Anatomically-Related Variations in Trace-Metal ... - CiteSeerX

8. Ahmed, J., and Smethurst, P., Radioimmunoassay of thyroxinebinding globulin: Evaluation of a kit and diagnostic application. Ann. Clin. Biochem. 17, 241-246 (1980). 9. Kbgedal, B., and KAllberg, M., Determination of thyroxine-binding globulin in human serum by single radial immunodiffusion and radioimmunoassay. Clin. Chem. 23, 1694-1699 (1977).

10. Askew, R. D., Bradwell, A. R., and Ramsden, D. B., The assay of thyroxine binding globulin in human serum using nephelometry. Clin. Chim. Acta 103, 1-5 (1980). 11. Bradwell, A. R., Burnett, D., Ramsden, D. B., etal., Preparation of monospecific antiserum to thyroxine binding globulin for its quantitation by rocket immunoelectrophoresis. Clin. Chim. Acta 71, 501-510 (1976).

CLIN. CHEM. 28/12, 2411-2413 (1982)

Anatomically-Related

Variations in Trace-Metal Concentrations in Hair

S. M. DeAntonlo, S. A. Katz,1 D. M. Schemer, and J. 0. Wood Scalp-hair analysis is used as an indicator of trace-metal concentrations in the human body. The major shortcoming of this approach is the inability to differentiate between endogenous and exogenous trace metals. Our analyses show no correlation between the concentrations of calcium, magnesium, copper, iron, or zinc in scalp hair and pubic hair in 67 donor-matched hair samples. We interpret this as a strong indication that exogenous trace metals

endogenous contributions would be constant at each anatomical site. Gross differences, if observed, could then be attributed to exogenous trace metals.

make a significant contribution to the results for trace metals in human hair.

fees for providing

Subjects and Methods Members

Ca, f, error

Cu, Fe, Zn

#{149}

variation,

sot4rce of

analytical

.

scalp hair has become popular as an index to environmental exposure (1, 2), systemic intoxication (3, 4), nutritional status (5, 6), and to certain diseases (7,8). Hair, unlike other tissues, is a metabolic Determination

of trace metals

in human

end product that is thought to incorporate trace metals into its structure during its growing process. During the growth phase of a hair cycle, the matrix cells at the papilla of the hair follicle show intense metabolic activity and produce hair at a rate of approximately 0.4 mm/day. The developing hair is exposed to the metabolic environment relatively briefly. As the growing hair approaches the skin surface, its outer layers become hardened and thereby seal in the trace metals accumulated from themetabolicenvironment during its formation. The

trace-metal

composition

of the hair reflects

the trace-

metal compositionofthemedium from which itwas formed. In this way, the hair can serve as a history of the trace metals associated with normal and abnormal metabolism as well as a recordof the tracemetals assimilated from the environment. While the determination of trace metals in human scalp hair is enthusiastically accepted in some laboratories (9, 10) for the purposescitedearlier, others(11, 12) raise serious questions regarding the significance of such measurements; specifically, how do such measurements differentiate between endogenous and exogenous trace metals? Our approach to this problem was to measure several trace metals in donor-

matched

scalp- and pubic-hair

samples,

assuming

Department of Chemistry, Rutgers-the State University Jersey, Camden, NJ 08102. 1 To whom correspondence should be addressed. Received June 14, 1982; accepted July 28, 1982.

that the

of New

of the college community samples

of their

were offered modest

scalp and pubic

hair. The

scalp-hair sampleswere cutclosetotheskinatthenape area with

AddItIonalKeyphrases:atomic absorptionspectrometry

of 1-lairSamples

Collection

stainless-steel

scalp-hair

scissors.

The

proximal

2.5 cm of each

sample was stored in a separate

plastic bag. The subjects were provided with scissors and plastic bags and instructedtocollect thehairfrom theentirepubicregion. Inthis way, 67 donor-matched samples of scalp and pubic hair were collected. The subjects also completed a brief questionnaire to provide information on the use of hair grooming products, hair toilet habits, possible sources of occupational exposure to metals, and personal data such as age,race,and sex.The population sampled consisted of 44 men and 23 women. Of the subjects, 21 were over 30 years old, seven between 29 and 25, 33 between 24 and 20,and two under 20.Four subjectsdeclined to state their ages.

of Flair Samples

Preparation

The hair samples in an ultrasonic

were washed by theIAEA procedure

cleaner.

(13)

The samples

were successively

soni-

catedfor60 s in 100 mL of acetone,water,water,water,and acetone.The hair samples were filtered from the wash liquid, allowedto airdry,and then desiccatedoverP2O5 for48 h. Portions

of each desiccated hair sample weighing 200 mg were with 2.5 mL ofredistilled nitric acidinhigh-pressure decomposition vessels (14). Digestion was complete when the decomposition vessels were kept at 100 #{176}C in an oven overnight. The digested samples were quantitatively transferred to 10-mL volumetric flasks and diluted with high-purity

wet-ashed

water.

Measurement Samples Calcium,

hair samples

of Trace-Metal

magnesium,

copper,

were measured

Concentration

in Hair

iron, and zinc in the digested

by conventional

flame atomic

absorption spectrometry. A Model 360 instrument (PerkinElmer Corp., Norwalk, CT 06856) was used. It was operated

intheconcentration mode withscaleexpansion, inaccordwith the manufacturer’s directions. Working ranges for each element were established within the linear region of the instrumental response to increasing concentrations of trace metal CLINICALCHEMISTRY,Vol. 28, No. 12, 1982 2411

Results

Table 1. Precision of Copper and Zinc Determinations in Hair (Five Subjects) Copper, pglg

Zinc, g/g Mean

Mean

SO

27.4

0.159

133

1.49

SD

17.8 10.3 25.9

0.333 0.429 1.75

132 170 232

1.06 3.32 12.2

14.5

0.547

132

4.96

DIscussion

standardsmade up in nitricacidmatrices.Samples giving signals beyond the working ranges were diluted and re-evaluated.

Quality Assurance

Procedures

(3-5 g) samples of scalp hair were obtained from five laboratoryworkers in the courseof normal cuttings. Each sample was cutinto2-5 mm piecesand washed by the IAEA procedure (13). Five replicates of each sample were wet-ashed inthe high-pressure decompositionvessels, and copperand zinc were determined by atomic absorption spectrometry. For a given sample, precision was good: the coefficients of variation ranged from 3 to 5%. The actualvaluesand theirstandard deviationsarelistedinTable 1. Dr. Robert Alverez provided us with a hair reference material prepared by the U.S. National Bureau of Standards (NBS). We determined a mean zinc concentration of 205 sg/g with a standard deviation of 4.7 ig/g in this material by atomic absorption spectroscopy. The NBS valueforZn inthismaterial is 189 tg/g. Large

We participated

in the

IAEA

interlaboratory

study

on

trace-element concentrations in a human hair reference material (HH-1) Our mean results for calcium, magnesium, copper, iron, and zinc were within the 95% confidence limits assignedto thereferencevaluesfortheseelements.Interms of precision, our mean results showed CVs corresponding to 2,4,2,2,and 3% forcalcium,magnesium, copper,iron,and zinc,respectively. Our data and thosefrom the IAEA study are presented in Table 2. (15).

To monitor contamination

the preparation by impurities

of the hair samples for possible in the nitric acid, residues in the

volumetricflasks, or residuesin the Teflon beakersof the decomposition vessels, we introduced a blank as every tenth item in the sample preparation/atomicabsorptionspectrometryprocess.In all,we evaluated14 blanks.Valuesfor copper and zincwere below the limits of detection (which correspondto concentrations in hairof 2 mg/kg) in all14 blanks.

Similarly,

the iron and calcium

below the limits of detection

concentrations

(corresponding

were

to 10 mg/kg) in

all 14 blanks. Two oftheblanksshowed magnesium concentrations corresponding to 51 mg/kg and 39 mg/kg; magnesium

concentrations limits

in the remaining

of detection

(corresponding

12 blanks

were below the

to 2 mg/kg).

Table 2. Results of Trace-Element Determinations In IAEA Human Hair Reference Material IAEA values Mean SD

This work Element

Mean

S_

-

I’gIg of hair

Ca Mg Cu Fe Zn

2412

528 64.7 12.8 22.9 206

10.2 2.51 0.290 0.550 6.70

CLINICAL CHEMISTRY,

522 62.0 10.2 23.7 174

Vol. 28, No. 12, 1982

160 9.58 3.17 9.76 31.5

Our results forcalcium, magnesium, copper,iron,and zinc inthepairedscalp-and pubic-hairsamplesaresummarized inTable 3,includingthecorrelation coefficients obtainedby linearregressionof the trace-metalconcentrationsin the paired samples of scalp and pubic hair.

Our datashow no significant difference between themean respective concentrations of Ca, Mg, Cu, Fe, and Zn in scalp hair and in pubic hair. Of more relevance, however,to the evaluation of exogenous trace metals is the lack of correlations on a subject-by-subject basis between the respective concentrations of these elements in scalp hair and pubic hair. It could be argued that the trace metal content of scalp hair shouldnotbe compared tothatofpubichairbecause (a) hair from different partsofthebody grows at different rates and (b) pubic hair and scalp hair willnotnecessarily have thesame toilet treatments. We firmly believe that our comparisons are valid, and we offer the following comments in support of this position: #{149} Pubic hair does grow more slowlythan does scalphair (16). Hopps (17) attributes this to the anagen (active growth phase)and telogen (resting phase) of pubic hair beingofequal duration while those of scalp hair are 900 days and 100 days, respectively. It appears to us, however, that differences in growth rateshouldnotinvalidate a comparisonoftrace-metal concentrations in scalp and pubic hair. Regardless of growth rate,the endogenous tracemetalsinscalphairand in pubic hair should show some consistentrelationship in donormatched samples. In a hypothetical example assuming equal metabolic activities of hair follicles in the scalp and in the pubicregionbut onlyhalfasmany activefollicles per unit area of skin at the latter site, we would expect the amount of trace elementsincorporated from the metabolicenvironmenttobe thesame perunitofsample weightforscalphairasforpubic hair. If, on the other hand, the number of active follicles per unit area of skin surface were the same at both anatomical sites but those of the scalp showed twice the metabolic activity as that of those in the public region, we would expect scalp hair to consistently show twice the concentration of endogenous trace metals as that shown by pubic hair. #{149} That the incorporation of exogenous trace metals into the hairmay depend upon the cosmetictreatmentofthe hairis exactlyone ofthe pointswe aretryingto make. Ifthe IAEA washing procedure, or any other washing procedure, were able to selectively measurements

remove exogenous trace metals from the hair, of endogenous trace metals would indeed be

possible. In practice, however,thisisnot thecase.Chittleborough (18) reviewed 25 pre-analysis treatments of hair, and Ryabukhin (13) summarized 21 pre-irradiation treatments used in the neutron activation analysis of hair. Although each of these many treatments is intended to remove external contaminants, none, in our opinion, is capable of differentiating between endogenous and exogenous trace metals in hair. Both scalp hair and pubic hair are exposed to a wide variety of cosmetic and pharmaceutical agents, and those used at one anatomical site are not necessarily used at the other.

These agents,inour opinion,areresponsible fordepositing, inoron the hair,tracemetalsthatcouldbe mistakenlyconsideredas reflecting metabolicprocessesinthe body. Creasonetal.(19) foundconcentrations ofcopper,zinc,and lead in scalp hair that significantly exceeded those in pubic hair,but observedno such differences inthe concentrations of iron.These conclusionswere made from data based on approximately 100 unmatched scalp- and pubic-hair samples from pregnant subjects. The scalp-hair samples were obtained during normal hair cuttings or trimmings several months

Table 3. Trace-Metal Concentrations in 67 Scalp Hair and 67 Pubic-HaIr Samples Mean

Ca

Cu

537 885 117 121 38 45 41 37 156 155

Pubic Scalp Pubic Scalp Pubic

Scalp Fe

Pubic

Scalp Zn

Pubic Scalp

preparturition; the pubic-hair samples were collected at the time of delivery. The scalp-hair trimmings probably consisted almost exclusively of distal ends. Concentrations of some trace metals in human scalp hair at the distal ends exceed those at the proximal ends (20,21). Hence, the observations of Creason et al. (19) have little relation to our work. Orlando and co-workers 23) reported that the concentrations of copper, zinc, lead, and iron in scalp hair significantly exceeded those in pubic hair. The samples were collected from 39 pregnant subjects at the time of delivery. The specific locations at which the scalp-hair samples were taken were not specified. They concluded that pubic hair was more appropriate than scalp hair for toxicological studies because the former was less likely to be contaminated. Linear regression analyses of our data show little if any correlation between the concentrations of calcium, magnesium, copper, iron, or zinc in scalp hair and pubic hair. We interpret this as a strong indication that each anatomical site of our subjects made its own unique exogenous contribution to the trace-metal content of hair from that site. If this is true in the general case, we urge extreme caution in interpreting the results of trace-metal determinations in human scalp-hair samples. The diagnosis of disease, the assessment of nutritional status, and the evaluation of systemic intoxication, if at all possible from hair analysis, must be made on the basis of only endogenous trace metals. (22,

We are grateful to R. Alverez, National Bureau of Standards, supplying the reference hair. Partial Charles and Joanna Busch Memorial gratefully acknowledged.

for financial support from the Fund (grant no. 28-9117) is

References 1. Obrusnick, I., and Bencko, V., INAA study of trace elements in hair of three selected population groups in Czechoslovakia. Radioanal. Lett. 38, 189-196 (1979). 2. Jervis, R. E., Tiefenbach, B., and Chattopadhyay, A., Scalp hair as a monitor of pollution exposure to environmental pollutants. J. Radioanal. Chem. 37, 751-760 (1977). 3. Valentine, J. L., Kang, H. K., and Spivey, G., Arsenic levels in human blood, urine and hair in response to exposure via drinking water. Environ. Res. 20, 24-32 (1979). 4. Hagedorn-Gotz, H., Kupper, G., and Stoeppler, M., On nickel contents in urine and hair in a case of exposure to nickel carbonyl. Arch. Toxicol. 38, 275-285 (1977). 5. Gibson, R. S., and DeWolfe, M. S., Copper, zinc, manganese, vanadium and iodine concentrations in the hair of Canadian low birth weight neonates. Am. J. Clin. Nutr. 32, 1728-1733 (1979).

Concn, mg/kg SD

508 960 68 105 17 22 20 13 50 50

Range

85-2570 38-3840 13-478 10-1010 16-81 17-98 20-89 22-77 64-285 90-294

I.

0.394 0.280 0.273

0.214 0.285

6. Deeming, S. B., and Weber, C. W. Evaluation of hair analysis for determination of zinc status using rats. Ibid. 30, 2047-2052 (1977). 7. Terai, M., Akabane, A., Ohno, K., et al., An application of neutron activation analysis to biological materials. II. The comparison of trace element contents in normal and diseased infant hairs. J. Radioanal. Chem. 52, 143-152 (1979). 8. Ryan, D. E., Holzbecher, J., and Stuart, D. C., Trace elements in scalp-hair of persons with multiple sclerosis and normal individuals. Clin. Chem. 24, 196-200 (1978). 9. Bland, J., Through the looking glass darkly: A story of trace mineral-induced behavioral disturbance and hair mineral analysis. Orthornol. Psychiatry 9,24-32 (1980). 10. Klevay, L. M., Hair as biopsy material: Progress and prospects. Arch. Intern. Med. 138,1127-1128 (1978). 11. Katz, S. A.,The useof hair asa biopsy material for trace elements in the body. Am. Lab. 11,44-52 (1979). 12. Hilderbrand, D. C., andWhite,D.H., Trace element analysis in hair: An evaluation. Clin. Chem. 20, 148-151(1974). 13. Ryabukhin, Y. S., Activation analysis of hair as an indicator of contamination of man by environmental trace element pollutants. Report IAEA/RL/50, International Atomic Energy Agency, Vienna, 1978. 14. lida, C., Uchida, T., and Kojima, I., Decomposition of bovine liver in a sealed Teflon vessel for determination of metals by atomic absorption spectrometry. Anal. Chim. Acta 113,365-368(1980). 15. M’Baku, S. B., and Parr, R. M., Interlaboratory study of trace and other elements in the IAEA powdered human hair reference material, HH-1. 6th mt. Conf. Modern Trends in Activation Anal., Toronto, June 1981. 16. Flesh, P., in Physiology and Biochemistry of the Skin, S. Rothman, Ed., Univ. Chicago Press, Chicago, IL, 1954, pp 601-661. 17. Hopps, H. C., The biological basis for using hair and nail for analysis of trace elements. Sci. Total Environ. 7,71-89 (1977). 18. Chittleborough, G., A chemist’s view of the analysis of human hair for trace elements. Ibid. 14, 53-75 (1980). 19. Creason, J. P., Svendsgaard, D., Bumgarner, J., et al., Maternalfetal tissue levels of 16 trace element substances in 8 selected continental United States communities. Proc. Conf., Trace Element Substances in Environmental Health, Univ. Missouri, Columbia, 1976. 20. Hambridge, K. M., Franklin, M. L., and Jacobs, M. A., Change in hair chromium concentrations with increasing distance from hair roots. Am. J. Clin. Nutr. 25,380-383 (1972). 21. Valkovic, V., Rendic, D., and Phillips, G. C., Elemental ratios along human hair as indicators of exposure to environmental pollutants. Environ. Sci. Techol. 9, 1150-1152 (1975). 22. Orlando, P., Perdelli, F., Casadio, M., and Pecorari, D., Sulla presenza di alcuni microelementi in gestanti a Genova. Estratto Giornale Igiene Me. Prey. 19, 62-67 (1978). 23. Orlando, P., Perdelli, F., Franco, J., and Alacevich, C., Sulla presenza di alcuni microelementi in capelli e peli pubici di alcune gestanti a Genova. Ibid. 19, 76-82 (1978).

CLINICALCHEMISTRY,Vol. 28, No. 12, 1982 2413