~
Concentration (mmol 1- t ) Na ÷
I 3 10 13 Present study
A WB A A A WB A
Concentration (#g !- ~) K+
Ca: +
5.35 6.1 0.4 6.94
83 60 51 0.51 0.55 0.41 0.73
54 17 75 55.9 55 40.2
Pb
150 0 5.2
"A, ann bag sweat; WB, whole-body sweat.
rinses collected prior to sweating. Three subjects had traces of lead in arm rinses ranging from 4 to l0 pg i -~. Their sweat lead levels ranged from 10.5 to 30 pg ! -~. These values were excluded from the data. A mean sweat lead value of 5.2 #g l-~ (range 1.5- ! 3.0 ltg l-~) was recorded in this study for subjects with blood lead levels of between 6 and 13.6 pg dl-~. This is in agreement with the value of 7pgl -~ reported by Rabinowitz et al. [14] using radioisotopes. A comparison of the electrolyte content of sweat in studies reporting lead values (Table 2) indicates that sweat samples collected in this study are comparable to those in previous investigations. Urinary lead levels recorded in the present study are comparable to those reported by other workers [15,16], and these values indicate that the lead concentration in sweat is -~ 25% that in urine, a value close to the 20% re~orted by Rabinowitz et al. [14]. The lead metabolism model postulated by Rabinowitz and co-workers indicates that sweat and salivary lead originate frora the same pool. Saliva samples were collected in this study to determine wt, ether or not their lead content bore any relationship to those of sweat. However, the poor correlation betwen saliva and sweat lead (r = -0.1083) was only improved to r = 0.3669 with the exclusion of subjects 9 and 17, who had the highest sweat and saliva values, respectively. It is possible that with a larger number of subjects a stronger relationship could be demonstrated. The mean salivary lead level of 4.8 pgl -I (range 2.5-10 pgl -j ) found in this study is lower than some previously reported values of 54pgl -~ [17] and 31 t~g 1-~ [! 8], but is somewhat higher, though mbre comparable to, the mean value of 1.2/agl -~ reported by Rabinowitz et al. [i4]. Current methods of collection of saliva samples are so readily prone to contamination that the reliability of salivary lead levels as an indicator of body burden is doubtful. Sweat samples may be of some importance in the estimation of body burdens
121
of lead, but the risk of contamination of samples is such th~.~tthey may yield misleading results, more so in occupationally exposed persons. Within the limitation of the use of a l h sweat collection period in estimating daily lead losses in sweat, the results of this study indicate that sweat losses of lead may be of some significance in persons losing large amounts of sweat, as may occur in hot industries and hot climates. However, a study of long-term sweat collections of persons working or living in hot conditions is required to determine if such losses do occur and it"they influence the physiological significance of environmental exposure to lead. ACKNOWLEDGEMENTS
We would like to express our appreciation to Mr Chris Wells for assistance with the development of the method for sweat lead analysis, and to Mr John Mohiuddin for technical assistance. REFERENCES I J.R. Cohn and E.A. Emmett, The excretion of trace metals in human sweat, Ann. Clin. Lab. Sci., 8 (1978) 270-275. 2 S.J. Ritchey, M.K. Korslund, L.M. Gilbert, D.C. Fay and M.F. Robinson, Zinc retention and losses of zinc in sweat by pre-adolescent girls, Am. J. Ciin. Nutr., 32 {1979) 799-803. 3 D.C. Hohnadel, F.W. Sunderman Jr, M.W. Nechay and M.D. McNeel, Atomic absorption spectrometry of nickel, copper, zinc, and lead in sweat collected from healthy subjects during sauna bathing, Clin. Chem., 99 (1973) 1288-1292. 4 J.J. Putman and P.W. Madden, Quicksilver and slow death, Natl. Geographic, 142 {1972) 507-527. 5 D. Szadkowski, H.P. Ehrhardt, K.H. Schaller and G. Lehnert, Der zinc und kupferspiegcl im serum hitzeadaptierter schwerabeiter, Int. Arch. Gewerbepathol. Gewerbehyg., 25 {1969) 2{)2-214. 6 D.O. Sheils, Tlae elimination of lead in sweat, Australar. Ann. Med., 3 (1954) 225-229. 7 J.W. Robinson and S. Weiss, The direct determination of cadmium in urine and perspiration using a carbon bed atomizer for atomic adsorbtion spectroscopy, J. Environ. Sci. Health, A ! 5(6) (1980) 635-662. 8 K.J. Collins, T.P. Eddy, A. Hibbs, A.L. Stock and E.F. Wheeler. Nutritional and environmental studies on an ocean going oil tanker. 2. Heat acclimatization and nutrient balances, Br. J. had. Med., 28 (1971)246-257. 9 Analytical Methods Committee, Recommendations for the definition, estimation and use of the detection limit, Analyst, 112 (1987) 199-204. 10 M. Asayama, T. Ogawa, T. Morimoto, Y. Futiki and K. Naito, Excretion of heavy metals in sweat (Japanese), J. Aichi Med. Univ. Assoc., 3 (1975) 230-235. !! J.L. Stauber and T.M. Florence, A comparative study ofcopper, lead, cadmium and zinc in human sweat and blood, Sci. Total Environ., 74 (1988) 235-247. 12 E.C. Grant, J.M. Howard, S. Davies, H. Chasty, B. Hornsby and J. Galbriath, Zinc deficiency in children with dyslexia: concentrations of zinc and other minerals in sweat and hair, Br. Med. J. (Clin. Res.), 296 (1988) 607-609. 13 O.D. Vellar, Studies on Sw:at Losses of Nutrients {Dissertation) University ofOslo, Oslo, Norway, 1969.
122 14 M.B. Rabinowitz, G.W. Wetherill and J.D. Kopple, Kinetic analysis of lead metabolism in healthy humans, J. Ciin. Invest., 58 (1976) 260-270. 15 A. Bernard, J.P. Buchet, H. Roels, P. Mason and R. Lauwerys, Renal excretion ofproteins and enzymes in workers exposed to cadmium, Eur. J. Ciin. Invest., 9 (1979) I 1-12. 16 S. Selander and K. Cramer, Interrelationships between lead in blood, lead in urine and ALA in urine during lead work, Br. J. Ind. Med., 27 (1970) 28-39. !7 A.Y. P'an, Lead levels in saliva and in blood, J. Toxicol. Environ. Health, 7 (1981) 273-280. 18 G.J. DiGregorio, A.P. Fuko, R.G. Sample, E. Bobyock, R. McMichael and W.S. Chernick, Lead and 6-aminolevulinic acid concentrations in human parotid saliva, Toxicol. Appl. Pharmacol., 27 (1974) 491-493.