Hair Analysis Panel Discussion:
Section: Appendix C, Thomas Clarkson

Appendix C
Pre-Meeting Comments

Hair Analysis: Exploring the State of the Science

Thomas Clarkson

Hopps (1977)

He provides background physiology and histology of human hair formation and growth

General questions

He gives no information

Topic 1 Analytical methods

No information Topic 2 Factors influencing the interpretation of analytical results

He notes various pathways of metal into hair:

1. via the follicle into the hair matrix
   
2. secretion of metals in the sebum on to the hair surface
   
3. secretion of metal in exocrine sweat on to the surface of the hair
   
4. secretion of metals in apocrine sweat on to the surface of the hair.
   

He notes that apocrine sweat may not be important for scalp hair.

He discusses the relative merits of head versus pubic hair and concludes that scalp hair is to be preferred

He discusses some reports where lead and arsenic have been measured in scalp hair. The metal level depend on the distance from the scalp/ Lead tends to increase towards the tip of the hair strand.

Arsenic appears to be accumulated in hair and may present a historical record of tissue levels. However hair can accumulate external arsenic in the form of arsenite. Animal experiments indicate arsenic is excreted in sweat.

Variable data have been obtained with cadmium

He gives a table of normal levels of metals in hair.

He notes that attempts to distinguish external versus internal uptake of metals have usually been unsuccessful

Topic 4 Toxicological consideration

No information

Topic 5 Data gaps and research needs

No information

Miekelay et al (1998)

Compared two methods of measuring metal in samples of scalp hair taken from 1,091 adults living in Rio de Janeiro. They also sent a test sample to commercial laboratories for comparison.

General questions

The article indicates the need to revise reference interval for normal levels of metals in hair

Topic 1 Analytical methods

The article claims that ICP-AES (inductively coupled plasma atomic emission spectrometry is out of date with poor detection limits but is still used by most commercial laboratories. The article claims that ICP-MS (inductively coupled mass spectrometry) is the method of choice.
Tables are presented comparing reference limits published by five commercial laboratories indicating wide differences between laboratories for certain metals. Tables are also presented indicating wide differences in results for certain metals on two hair samples circulated blind to the same five commercial laboratories. However, results for some metals yielded reasonable agreement. These metals included Na, Ca, Mg, Mn, Cu, Fe, Zn. The following metals gave reasonable agreement if results from one of the laboratories were excluded: Pb, Cd, Ba, Ni, Li, P, B, Cr, Mo.

Topic 2 Factors influencing the interpretation of analytical results

No information

Topic 3 Toxicological consideration

No information

Topic 4 Data gaps and research needs

The study indicates the need to revise reference limits for some metals.

Sky-Peck (1990)

He performed X-ray fluorescence analysis in six carefully aligned samples of hair from 987 employees and their families at a major medical center in Cook County, Illinois. The purpose was to elucidate factors that might affect concentrations of trace elements in human scalp hair
General questions

He concludes that hair analysis should only be used as a screening method along with other measures of the nutritional status of the patient.

More data are needed on factors affecting trace elements in hair before hair can be used as a quantitative tool to assess the nutritional status of any trace element.

Topic 1 Analytical methods

He used only X-ray fluorescence analysis. He did not describe how the weight of hair was obtained. Usually Compton scattering is used to measure the hair mass. This does not appear to be the method used in this report

Topic 2 Factors influencing the interpretation of analytical results

The method of washing the hair sample can influence the levels of certain trace elements. The mild washing procedure used in the report did not affect levels of 14 selected trace elements. Treatment with peroxide produced a statistically significant reduction in S, Ca, Fe, and Zn. The reduction in Ca was almost complete and Zn was reduced substantially. Hg levels were not affected.

Permanent waving produced a statistically significant increase in levels of 6 trace elements. Levels of Ca, Ni and As were more than doubled. Mercury was unaffected.

Brunettes and blondes differed significantly in only three trace elements, F (slightly lower in blondes); Mn (slightly lower in blondes); and Pb (almost double in blondes). Compared to brunettes and redheads differed statistically in 5 trace elements. Iron was almost doubled in red heads. Mercury was slightly reduced.

Blacks differed from Caucasians in 10 trace elements. Ni, As, and Pb in blacks were more than twice as high as in Caucasians. Orientals differed from Caucasians in 9 trace elements. Ca and Pb in Orientals were a factor of 2 below corresponding levels in Caucasians. Mercury was the same.

Note: Elements differed according to age. Ca in the older group was less than 50% of the younger group. Br was five times higher. Hg was unaffected.

The longitudinal profiles differ according to the trace element. The levels of As, Hg, Cu, Fe, Zn, S and Se were steady and unaffected by distance from the root end. On the other hand, the levels of Pb, Ni and Mn rose sharply towards the tip of the hair strands suggestion external contamination. Ca and Sr showed less pronounced changes.

The results indicate that the levels of certain trace elements are influenced by a number of factors. It would appear that Pb, Ni and Mn are affected by external contamination.

On the other hand, levels of Hg appear to be robust and unaffected by all but one of the factors tested in this report. For one factor, natural hair color of redheads versus brunettes, there was a statistically significant difference in mercury levels, but this difference was quantitatively small.

The most stable trace elements were S, Cu, Zn, Se, Cr, and Rb because these were not changed by more than a factor of 2 by any of the factors tested in this study.

The most unstable elements were Ca, which was affected by more than a factor of 2 by five of the six factors tested. Pb was affected by four factors, and Ni. Br, and Sr by three factors.

Topic 3 Toxicological consideration

No information

Topic 4 Data gaps and research needs

The paper stresses the need for more data on factors affecting levels of trace elements in hair

Seidel et al. (2001)

The authors sent a common hair sample to six commercial laboratories for trace element analysis Different levels were obtained However, it is difficult to evaluate the data without knowing the correct level. These levels can be compared to the normal ranges for each laboratory.

The authors also checked on the accreditation of the labs and on the dietary advice given on the basis on the findings

General questions

The authors argue that there are few if any trace elements that have been validated as indicators of dietary sufficiency or of toxicity. Methyl mercury may be the only substance for which toxic dose response relationships have been established.

Topic 1 Analytical methods

The labs tests used atomic fluorescence or mass spectrometry detection methods. The authors note that the mass spec. method is much lower detection limits

Topic 2 Factors influencing the interpretation of analytical results

The labs can be compared in terms of identifying with elements are outside their normal range. All six labs agreed that the following elements were with their normal range: Ba, Be, B, Cd, S and Ti. All labs agreed that Mn and Mo were outside their normal range. For the following elements all labs except one agreed on classifying according to their normal range: Al, As, Pb, Mg, Hg, Ni, and Zn.

Thus, for approximately half the elements tested, there was reasonable agreement between the commercial labs.

Topic 3 Toxicological consideration

There is lack of toxicological information of the value of hair element concentration as a biomarker for tissue levels, especially levels in the target tissue. This information is available only for methyl mercury.

Topic 4 Data gaps and research needs

As mentioned above, the data gaps are in dose response information and in relating hair levels to levels in the target tissue

Steindel & Howanitz (2001)

The authors provide editorial comment on the paper by Seidel and provide a discussion of proficiency testing in clinical chemistry laboratories.

General questions

They point out that the current lack of normal ranges for trace elements in hair make interpretation of results impossible. They comment of the difficulty of making nutritional conclusions from hair data

Topic 1 Analytical methods

No information

Topic 2 Factors influencing the interpretation of analytical results

The authors listed many problems in interpretation of hair data including external contamination and the absence of reliable reference standards and uniform methods for processing the hair samples.

Topic 3 Toxicological considerations

No information

Topic 4 Data gaps and research needs

More data are needed on inter-laboratory comparisons

Wennig (2000)

This is a review article on the incorporation of drugs into hair. It presents a useful review of hair physiology and biochemistry. It gives recommendations for collection and storage of hair samples.

It gives no information on trace elements in hair.

Yoshinaga et al (1990)

The paper compares the concentration of a number of trace elements in hair with corresponding concentrations in several organs and tissues obtained at autopsy.

Unfortunately, little detail was given on how the hair samples were collected or on the length of the hair samples.

General questions

Topic 1 Analytical methods

A commonly used analytical method was used (ICP-AES). Quality control tests were made.

Topic 2 Factors influencing the interpretation of analytical results

The varying length of the hair samples may have influenced the result and accounted for the poor correlations.

Topic 3 Toxicological consideration

The main finding was that levels of Ca, Mg, P, and Zn in hair did not correlate with tissue levels or body burden

They were not able to draw any conclusions about Fe, Cu or Se as the appropriate tissues were not available for analysis

Topic 4 Data gaps and research needs

More information is needed on hair versus levels in autopsy tissues. The hair length should be restricted to a short segment close to the scalp.

References

• Hopps HC. (1977). The biologic bases for using hair and nail for analyses of trace elements. The Science of the Total Environment, 7:71-89.
  
• Miekeley N, Dias Carneiro MW, Porto da Silveira CL. (1998). How reliable are human hair reference intervals for trace elements? The Science of the Total Environment, 218:9-17.
  
• Sky-Peck HH (1990). Distribution of trace elements in human hair. Clin. Physiol. Biochem. 8:70-80.
  
• Seidel S, Kreutzer R, Smith D, McNeel S, Gilliss D. (2001). Assessment of commercial laboratories performing hair mineral analysis. JAMA 285(1):67-72.
  
• Steindel SJ, Howanitz PJ. (2001). The uncertainty of hair analysis for trace metals. JAMA 285(2):83-85.
  
• Wennig R. (2000). Potential problems with the interpretation of hair analysis results. Forensic Science International 107:5-12.
  
• Yoshinaga J, et al.(1990). The Science of the Total Environment 99:125-135.
  

Next | Table of Contents

Top of Page