- Email: [email protected]
The interpretation of human hair trace element concentrations
The Science of the Total Environment, 39 (1984) 93--101 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands
93
THE INTERPRETATION OF HUMAN HAIR TRACE ELEMENT CONCENTRATIONS
ROSALIND S. GIBSON
Applied Human Nutrition, Department of Family Studies, University of Guelph, Guelph, Ontario, NIG 2W1 (Canada) IAN L. GIBSON
Department of Earth Sciences, University of Waterloo, Waterloo, Ontario, N2L 3G1 (Canada) (Received February 1st, 1984, accepted March 16th 1984)
ABSTRACT Scalp hair samples collected from 86 Canadian elderly, non-institutionalized women (mean age 66.6 -+ 6.2 y) on two separate occasions ten weeks apart, were analyzed for Zn, Cu and Mn content. Fasting blood samples were also collected and serum Zn, Cu and albumin content determined. Median hair Zn, Cu and Mn and mean serum Zn and albumin concentrations were not siginificantly different at the pre- and post-study periods. Furthermore, hair Zn concentrations at the two sampling times were positively and significantly correlated as were hair Cu, hair Mn, serum Zn and serum albumin values. In contrast no significant correlation related the hair and serum Zn values and hair and serum Cu values at any time. The constancy of the hair Zn, Cu and Mn concentrations at the two sampling periods may reflect the homeostatic regulation which controls absorption and hence body trace element content. In addition, results indicate that, after careful laboratory washing, the effects of adventitious contamination on hair trace element content are small, and can be effectively ignored.
INTRODUCTION T h e use o f scalp hair as a b i o p s y material f o r the assessment o f trace e l e m e n t status in h u m a n s has received m u c h a t t e n t i o n in r e c e n t y e a r s [1, 2 ] . This is because sampling is relatively non-invasive a n d h e n c e samples are easily o b t a i n e d a n d stored. M o r e o v e r trace e l e m e n t s in hair are m o r e conc e n t r a t e d t h a n in b l o o d , t h u s facilitating analysis. One o f t h e possible p r o b l e m s w i t h t h e use o f scalp hair is its susceptibility t o e x o g e n o u s cont a m i n a n t s such as a t m o s p h e r i c p o l l u t a n t s , water, a n d sweat [ 2 ] . A wide v a r i e t y o f hair b e a u t y t r e a t m e n t s m a y also m o d i f y t h e trace e l e m e n t c o m p o sition o f hair [ 3 ] . A large n u m b e r o f w a s h i n g p r o c e d u r e s have been d e v e l o p e d f o r r e m o v i n g c o m p l e t e l y a n d exclusively a d v e n t i t i o u s c o n t a m i n a t i o n [4, 5 ] . H o w e v e r , t h e i r effectiveness has been q u e s t i o n e d b y m a n y investigators [ 3,6 ]. 0048-9697/84/$03.00
© 1984 Elsevier Science Publishers B.V.
94 To date most of the studies of hair trace element concentrations of adult population groups have been cross sectional [7, 8] ; very few investigators have analyzed hair samples taken on more than one occasion from a large number of adult subjects. This is unfortunate as valuable information on the significance of hair trace elements levels and their long term variability with time can be obtained utilizing this approach. The effects of differing hair beauty treatments also become apparent. An alternative procedure is to analyze along the length of individual hair strands using Proton-Induced X-ray Emission Spectrometry (PIXE) [9]. However, this requires very sophisticated instrumentation. We therefore analyzed the zinc, copper and manganese content of hair samples collected from the occipital region of the scalp, on two separate occasions ten weeks apart, from 86 elderly Canadian women. The relationship between serum zinc and hair zinc, and serum copper and hair copper concentrations were also examined.
EXPERIMENTAL Hair samples, cut close to the occipital region of the scalp with teflon coated stainless steel scissors, were collected from 86 Canadian, noninstitutionalized elderly w o m e n (mean age 66.6 + 6.2; range 58--89 y) consuming se[t-selected diets, on two separate occasions, exactly ten weeks apart. Only the proximal 1--1.5 cm were retained for analysis and as hair grows on average 1 cm per month [2], this time interval ensured that t w o separate periods of hair growth were sampled. Hair samples were washed with a nonionic detergent solution (Actinox-Sherwood Medical Industries) using a modification of the procedure described by Harrison et al. [10] published earlier [11]. Hair zinc, copper and manganese analyses were performed by instrumental neutron activation procedures (INAA), described previously [11]. The accuracy and precision of the INAA procedures was studied by analyzing six aliquots of an International Atomic Energy Authority powdered hair sample. The mean + S D for zinc, copper and manganese were: 168 ± 10.2, 10.2 ± 0.7, 0.89 + 0 . 1 0 p p m compared to published values of: zinc 174 ppm; copper 102 ppm; manganese 0.85 ppm [12]. Details of the-health status of the subjects, use of vitamin and mineral supplements, and medications prescribed, were obtained both from the subjects and their respective physicians. In addition information on the brands of shampoos used, hair beauty treatments and use of softened and unsoftened water were also obtained via a questionnaire. More than half of the subjects had hair beauty treatments during the ten week study period. The study protocol was approved by tne tlniversity of Guelph Human Ethics Committee and written consent of each subject obtained. Overnight fasting blood samPles were taken at the same time as the hair samples, wa venipuncture, using trace-element free vacutamers (Becton-Dickinson) with subjects in the recumbent position. After clotting,
95 blood samples were centrifuged and serum separated using acid washed glass pipets. The serum was then frozen at -- 20°C. Serum zinc and copper analyses were p e r f o r m e d by flame and flameless atomic absorption procedures respectively, using a Perkin-Elmer AA s p e c t r o p h o t o m e t e r (model 372) fitted with a graphite furnace a t t a c h m e n t (HGA 2100). T he m e t h o d of Fuw a et al. [13] was used f or ihe serum zinc analyses. Serum copper concentrations were d et e r m i ne d via the m e t h o d of Evenson and Warren [ 1 4 ] . For the atomic absorption procedures, a pool ed serum and a quality cont rol serum (Cation-Cal; Dade Chemical Co.) were included with each batch of serum samples to assess the precision and accuracy of the methods, respectively. Mean values f or 4 analyses of p o o l e d serum were: zinc 14.8 + 0.2 pmol/1; co p p er 19.4 + 2.5/~mol/1. Mean values for 5 analyses of Cation-Cal were zinc, 46.5 + 0.5 pmol/1; copper, 28.6 + 2.3 pmol/1 c o m p a r e d to certified values of zinc, 46.2pmol/1; copper, 3 0 . 3 p m o l / l . Serum albumin concentrations were also d e t e r m i n e d on the serum samples collected at the beginning and the end of the ten week period using the m e t h o d of R o d k e y [ 1 5 ] . A pool ed serum and quality c ont r ol serum (Monitrol; Dade Chemical C o . ) w e r e also used. The mean + SD f or 18 analyses of a p o o l e d serum and quality cont rol serum were 0.44 + 0.02 and 0.41 + 0.02 g/l, respectively. The certified value for Monitrol was 0.40 g/1. Hair trace e l e m ent levels are n o t usually normally distributed [ 1 6 ] , so th at non-parametric statistics were e m p l o y e d , the median being used to indicate central t e n d e n c y . The non-parametric Spearman rank coefficient was used as a measure of correlation and the Wilcoxon m a t c h e d pairs test used to assess differences in hair trace element concentrations between sampling periods.
RESULTS
Descriptive statistics f or hair trace e l e m e n t concentrations and serum zinc, co p p er and albumin values are shown in Table 1 and Table 2, respectively. TABLE 1 PRE- AND POST-STUDY HAIR ZINC, COPPER AND MANGANESE CONCENTRATIONS (ppm) Mean ± SD
Pre-study Zn Post-study Zn Pre-study Cu Post-study Cu Pre-study Mn
Post-study Mn
155 151 14.9 14.5
± 42 ± 36 ± 9.0 ± 7.5 0.26 ± 0.26 0.27 ± 0.24
Median
156 152 12.8 12.8 0.16 0.20
Quartile 25th
75th
131 130 10.8 10.7 0.10 0.12
183 180 16.0 15.5 0.31 0.33
96
TABLE 2 PRE- AND POST-STUDY SERUM ZINC, COPPER AND ALBUMIN CONCENTRATIONS Pre-Study
Post-Study
S e r u m Zinc ~ m o l / l
16.6 + 2.0
16.9 +
Serum Copper/~mol/l Albumin g/1
19.3 + 1.9 0.42 + 0.04
0.42 + 0.04
1.9
Even though two separate hair samples representing t w o different periods of hair growth were analyzed, median hair zinc, copper and manganese concentrations were n o t significantly different at the pre- and post study periods. Furthermore the hair zinc concentrations at the t w o sampling times were positively correlated (r = 0.60, p < 0.0002; Fig. 1). The hair copper values showed a similar correlation (r = 0 . 5 9 , p <: 0.0002; Fig. 2). and so did the hair manganese levels (r = 0.26, p < 0.03; Fig. 3). Hair trace element levels were independent of age, type of hair beauty treatment and shampoos, consumption of soft or hard water and use of vitamin and mineral supplements. The latter were generally iron supplements without appreciable levels of zinc, copper and manganese. The mean serum zinc and albumin concentrations were also comparable at the beginning and end of the study. Again highly significant positive
225-
•
8o
•
Pre •
• •
••
150-
•
• °°°° o• e• • °o
°o
0o" ° -
Zn
•
~°me•
•
o•
08
•
75-
7'5
1~0
Post
2~5
Fig. 1. Correlation of pre- and p o s t - s t u d y hair zinc concentrations (ppm).
97
Pre
Cu
35-
25'
"'*
15"
:l
.:
15
2'5
Post
315
Fig. 2. Correlation of pre- and post-study hair copper concentrations (ppm). 1.5-
Pre
Mn
1'0-
0~,-
o o
• °oeCm
o •
• o
•
e °
•
0'.5
1!0
Post
1.'5
Fig. 3. Correlation of pre- and post-study hair manganese concentrations (ppm).
98 correlations were noted for pre- and post-study serum zinc concentrations {r = 0.51, p < 0.0002) and the pre- and post serum albumin values (r = 0.51, p < 0.0002). In addition serum zinc and serum albumin values correlated positively at both times, although these correlations were only significant at the end of the study (r = 0.40, p < 0.003). In contrast, no significant correlations related the hair and serum zinc values at any time. Similarly no relationship was found between the hair and serum copper values.
DISCUSSION Evidence from both animal and human studies suggest zinc absorption is homeostatically regulated [17, 18]. As a result serum zinc, and perhaps b o d y zinc content remains at a relatively constant level. We suggest that the constancy of the hair zinc concentrations of the adult subjects of this study reflects this homeostatic regulation. The strong positive correlation of the pre- and post-study hair zinc concentrations is taken to indicate that the level of zinc regulation varies from subject to subject. We suggest that the same regulatory mechanism controls the copper and manganese content of the body. The absence of any positive correlation between hair zinc and serum zinc and hair copper and serum copper concentrations, noted in this study and by others [7, 8, 19] has been attributed to differences in the time scales reflected by the two indices [20]. Serum concentrations provide an acute index related to the supply of trace elements over a very short time period but are affected by many other factors such as infection, stress, diurnal variation etc. [21]. On the other hand, hair trace element content provides a retrospective index of trace element status during a relatively longer period of hair growth. However, the serum zinc concentrations in this study were comparable at the beginning and end of the ten week period suggesting that the effect, if any, of these complicating short term factors on serum zinc values was not very significant. Nevertheless the serum zinc concentrations did n o t correlate with hair zinc values. Instead, serum zinc correlated positively with serum albumin at both sampling times, a finding noted by others [22]. As 60%-70% of zinc is loosely b o u n d to albumin in serum [23], this relationship is n o t surprising. It is possible that the lack of any correlation between total serum zinc and hair zinc content is because the latter is related to one of the smaller nonalbumin-bound fractions of zinc in serum, such as that b o u n d to alpha-2macroglobulin [23] or to amino acids [24]. This would account for the lack of any gross correlation between hair zinc and serum zinc levels. We further conclude from our study that by sampling and washing hair samples by standardized procedures, effects of any adventitious trace element contamination are small and can be effectively ignored. Furthermore, as no
99
relationship was f o u n d between hair zinc, copper and manganese concentrations and type of hair beauty treatment, the latter may n o t have such a significant effect on these hair trace element values as suggested by other investigators [ 3, 7 ]. In general exogenous contamination does n o t appear to be the reason for the lack of correlation between hair and serum trace element content. DeAntonio et al. [7] showed no significant differences between the mean trace element concentrations of scalp and pubic hair samples taken from a group of 67 subjects. These findings are of particular interest, in view of the marked differences in the rate of hair growth in the two anatomical sites. Hopps [2] has suggested that pubic hair has an anagen and telogen phase of equal duration in contrast to that of scalp hair which has anagen and telogen phases of 900 and 100 days, respectively. If this is the case, the results of DeAntonio et al. [7] suggest that the relative rates of hair growth are not significant factors in controlling hair trace element levels. In severe zinc deficiency states, e.g. acrodermatitus enteropathica [25], and in some cases of severe protein energy nutrition [26], hair growth is impaired. However, hair zinc concentrations are often normal and there is no correlation between hair and serum zinc levels. If impairment of hair growth is n o t an important factor in controlling hair zinc levels, we speculate that this lack of correlation may be due to a relatively minor zinc fraction of the serum regulating hair zinc concentrations. Certainly changes in the very small concentrations of non-essential heavy metals in blood or serum are readily reflected by dramatic changes in hair heavy metal concentrations [27]. To date there is very little information on correlations of human hair trace element levels with components of blood other than serum or plasma. Correlations of hair trace element levels with concentrations in other organs, tissues, or b o d y pools are also poorly known. As emphasized by Hambidge [ 2 8 ] , these data are urgently required in order to better understand the significance of hair trace element concentrations.
ACKNOWLEDGEMENTS This work was supported by the Human Nutrition Research Council of Ontario.
REFERENCES 1 L.M. Klevay, Hair as a biopsy material. 1. Assessment of zinc nutriture, Am. J. Clin. Nutr., 23 (1970) 284--289. 2 H.C. Hopps, The biologic bases for using hair and nails for analyses of trace elements, Sci. Total Environ., 7 (1977) 71--89. 3 J.M. McKenzie, Alteration of the zinc and copper concentration of hair, Am. J. Clin. Nutr., 31 (1978) 470--476.
100 4 S. Salmela, E. Vuori and J.O Kilpio, The effect of washing procedures on trace element content of human hair, Anal. Chim. Acta, 125 (1981) 131--137. 5 G.S. Assarian and D. Oberleas, Effect of washing procedures on trace element content of hair, Clin. Chem., 23 (1977) 1771--1772. 6 K.M. Hambidge, M.L. Franklin and J.A. Jacobs, Hair chromium concentrations: Effect of sample washing and external environment, Am. J. Clin. Nutr., 25 (1972) 384--389. 7 S.M. DeAntonio, S.A. Katz, D.M. Scheiner and J.D. Wood, Anatomically related variations in trace-metal concentrations in hair, Clin. Chem., 28 (1982) 2411--2413. 8 S.C. Vir and A.H.G. Love, Zinc and copper status of the elderly, Am. J. Clin. Nutr., 32 (1979) 1472--1476. 9 J.L. Campbell, S. Faiq, R.S. Gibson, S.B. Russell and C.W. Schulte, Determination of trace element profiles and concentrations in human hair by proton-induced X-ray emission spectrometry, Anal. Chem., 53 (1981) 1249-=1253. ° 10 W.W. Harrison, J.P. Yurachek and C.A. Benson, The determination of trace elements in human hair by atomic absorption spectroscopy, Clin. Chim. Acta, 23 (1969) 83--91. 11 R.S. Gibson and M.S. DeWolfe, Copper, zinc, manganese, vanadium and iodine concentrations in hair in Canadian neonates, Pediatr. Res., 13 (1979) 959--962. 12 S.B. M'Baku and R.M. Parr, Interlaboratory study of trace and other elements in the IAEA powdered human hair reference material, HH-1, J. Radioanal. Chem., 69 (1982) 171--180. 13 P.P. Fuwa, R. McKay and B.L. Valee, Determination of zinc in biological materials by atomic absorption spectroscopy. Anal. Chem., 36 (1964) 2407--2411. 14 M.A. Evenson and B.L. Warren, Determination of serum copper by atomic absorption with the use of the graphite cuvette, Clin. Chem., 21 (1975) 619--625. 15 F.L. Rodky, Direct spectrophotometric determination of albumin in human serum, Clin. Chem., 11 (1965) 478--487. 16 R.F. Coleman, F.H. Cripps, A. Stimson and H.D. Scott, The determination of trace elements in human hair by neutron activation and the application of forensic science, Atomic Weapons Research Establishment, Aldermaston Report No. 0-86/66, 1966. 17 E. Weigand and M. Kirchgessner, Total true efficiency of zinc utilization: Determination and homeostatic dependence upon the zinc supply status in young rats, J. Nutr., 110 (1980) 469--480. 18 J.H. Freeland-Graves, M.L. Ebangit and P.J. Hendrikson, Alteration in zinc absorption and salivary sediment zinc after a lacto-ovo-vegetarian diet, Am. J. Clin. Nutr., 33 (1980) 1757--1766. 19 L.D. McBean, M. Mahloudji, J.G. Reinhold and J.A. Halsted, Correlation of zinc concentrations in human plasma and hair, Am. J. Clin. Nutr., 24 (1971) 506--509. 20 M. Laker, On determining trace element levels in man: the use of blood and hair, Lancet, 8292, vol. 2 ( 1 9 8 2 ) 260--262. 21 N.W. Solomons, On the assessment of zinc and copper nutriture in man, Am. J. Clin. Nutr., 32 (1979) 856--871. 22 E.M. MacMillan and D.J. Rowe, Plasma zinc-serum albumin correlation: Relevance to assessment of zinc status in humans, Clin. Exp. Derm., 7 (1982) 599--604. 23 A.F. Parisi and B.L. Vallee, Isolation of a zinc alpha-2-macroglobulin from human serum, Biochem., 9 (1970) 2421--2426. 24 E.L. Giroux and R.I. Henkin, Competition for zinc among serum albumin and amino acids, Biochim. Biophys. Acta, 273 (1972) 64--72. 25 K.M. Hambidge, P.A. Walravens and K.H. Neldner, The role of zinc in the pathogenesis and treatment of acriodermatitis enteropathica, in: G.J. Brewer, A.S. Prasad (Eds.), Zinc Metabolism: Current Aspects in Health and Disease, Alan R. Liss, New York, 1977, pp. 329--340. 26 J. Ertern, A. Arcasoy, A.O. Caudar and S. Cin, Hair zinc levels in healthy and malnourished children, Am. J. Clin. Nutr., 31 (1978) 1172--1174.
101 27 A. Olguin, P. Jauge, M. Cebrian and A. Albores, Arsenic levels in blood, urine, hair and nails from a chronically exposed human population, Proc. West Pharmacol. Soc., 26 (1983) 175--177. 28 K.M. Hambidge, Hair analysis: worthless for vitamins, limited for minerals, Am. J. Clin. Nutr., 36 (1982) 943--949.
93
THE INTERPRETATION OF HUMAN HAIR TRACE ELEMENT CONCENTRATIONS
ROSALIND S. GIBSON
Applied Human Nutrition, Department of Family Studies, University of Guelph, Guelph, Ontario, NIG 2W1 (Canada) IAN L. GIBSON
Department of Earth Sciences, University of Waterloo, Waterloo, Ontario, N2L 3G1 (Canada) (Received February 1st, 1984, accepted March 16th 1984)
ABSTRACT Scalp hair samples collected from 86 Canadian elderly, non-institutionalized women (mean age 66.6 -+ 6.2 y) on two separate occasions ten weeks apart, were analyzed for Zn, Cu and Mn content. Fasting blood samples were also collected and serum Zn, Cu and albumin content determined. Median hair Zn, Cu and Mn and mean serum Zn and albumin concentrations were not siginificantly different at the pre- and post-study periods. Furthermore, hair Zn concentrations at the two sampling times were positively and significantly correlated as were hair Cu, hair Mn, serum Zn and serum albumin values. In contrast no significant correlation related the hair and serum Zn values and hair and serum Cu values at any time. The constancy of the hair Zn, Cu and Mn concentrations at the two sampling periods may reflect the homeostatic regulation which controls absorption and hence body trace element content. In addition, results indicate that, after careful laboratory washing, the effects of adventitious contamination on hair trace element content are small, and can be effectively ignored.
INTRODUCTION T h e use o f scalp hair as a b i o p s y material f o r the assessment o f trace e l e m e n t status in h u m a n s has received m u c h a t t e n t i o n in r e c e n t y e a r s [1, 2 ] . This is because sampling is relatively non-invasive a n d h e n c e samples are easily o b t a i n e d a n d stored. M o r e o v e r trace e l e m e n t s in hair are m o r e conc e n t r a t e d t h a n in b l o o d , t h u s facilitating analysis. One o f t h e possible p r o b l e m s w i t h t h e use o f scalp hair is its susceptibility t o e x o g e n o u s cont a m i n a n t s such as a t m o s p h e r i c p o l l u t a n t s , water, a n d sweat [ 2 ] . A wide v a r i e t y o f hair b e a u t y t r e a t m e n t s m a y also m o d i f y t h e trace e l e m e n t c o m p o sition o f hair [ 3 ] . A large n u m b e r o f w a s h i n g p r o c e d u r e s have been d e v e l o p e d f o r r e m o v i n g c o m p l e t e l y a n d exclusively a d v e n t i t i o u s c o n t a m i n a t i o n [4, 5 ] . H o w e v e r , t h e i r effectiveness has been q u e s t i o n e d b y m a n y investigators [ 3,6 ]. 0048-9697/84/$03.00
© 1984 Elsevier Science Publishers B.V.
94 To date most of the studies of hair trace element concentrations of adult population groups have been cross sectional [7, 8] ; very few investigators have analyzed hair samples taken on more than one occasion from a large number of adult subjects. This is unfortunate as valuable information on the significance of hair trace elements levels and their long term variability with time can be obtained utilizing this approach. The effects of differing hair beauty treatments also become apparent. An alternative procedure is to analyze along the length of individual hair strands using Proton-Induced X-ray Emission Spectrometry (PIXE) [9]. However, this requires very sophisticated instrumentation. We therefore analyzed the zinc, copper and manganese content of hair samples collected from the occipital region of the scalp, on two separate occasions ten weeks apart, from 86 elderly Canadian women. The relationship between serum zinc and hair zinc, and serum copper and hair copper concentrations were also examined.
EXPERIMENTAL Hair samples, cut close to the occipital region of the scalp with teflon coated stainless steel scissors, were collected from 86 Canadian, noninstitutionalized elderly w o m e n (mean age 66.6 + 6.2; range 58--89 y) consuming se[t-selected diets, on two separate occasions, exactly ten weeks apart. Only the proximal 1--1.5 cm were retained for analysis and as hair grows on average 1 cm per month [2], this time interval ensured that t w o separate periods of hair growth were sampled. Hair samples were washed with a nonionic detergent solution (Actinox-Sherwood Medical Industries) using a modification of the procedure described by Harrison et al. [10] published earlier [11]. Hair zinc, copper and manganese analyses were performed by instrumental neutron activation procedures (INAA), described previously [11]. The accuracy and precision of the INAA procedures was studied by analyzing six aliquots of an International Atomic Energy Authority powdered hair sample. The mean + S D for zinc, copper and manganese were: 168 ± 10.2, 10.2 ± 0.7, 0.89 + 0 . 1 0 p p m compared to published values of: zinc 174 ppm; copper 102 ppm; manganese 0.85 ppm [12]. Details of the-health status of the subjects, use of vitamin and mineral supplements, and medications prescribed, were obtained both from the subjects and their respective physicians. In addition information on the brands of shampoos used, hair beauty treatments and use of softened and unsoftened water were also obtained via a questionnaire. More than half of the subjects had hair beauty treatments during the ten week study period. The study protocol was approved by tne tlniversity of Guelph Human Ethics Committee and written consent of each subject obtained. Overnight fasting blood samPles were taken at the same time as the hair samples, wa venipuncture, using trace-element free vacutamers (Becton-Dickinson) with subjects in the recumbent position. After clotting,
95 blood samples were centrifuged and serum separated using acid washed glass pipets. The serum was then frozen at -- 20°C. Serum zinc and copper analyses were p e r f o r m e d by flame and flameless atomic absorption procedures respectively, using a Perkin-Elmer AA s p e c t r o p h o t o m e t e r (model 372) fitted with a graphite furnace a t t a c h m e n t (HGA 2100). T he m e t h o d of Fuw a et al. [13] was used f or ihe serum zinc analyses. Serum copper concentrations were d et e r m i ne d via the m e t h o d of Evenson and Warren [ 1 4 ] . For the atomic absorption procedures, a pool ed serum and a quality cont rol serum (Cation-Cal; Dade Chemical Co.) were included with each batch of serum samples to assess the precision and accuracy of the methods, respectively. Mean values f or 4 analyses of p o o l e d serum were: zinc 14.8 + 0.2 pmol/1; co p p er 19.4 + 2.5/~mol/1. Mean values for 5 analyses of Cation-Cal were zinc, 46.5 + 0.5 pmol/1; copper, 28.6 + 2.3 pmol/1 c o m p a r e d to certified values of zinc, 46.2pmol/1; copper, 3 0 . 3 p m o l / l . Serum albumin concentrations were also d e t e r m i n e d on the serum samples collected at the beginning and the end of the ten week period using the m e t h o d of R o d k e y [ 1 5 ] . A pool ed serum and quality c ont r ol serum (Monitrol; Dade Chemical C o . ) w e r e also used. The mean + SD f or 18 analyses of a p o o l e d serum and quality cont rol serum were 0.44 + 0.02 and 0.41 + 0.02 g/l, respectively. The certified value for Monitrol was 0.40 g/1. Hair trace e l e m ent levels are n o t usually normally distributed [ 1 6 ] , so th at non-parametric statistics were e m p l o y e d , the median being used to indicate central t e n d e n c y . The non-parametric Spearman rank coefficient was used as a measure of correlation and the Wilcoxon m a t c h e d pairs test used to assess differences in hair trace element concentrations between sampling periods.
RESULTS
Descriptive statistics f or hair trace e l e m e n t concentrations and serum zinc, co p p er and albumin values are shown in Table 1 and Table 2, respectively. TABLE 1 PRE- AND POST-STUDY HAIR ZINC, COPPER AND MANGANESE CONCENTRATIONS (ppm) Mean ± SD
Pre-study Zn Post-study Zn Pre-study Cu Post-study Cu Pre-study Mn
Post-study Mn
155 151 14.9 14.5
± 42 ± 36 ± 9.0 ± 7.5 0.26 ± 0.26 0.27 ± 0.24
Median
156 152 12.8 12.8 0.16 0.20
Quartile 25th
75th
131 130 10.8 10.7 0.10 0.12
183 180 16.0 15.5 0.31 0.33
96
TABLE 2 PRE- AND POST-STUDY SERUM ZINC, COPPER AND ALBUMIN CONCENTRATIONS Pre-Study
Post-Study
S e r u m Zinc ~ m o l / l
16.6 + 2.0
16.9 +
Serum Copper/~mol/l Albumin g/1
19.3 + 1.9 0.42 + 0.04
0.42 + 0.04
1.9
Even though two separate hair samples representing t w o different periods of hair growth were analyzed, median hair zinc, copper and manganese concentrations were n o t significantly different at the pre- and post study periods. Furthermore the hair zinc concentrations at the t w o sampling times were positively correlated (r = 0.60, p < 0.0002; Fig. 1). The hair copper values showed a similar correlation (r = 0 . 5 9 , p <: 0.0002; Fig. 2). and so did the hair manganese levels (r = 0.26, p < 0.03; Fig. 3). Hair trace element levels were independent of age, type of hair beauty treatment and shampoos, consumption of soft or hard water and use of vitamin and mineral supplements. The latter were generally iron supplements without appreciable levels of zinc, copper and manganese. The mean serum zinc and albumin concentrations were also comparable at the beginning and end of the study. Again highly significant positive
225-
•
8o
•
Pre •
• •
••
150-
•
• °°°° o• e• • °o
°o
0o" ° -
Zn
•
~°me•
•
o•
08
•
75-
7'5
1~0
Post
2~5
Fig. 1. Correlation of pre- and p o s t - s t u d y hair zinc concentrations (ppm).
97
Pre
Cu
35-
25'
"'*
15"
:l
.:
15
2'5
Post
315
Fig. 2. Correlation of pre- and post-study hair copper concentrations (ppm). 1.5-
Pre
Mn
1'0-
0~,-
o o
• °oeCm
o •
• o
•
e °
•
0'.5
1!0
Post
1.'5
Fig. 3. Correlation of pre- and post-study hair manganese concentrations (ppm).
98 correlations were noted for pre- and post-study serum zinc concentrations {r = 0.51, p < 0.0002) and the pre- and post serum albumin values (r = 0.51, p < 0.0002). In addition serum zinc and serum albumin values correlated positively at both times, although these correlations were only significant at the end of the study (r = 0.40, p < 0.003). In contrast, no significant correlations related the hair and serum zinc values at any time. Similarly no relationship was found between the hair and serum copper values.
DISCUSSION Evidence from both animal and human studies suggest zinc absorption is homeostatically regulated [17, 18]. As a result serum zinc, and perhaps b o d y zinc content remains at a relatively constant level. We suggest that the constancy of the hair zinc concentrations of the adult subjects of this study reflects this homeostatic regulation. The strong positive correlation of the pre- and post-study hair zinc concentrations is taken to indicate that the level of zinc regulation varies from subject to subject. We suggest that the same regulatory mechanism controls the copper and manganese content of the body. The absence of any positive correlation between hair zinc and serum zinc and hair copper and serum copper concentrations, noted in this study and by others [7, 8, 19] has been attributed to differences in the time scales reflected by the two indices [20]. Serum concentrations provide an acute index related to the supply of trace elements over a very short time period but are affected by many other factors such as infection, stress, diurnal variation etc. [21]. On the other hand, hair trace element content provides a retrospective index of trace element status during a relatively longer period of hair growth. However, the serum zinc concentrations in this study were comparable at the beginning and end of the ten week period suggesting that the effect, if any, of these complicating short term factors on serum zinc values was not very significant. Nevertheless the serum zinc concentrations did n o t correlate with hair zinc values. Instead, serum zinc correlated positively with serum albumin at both sampling times, a finding noted by others [22]. As 60%-70% of zinc is loosely b o u n d to albumin in serum [23], this relationship is n o t surprising. It is possible that the lack of any correlation between total serum zinc and hair zinc content is because the latter is related to one of the smaller nonalbumin-bound fractions of zinc in serum, such as that b o u n d to alpha-2macroglobulin [23] or to amino acids [24]. This would account for the lack of any gross correlation between hair zinc and serum zinc levels. We further conclude from our study that by sampling and washing hair samples by standardized procedures, effects of any adventitious trace element contamination are small and can be effectively ignored. Furthermore, as no
99
relationship was f o u n d between hair zinc, copper and manganese concentrations and type of hair beauty treatment, the latter may n o t have such a significant effect on these hair trace element values as suggested by other investigators [ 3, 7 ]. In general exogenous contamination does n o t appear to be the reason for the lack of correlation between hair and serum trace element content. DeAntonio et al. [7] showed no significant differences between the mean trace element concentrations of scalp and pubic hair samples taken from a group of 67 subjects. These findings are of particular interest, in view of the marked differences in the rate of hair growth in the two anatomical sites. Hopps [2] has suggested that pubic hair has an anagen and telogen phase of equal duration in contrast to that of scalp hair which has anagen and telogen phases of 900 and 100 days, respectively. If this is the case, the results of DeAntonio et al. [7] suggest that the relative rates of hair growth are not significant factors in controlling hair trace element levels. In severe zinc deficiency states, e.g. acrodermatitus enteropathica [25], and in some cases of severe protein energy nutrition [26], hair growth is impaired. However, hair zinc concentrations are often normal and there is no correlation between hair and serum zinc levels. If impairment of hair growth is n o t an important factor in controlling hair zinc levels, we speculate that this lack of correlation may be due to a relatively minor zinc fraction of the serum regulating hair zinc concentrations. Certainly changes in the very small concentrations of non-essential heavy metals in blood or serum are readily reflected by dramatic changes in hair heavy metal concentrations [27]. To date there is very little information on correlations of human hair trace element levels with components of blood other than serum or plasma. Correlations of hair trace element levels with concentrations in other organs, tissues, or b o d y pools are also poorly known. As emphasized by Hambidge [ 2 8 ] , these data are urgently required in order to better understand the significance of hair trace element concentrations.
ACKNOWLEDGEMENTS This work was supported by the Human Nutrition Research Council of Ontario.
REFERENCES 1 L.M. Klevay, Hair as a biopsy material. 1. Assessment of zinc nutriture, Am. J. Clin. Nutr., 23 (1970) 284--289. 2 H.C. Hopps, The biologic bases for using hair and nails for analyses of trace elements, Sci. Total Environ., 7 (1977) 71--89. 3 J.M. McKenzie, Alteration of the zinc and copper concentration of hair, Am. J. Clin. Nutr., 31 (1978) 470--476.
100 4 S. Salmela, E. Vuori and J.O Kilpio, The effect of washing procedures on trace element content of human hair, Anal. Chim. Acta, 125 (1981) 131--137. 5 G.S. Assarian and D. Oberleas, Effect of washing procedures on trace element content of hair, Clin. Chem., 23 (1977) 1771--1772. 6 K.M. Hambidge, M.L. Franklin and J.A. Jacobs, Hair chromium concentrations: Effect of sample washing and external environment, Am. J. Clin. Nutr., 25 (1972) 384--389. 7 S.M. DeAntonio, S.A. Katz, D.M. Scheiner and J.D. Wood, Anatomically related variations in trace-metal concentrations in hair, Clin. Chem., 28 (1982) 2411--2413. 8 S.C. Vir and A.H.G. Love, Zinc and copper status of the elderly, Am. J. Clin. Nutr., 32 (1979) 1472--1476. 9 J.L. Campbell, S. Faiq, R.S. Gibson, S.B. Russell and C.W. Schulte, Determination of trace element profiles and concentrations in human hair by proton-induced X-ray emission spectrometry, Anal. Chem., 53 (1981) 1249-=1253. ° 10 W.W. Harrison, J.P. Yurachek and C.A. Benson, The determination of trace elements in human hair by atomic absorption spectroscopy, Clin. Chim. Acta, 23 (1969) 83--91. 11 R.S. Gibson and M.S. DeWolfe, Copper, zinc, manganese, vanadium and iodine concentrations in hair in Canadian neonates, Pediatr. Res., 13 (1979) 959--962. 12 S.B. M'Baku and R.M. Parr, Interlaboratory study of trace and other elements in the IAEA powdered human hair reference material, HH-1, J. Radioanal. Chem., 69 (1982) 171--180. 13 P.P. Fuwa, R. McKay and B.L. Valee, Determination of zinc in biological materials by atomic absorption spectroscopy. Anal. Chem., 36 (1964) 2407--2411. 14 M.A. Evenson and B.L. Warren, Determination of serum copper by atomic absorption with the use of the graphite cuvette, Clin. Chem., 21 (1975) 619--625. 15 F.L. Rodky, Direct spectrophotometric determination of albumin in human serum, Clin. Chem., 11 (1965) 478--487. 16 R.F. Coleman, F.H. Cripps, A. Stimson and H.D. Scott, The determination of trace elements in human hair by neutron activation and the application of forensic science, Atomic Weapons Research Establishment, Aldermaston Report No. 0-86/66, 1966. 17 E. Weigand and M. Kirchgessner, Total true efficiency of zinc utilization: Determination and homeostatic dependence upon the zinc supply status in young rats, J. Nutr., 110 (1980) 469--480. 18 J.H. Freeland-Graves, M.L. Ebangit and P.J. Hendrikson, Alteration in zinc absorption and salivary sediment zinc after a lacto-ovo-vegetarian diet, Am. J. Clin. Nutr., 33 (1980) 1757--1766. 19 L.D. McBean, M. Mahloudji, J.G. Reinhold and J.A. Halsted, Correlation of zinc concentrations in human plasma and hair, Am. J. Clin. Nutr., 24 (1971) 506--509. 20 M. Laker, On determining trace element levels in man: the use of blood and hair, Lancet, 8292, vol. 2 ( 1 9 8 2 ) 260--262. 21 N.W. Solomons, On the assessment of zinc and copper nutriture in man, Am. J. Clin. Nutr., 32 (1979) 856--871. 22 E.M. MacMillan and D.J. Rowe, Plasma zinc-serum albumin correlation: Relevance to assessment of zinc status in humans, Clin. Exp. Derm., 7 (1982) 599--604. 23 A.F. Parisi and B.L. Vallee, Isolation of a zinc alpha-2-macroglobulin from human serum, Biochem., 9 (1970) 2421--2426. 24 E.L. Giroux and R.I. Henkin, Competition for zinc among serum albumin and amino acids, Biochim. Biophys. Acta, 273 (1972) 64--72. 25 K.M. Hambidge, P.A. Walravens and K.H. Neldner, The role of zinc in the pathogenesis and treatment of acriodermatitis enteropathica, in: G.J. Brewer, A.S. Prasad (Eds.), Zinc Metabolism: Current Aspects in Health and Disease, Alan R. Liss, New York, 1977, pp. 329--340. 26 J. Ertern, A. Arcasoy, A.O. Caudar and S. Cin, Hair zinc levels in healthy and malnourished children, Am. J. Clin. Nutr., 31 (1978) 1172--1174.
101 27 A. Olguin, P. Jauge, M. Cebrian and A. Albores, Arsenic levels in blood, urine, hair and nails from a chronically exposed human population, Proc. West Pharmacol. Soc., 26 (1983) 175--177. 28 K.M. Hambidge, Hair analysis: worthless for vitamins, limited for minerals, Am. J. Clin. Nutr., 36 (1982) 943--949.