Reference values of elements in human hair: A

e n v i r o n m e n t a l t o x i c o l o g y a n d p h a r m a c o l o g y 3 6 ( 2 0 1 3 ) 1077–1086

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/etap

Reference values of elements in human hair: A systematic review Marcin Mikulewicz a,∗ , Katarzyna Chojnacka b , Thomas Gedrange c , Henryk Górecki b a b c

Department of Dentofacial Orthopeadics and Orthodontics, Medical University of Wrocław, Poland Institute of Inorganic Technology and Mineral Fertilizers, Wrocław University of Technology, Poland Department of Orthodontics, Technische Universität Dresden, Germany

a r t i c l e

i n f o

a b s t r a c t

Article history:

Background: The lack of systematic review on reference values of elements in human hair

Received 22 June 2013

with the consideration of methodological approach. The absence of worldwide accepted

Received in revised form

and implemented universal reference ranges causes that hair mineral analysis has not

13 September 2013

become yet a reliable and useful method of assessment of nutritional status and exposure

Accepted 20 September 2013

of individuals.

Available online 30 September 2013

Objectives: Systematic review of reference values of elements in human hair. Data sources: PubMed, ISI Web of Knowledge, Scopus.

Keywords:

Study eligibility criteria: Humans, hair mineral analysis, elements or minerals, reference val-

Hair mineral analysis

ues, original studies.

Reference ranges

Results: The number of studies screened and assessed for eligibility was 52. Eventually,

Reference values

included in the review were 5 papers. The studies report reference ranges for the content of elements in hair: macroelements, microelements, toxic elements and other elements. Reference ranges were elaborated for different populations in the years 2000–2012. The analytical methodology differed, in particular sample preparation, digestion and analysis (ICP-AES, ICP-MS). Consequently, the levels of hair minerals reported as reference values varied. Conclusions: It is necessary to elaborate the standard procedures and furtherly validate hair mineral analysis and deliver detailed methodology. Only then it would be possible to provide meaningful reference ranges and take advantage of the potential that lies in Hair Mineral Analysis as a medical diagnostic technique. © 2013 Elsevier B.V. All rights reserved.

1.

Introduction

Biomonitoring has recently been approved by obligatory law as technique which assesses the impact of chemical elements present in the environment on living organisms (The EU Water Framework Directive (2000/60/WE), U.S. EPA-600/479-049. August 1979). In the case of human biomonitoring, only non-invasive matrices, e.g. hair, urine, saliva can be used. Hair

∗

provides information on prolonged, not momentary exposure (U.S. EPA). Hair mineral analysis (HMA) has also became an interesting diagnostic tool in biomonitoring of exposure to toxic elements, in the assessment of health and nutritional status (Li et al., 2011). Among all available surveys, different methodological approaches (selection of population, sampling, samples preparation, analytical methods) can be found. Consequently, it appears that the results (reference ranges and values) from

Corresponding author at: ul. Krakowska 26, 50-425 Wrocław, Poland. Tel.: +48 71 784 02 99. E-mail address: [email protected] (M. Mikulewicz). 1382-6689/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.etap.2013.09.012

1078


different studies (or commercial laboratories) are difficult to compare and it is hard to draw meaningful conclusions (Dongarrà et al., 2011). Although, many of commercial laboratories provide the service of HMA and each has established its own reference ranges which are not comparable. Some papers even doubt whether the results of HMA determined by these laboratories are reliable and meaningful (Drasch and Roider, 2002). Reference values (RV’s) and Reference intervals (ranges) (RI’s) were defined by International Federation of Clinical Chemistry (IFCC) (Poulsen et al., 1997). Reference range (standard range) is defined as the prediction interval between which 95% of values of a reference group is found. RVs concern healthy subjects and use reference groups and are also called normal ranges or normal values. Specific ranges are made for each sex, age group, race, etc. (Poulsen et al., 1997). Basing on the available scientific literature it must be understood that the determination of reference ranges is a complicated task and is essential if HMA would be useful and interpretable. In elaboration of reference intervals, the populations should be additionally divided into subgroups according to: gender, age, ethnicity, type of hair (curly or straight), geographical location (Rodushkin et al., 2000; Chojnacka et al., 2006; Gellein et al., 2008; Kempson and Lombi, 2011). Many factors that can affect the reference ranges may influence the final outcome. It seems that a transfer of data obtained in one population may not be meaningful to another. The results should be permanently updated along with modernization of the available analytical instruments (Druyan et al., 1998). Reliable reference intervals which are comparable have not been reported, yet. The rationale for the present systematic review is the lack of the review of reference values, to bring perspective (by comparing the methodology and results) toward the development of useful values that can be used to interpret individual results HMA.

2.

Materials and methods

Information sources: PubMed, ISI Web of Knowledge and Scopus (access 2000-01-01–2013-02-14). Full electronic search strategy for PubMed and other two databases was as follows: ((hair[Title]) AND (“reference values” or “reference interval”[Title])) AND (minerals or elements or metal[Title]). The following eligibility criteria were considered for study characteristics: humans, hair mineral analysis, elements or minerals, reference values, original studies. The report characteristics were as follows: publication since the year 2000, language: English, published papers. Exclusion criteria: language (other than English), hair (other than human scalp), analytes (other than chemical elements), reported results (other than reference ranges), results reported less than 5 elements, less than 120 participants in the population. The included papers were read carefully by two independent readers. The outcomes were compared. The data items were defined: methodology, studied population, references ranges for macroelements, essential and other trace elements and toxic elements. The following assumptions were used for assessing the risk of bias in individual studies: the number of individuals in the studied

population, sample preparation methodology, non-adequate analytical techniques, exposed population (occupationally or environmentally). Only healthy individuals were considered. The results of studies were combined in the tables which included: methodology (sampling, laboratory preparation, digestion, dilution, analysis), population (N, age, sex, country, year, comment), reference values (reference range (lower and upper percentile)). The study was written according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement guidelines for systematic reviews.

3.

Results

PubMed yielded 52, ISI Web of Knowledge – 10 papers, Scopus retrieved 13 papers. Papers from latter two databases were covered by PubMed. Abstracts of all retrieved papers were read and on this basis 5 papers (Dongarrà et al., 2011; Senofonte et al., 2000; Park et al., 2007; Vanaelst et al., 2012; Carneiro et al., 2011a,b) (Table 1) were included in the present systematic review and 47 papers (Table 1) were excluded (Fig. 1 and Appendix References) for the following reasons: language, different matrices, sampling before the year 2000, not healthy participants, ≤5 elements, no reference values in original work, less than 120 individuals in the population. Finally, 5 studies were included in the systematic review. For each study data were combined and reported in Tables 2–6. The methodology of sample preparation differed between the studies considerably (Table 2). The mass of samples varied in the range 0.02–1.3 g. The mass of sample depended on the analytical technique. ICP-MS required ca. 0.1 g, while ICP-AES ca. 0.5 g because of higher detection limits of the latter technique. The laboratory sample preparation approach also varied, as well as the composition of digestion solution. Alternate washing with water and acetone was the most common. Sometimes laboratories used additional washing steps: with ethyl ether, EDTA, TMAH, USN bath. Digestion in microwave oven was practiced in 3 cases. In the remaining studies, incubation for different periods of time was undertaken, sometimes in elevated temperature. Generally HNO3 was used or a mixture of HNO3 and H2 O2 in various proportions. After digestion, the samples were diluted with different volumes of water (0.1–50 ml). Afterwards the samples underwent multielemental analysis (with either ICP-AES or ICP-MS). The studied populations concerned different number of individuals (137–655) and age (children and teenagers, geographical locations (all continents except Africa) – generally urban populations). The research was performed in the years 2000–2012. All the populations were from urban areas and included only healthy volunteers. In some cases the groups was divided according to gender. In the elaboration of reference ranges, different percentiles were considered – the most commonly 5–95 and 10–90. Hair was analyzed for the content of macroelements (Table 3), essential trace elements (Table 4), toxic elements (Table 5) and other trace elements (Table 6). The values reported by different laboratories differed considerably. The results related to reference ranges from different laboratories yielded sometimes contradictory

1079


Table 1 – Studies included and excluded from the review with the reasons. Study

Reason

Included 1 2 3 4 5

Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Vanaelst et al. (2012) Carneiro et al. (2011a,b)

Fulfilled inclusion criteria Fulfilled inclusion criteria Fulfilled inclusion criteria Fulfilled inclusion criteria Fulfilled inclusion criteria

Excluded 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

´ Wozniak et al. (2012) Lee et al. (2012) Carraro et al. (2012) Afridi et al. (2012a) Afridi et al. (2012b) Suliburska (2011) Elenge et al. (2011) Carneiro et al. (2011a,b) Benderli Cihan and Oztürk Yıldırım (2011) Chojnacka et al. (2011) Li et al. (2011) Chojnacka et al. (2010) Liu et al. (2008) Farias et al. (2008) Vaghri et al. (2008) Gray et al. (2008) Błach-Legawiec et al. (2002) Zerbino and Solomenchuk (2007) Dunicz-Sokolowska et al. (2006a) Dunicz-Sokolowska et al. (2006b) Chojnacka et al. (2006) Asano et al. (2005) Goullé et al. (2005) Hamilton and Schweinsberg (2004) Johnsson et al. (2004) McDowell et al. (2004) Pereira et al. (2004) Samanta et al. (2004) Klevay et al. (2004) Hac´ et al. (2003) Tang et al. (2003) Hinwood et al. (2003) Mortada et al. (2002) Black et al. (2002) Lech (2002) Gill et al. (2002) Santos et al. (2002) Asano et al. (2002) Li and Kuo (2002) Soares et al. (2002) Drasch and Roider (2002) Li et al. (2001) Watts and Mercola (2001) Goldberg and Silapunt (2001) Seidel et al. (2001) Ono and Tateshita (2000) Lukasiak et al. (2000)

Not healthy Other than human Language ≤5 elements ≤5 elements ≤5 elements Less than 120 participants Different material (nails) Not healthy ≤5 elements Centenarians Less than 120 participants ≤5 elements Portuguese ≤5 elements Other than human Language Language Sampling since 1994 Sampling since 1994 Less than 120 participants Other than human Less than 120 participants Language ≤5 elements ≤5 elements No reference values ≤5 elements No reference values ≤5 elements ≤5 elements ≤5 elements ≤5 elements ≤5 elements ≤5 elements ≤5 elements ≤5 elements, other than human Other than human ≤5 elements ≤5 elements Review Other than human Letter to the Editor; Authors response Other topic No reference values Other topic ≤5 elements

interpretation concerning the deficiency or excess (e.g. Ca, Mg and many other elements). The differences are related with diversified washing procedures and analytical techniques.

4.

Discussion

Hair is included in reticulo-endothelial system of tissues. Elements are irreversibly incorporated and this is a part of

excretory mechanism for metabolic elimination (Chojnacka et al., 2010). Elements bound in hair have both endogenous and exogenous origin (Senofonte et al., 2000). Exogenously bound elements do not necessarily originate from external deposition. The extra-epithelial portion is exposed to washing, drying, chemical alteration, cosmetics, environmental pollutants present in the environment (water or air) and other chemical and physical factors. Consequently, the elements

1080

Idenficaon


Records idenfied through database searching (n = 51 )

Addional records idenfied through other sources (n = 1 )

Included

Eligibility

Screening

Records aer duplicates removed (n =52 )

Records screened (n =15 )

Records excluded (n = 37 )

Full-text arcles assessed for eligibility (n = 5 )

Full-text arcles excluded, with reasons (n = 10 )

Studies included in systemac review (n = 5 )

Fig. 1 – PRISMA flow diagram.

can be either adsorbed or desorbed by the hair. Therefore washing techniques and sample preparation procedures do not necessarily should aim to remove only elements bound on the surface of hair.

Reference ranges in diagnostics usually report physiologically normal lower and upper limits. However, in HMA these ranges are elaborated statistically and represent the most probable values for healthy individuals (Druyan et al., 1998).

Table 2 – Methodology of human hair samples preparation and analysis. Ref.

Sampling (g)

Laboratory preparation

Senofonte et al. (2000)

0.1–1.3

Dongarrà et al. (2011)

0.15

Park et al. (2007)

2–5

3 + 1 (v/v) ethyl ether + acetone, 85 ◦ C 1 h, 5% EDTA 1 h, water, 85 ◦ C, 16 h Sequence acetone–water–water– water–acetone, 20 ml acetone or water, USN bath – 15 min, 40 ◦ C, 24 h Demineralized water

Vanaelst et al. (2012)

0.1

Carneiro et al. (2011a,b)

0.02

TMAH, tetramethylammonium hydroxide.

Acetone + ultrapure water in USN bath; rinsing with MilliQ Water Acetone/water/acetone

Digestion

Dilution

Analysis

5 ml HNO3 + 1 ml H2 O2 , MW oven

20 ml

ICP-AES

3 ml HNO3 (24 h), 0.5 ml H2 O2 (24 h)

25 ml

ICP-MS

HNO3 (as in Rodushkin et al., 2000) 1 ml (14 M HNO3 + 9.8 M H2 O2 ), MW oven 1 ml 25% m/v TMAH overnight

(as in Rodushkin et al., 2000)

ICP-MS

n.a.

ICP-MS

up to 10 ml, 1% (v/v) HNO3

ICP-MS

1081


Table 3 – Reference values for macroelements levels in human hair (mg/kg). Ref.

Reference range Lower

Upper

Ca Park et al. (2007)

120

365

Vanaelst et al. (2012)

160

1460

Mg Senofonte et al. (2000) Park et al. (2007) Vanaelst et al. (2012) P Senofonte et al. (2000) Vanaelst et al. (2012) Na Vanaelst et al. (2012)

7.4 6 8.67

Percentile Lower

Upper Values within 90% of the confidence interval 90

10

72.9 24 65.3

Population N

Age

Sex

Country

655

3–6

F/M

Korea

218

6–10

F

Belgium

3–15 3–6 6–10

F/M F/M F

Italy Korea Belgium

5

95

10

90

411 655 218

96 108

299 173

5 10

95 90

412 218

3–15 6–10

F/M F

Italy Belgium

2

123

10

90

50

6–10

F

Belgium

Table 4 – Reference values for essential trace elements levels in human hair (mg/kg). Ref.

Reference range

Percentile

Lower

Upper

Cr Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007)

0.08 0.001 0.20

4.56 0.48 0.90

5 2.5

Co Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Carneiro et al. (2011a,b)

0.03 0.01 0.01 0.001

2.95 1.20 0.02 0.017

5 2.5

Cu Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Vanaelst et al. (2012)

7.2 9.1 8 10.1

82.7 59.7 36 46.6

Fe Senofonte et al. (2000) Park et al. (2007) Vanaelst et al. (2012)

5.9 7 3.66

Mn Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Carneiro et al. (2011a,b)

0.04 0.002 0.10 0.03

0.77 0.91 0.60 0.75

Mo Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Carneiro et al. (2011a,b)

0.04 0.0001 0.04 0.024

0.98 1.78 0.10 0.19

Se Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Carneiro et al. (2011a,b)

0.30 0.13 0.50 0.03

1.51 1.28 1.00 0.20

64 96.86 30 150

228 329.19 130 327

Zn Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Vanaelst et al. (2012)

36.8 21 17.3

Lower

10

5 2.5

Upper

N

Age

Sex

Country

95 97.5

160 131 655

3–15 11–13 3–6

F/M F/M F/M

Italy Italy Korea

95 97.5

137 136 655 167

3–15 11–13 3–6 12–18

F/M F/M F/M F/M

Italy Italy Korea Brazil

412 135 655 218

3–15 11–13 3–6 6–10

F/M F/M F/M F

Italy Italy Korea Belgium

408 655 218

3–15 3–6 6–10

F/M F/M F

Italy Korea Belgium

378 133 655 167

3–15 11–13 3–6 12–18

F/M F/M F/M F/M


195 135 655 167

3–15 11–13 3–6 12–18

F/M F/M F/M F/M


327 134 655 167

3–15 11–13 3–6 12–18

F/M F/M F/M F/M


411 130 655 218

3–15 11–13 3–6 6–10

F/M F/M F/M F

Italy Italy Korea Belgium

90

95 97.5

10

90

5

95

10

90

5 2.5 10

5 2.5 10

5 2.5 10

5 2.5 10

Population

95 97.5 90

95 97.5 90

95 97.5 90

95 97.5 90

1082


Table 5 – Reference values for toxic elements levels in human hair (mg/kg). Ref.

Reference range

Percentile

Lower

Upper

Al Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007)

2.4 0.01 3

20.0 12.75 16

5 2.5

Sb Dongarrà et al. (2011) Carneiro et al. (2011a,b)

0.0002 0.0003

0.11 0.02

As Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Carneiro et al. (2011a,b)

0.14 0.0003 0.05 0.0011

0.24 0.03 0.20 0.016

Ba Dongarrà et al. (2011) Park et al. (2007)

0.18 0.10

2.70 0.70

Bi Park et al. (2007)

0

0.10

Cd Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Carneiro et al. (2011a,b)

0.04 0.0004 0.01 0.0003

0.61 0.16 0.20 0.0128

Pb Senofonte et al. (2000) Dongarrà et al. (2011) Park et al. (2007) Carneiro et al. (2011a,b)

1.0 0.28 0.008

19.8 3.03