v>EPA
United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
EPA-600/1-78-037 a
May 1978
Research and Development
Human Scalp Hair:
An Environmental
Exposure Index for
Trace Elements
I. Fifteen Trace
Elements in
New York, N.Y.
(1971-72)
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development. U.S Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The nine series are1
1 Environmental Health Effects Research
2. Environmental Protection Technology
3 Ecological Research
4 Environmental Monitoring
5 Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8 "Special" Reports
9 Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series This series describes projects and studies relating to the toler-
ances of man for unhealthful substances or conditions. This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques utilizing ani-
mals but always with intended application to human health measures
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/l-78-037a
May 1978
HUMAN SCALP HAIR: AN ENVIRONMENTAL
EXPOSURE INDEX FOR TRACE ELEMENTS
I. Fifteen Trace Elements in New York, N. Y. (1971-72)
by
John P. Creason
Statistics and Data Management Office
Health Effects Research Laboratory
and
Thomas A. Hinners
Joseph E. Bumgarner
Cecil Pinkerton
Environmental Monitoring and Support Laboratory
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
HEALTH EFFECTS RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, N.C. 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research Laboratory,
U.S. Environmental Protection Agency, and approved for publication. Mention
of trade names or commercial products does not constitute endorsement or
recommendation for use.
ii
-------
FOREWORD
The many benefits of our modern, developing, industrial society
are accompanied by certain hazards. Careful assessment of the relative
risk of existing and new man-made environmental hazards is necessary
for the establishment of sound regulatory policy. These regulations
serve to enhance the quality of our environment in order to promote the
public health and welfare and the productive capacity of our Nation's
population.
The Health Effects Research Laboratory, Research Triangle Park,
conducts a coordinated environmental health research program in toxicology,
epidemiology, and clinical studies using human volunteer subjects.
These studies address problems in air pollution, non-ionizing
radiation, environmental carcinogenesis and the toxicology of pesticides
as well as other chemical pollutants. The Laboratory participates in
the development and revision of air quality criteria documents on
pollutants for which national ambient air quality standards exist or
are proposed, provides the data for registration of new pesticides or
proposed suspension of those already in use, conducts research on
hazardous and toxic materials, and is primarily responsible for providing
the health basis for non-ionizing radiation standards. Direct support
to the regulatory function of the Agency is provided in the form of
expert testimony and preparation of affidavits as well as expert advice
to the Administrator to assure the adequacy of health care and surveillance
of persons having suffered imminent and substantial endangerment of
their health.
These data are provided for those researchers interested in developing
a reliable and easily collected index of environmental exposure to certain
trace elements, and as well, the data shed light on the influences of
personal covariates on the trace element content of hair. These data are
timely with regard to the current concerns regarding low-level environmental
exposure to trace elements and their uptake by exposed populations.
F. G. Hueter, Ph. D.
Acting Director,
Health Effects Research Laboratory
iii
-------
ABSTRACT
Previous studies have revealed that hair trace element concentrations
can reflect exposure in cases of frank poisoning and deficiency. Corre-
lations have been found also in some populations living in regions where
metallurgic processes are conducted.
This study reports significant correlations between hair barium,
chromium, lead, mercury, nickel, tin, and vanadium content and exposures
(as measured by analyses for the corresponding elements in dustfall or
housedust) within a single metropolitan area. Age, sex, hair color, and
smoking habits were included in the statistical evaluation. Several
metals showed a tendency to increase and decrease together in the hair
specimens in agreement with trends reported for other human tissues.
It is acknowledged that hair has the capacity to adsorb and to
release trace elements in certain situations. However, population studies
can compensate for confounding influences by (1) a randomizing effect,
by (2) an averaging effect, and (3) by statistical rejection of unrepre-
sentative data values. The relationship of hair content to (a) content
in other tissues and to (b) metabolic status are separate and complex
issues that should not be confused with (c) exposure relationships.
IV
-------
CONTENTS
Foreword 1!1
Abstract
Figures Y1.
Tables V.V.
Acknowledgment v111
1. Introduction J
2. Methods J
Environmental Monitoring }
Scalp Hair and Covariate Information *
Chemical Analysis |
Statistical Analysis °
3. Results °
Study Population Characteristics °
Sample Hair Trace Metal Characteristics 8
Pollution Media Trace Metal Characteristics 9
Hair Trace Metal Concentrations in Relation
to Media Indices of Trace Metal Exposure
and to Personal Covariates '1
4. Summary and Discussion 15
References
-------
FIGURES
Number Page
1 Trace metal levels by element and community for
dustfall 20
2 Trace metal levels by element and community for
housedust 21
3 Trace metal levels by element and community for soil 23
4 Geometric mean scalp hair Pb for children by age and
sex 24
5 Geometric mean scalp hair Mn for children by age and
sex 25
6 Geometric mean scalp hair Ba for children by age and
sex 26
7 Geometric mean scalp hair Ba for adults by age and
sex 27
8 Geometric mean scalp hair Cu for adults by age and
sex 28
VI
-------
TABLES
Number Page
1 Sample Preparation and Analytical Methods 29
2 Number of Participants by Age, Sex, and Area of
Residence 30
3 Demographic Characteristics of Adults by Area of
Residence 31
4 Hair Color by Age Category and Sex 32
5 Trace Metal Levels in Human Hair in Children
(Ages 0-15) and Adults (Ages +16) (ug/g) 33
6 A Significance Table for Hair Element/Element
Correlations 34
7 Trace Metals in Dustfall by Area (mg/m2/mo) 35
8 Trace Metals in Housedust by Area (yg/g) 36
9 Trace Metals in Soil by Area (yg/g) 38
10 Correlation Coefficients of Logs of Media Trace
Metal Levels 39
11 Tests of Significance of the Effect of Selected
Factors on Soil Trace Metal Levels 40
12 Tests of Significance of the Effect of Selected
Factors on Scalp Hair Trace Metal Levels, Using
(1) Dustfall and (2) Housedust as a Measure of
Environmental Exposure 41
13 Dustfall, Housedust, and Scalp Hair Mean Trace Metal
Levels by Community for Scalp Hair Metals with a
Significant Environmental Exposure Effect 42
14 Geometric Mean Scalp Hair Trace Metal Concentrations
by Sex for Children and Adults (yg/g) 43
VII
-------
ACKNOWLEDGMENT
The authors gratefully acknowledge Ms. Peggy Stewart, Dr. Anna Yokum
and the personnel of Stewart Laboratories, Inc., Knoxville, Tennessee,
for their major contributions and dedication to this study.
vm
-------
SECTION 1
INTRODUCTION
Human scalp hair has been shown to reflect increased environmental
exposure to metals such as lead, mercury, cadmium and arsenic (1-6).
Hair chromium and zinc concentrations have been reported to reflect de-
ficiency conditions in humans (11,22,38,40-46). Similar trends for
these and other elements have been found in animal studies (11,38,43,44
47). Scalp hair is an almost ideal tissue for population sampling in
that it is painlessly removed, normally discarded, easily collected and
easily stored (1,2,38). The CHESS* program of the United States
Environmental Protection Agency (7) provided a means for community-wide
sampling of scalp hair.
The purpose of this study was to evaluate further the utility of scalp
hair as a method of environmental monitoring of humans for trace metals
exposure. The relationship of scalp hair trace metal levels to important
personal covariates such as hair color, age, sex, and socioeconomic level
was also of concern. Since for some metals, smoking has been implicated
as an important exposure covariate (8,9), it was included in this study.
A knowledge of the effects of these personal covariates on scalp hair trace
metal levels will permit a more accurate assessment of the quantitative
relationships of hair trace metal levels to environmental exposure.
Dustfall has been used as an environmental index of trace substance
exposure (19,48). More intimate indices of trace substance exposure such
as household dust, soil and water from CHESS-participant homes have already
been considered (9,10).
*CHESS stands for Community Health and Environmental Surveillance System.
1
-------
The specific hypotheses tested in this study were:
1. Significant relationships exist between dustfall, household
dust (housedust), and soil trace metal levels.
2. There are significant variations in selected scalp hair
trace metal levels due to personal covariates such as age,
sex, hair color, socioeconomic status and smoking habits.
3. Environmental exposure, as measured by one of the above media,
has a significant effect upon selected scalp hair trace metal
levels, even after adjusting for any effects due to personal
covariates.
One covariate of great interest that could not be evaluated in this
study was race. There were too few non-white respondents to make such
an investigation possible. This covariate will be examined in later
studies.
Other than the above specific hypotheses, it was of interest to
establish the distributional characteristics of each scalp hair trace
metal, including baseline levels, ranges, and skewness of the distributions.
Although measurements of nineteen (19) different scalp hair trace
elements were attempted, only the following fifteen (15) will be dealt with
in this report: barium (Ba), boron (B), cadmium (Cd), chromium (Cr),
copper (Cu), iron (Fe), lead (Pb), lithium (Li), manganese (Mn),
mercury (Hg), nickel (Ni), selenium (Se), silver (Ag), tin (Sn) and
vanadium (V). Arsenic (As), beryllium (Be) and cobolt (Co) were excluded
from consideration because many hair sample values were below detection
limits for the sample sizes available. Zinc (Zn) was measured in all
three environmental samples collected as well as in scalp hair, but the scalp
hair concentrations were found to be low in comparison to most normal values
-------
in the literature (11,22). A report on this metal has therefore been
postponed. Five metals (Cd, Cu, Pb, Mn and Ni) were measured in all the
media (dustfall, housedust, and soil) and the remaining nine metals were
measured in housedust only.
-------
SECTION 2
METHODS
Environmental Monitoring
Three CHESS communities were selected in the New York City metropolitan
complex: the Westchester section of Bronx, the Howard Beach section of
Queens, and the town of Riverhead, Long Island. These communities were
believed to exhibit a general pollution exposure gradient from low
(Riverhead) to intermediate (Queens) to high (Bronx); this belief was
based on previous observations of total suspended particulate levels and
dustfall levels, as well as some preliminary examinations of levels of
selected trace metals in dustfall (10,12). Atmospheric studies including
dustfall collections, were made at CHESS air monitoring sites. These
sites were located such that the families from each of the three communities
of interest were within 2.5 km. of the respective sites.
Dustfall data were obtained monthly over a period of 22 months, from
September, 1970 through June, 1972, at the central site within each
community. Standard procedures of collection were followed (13). Four
soil samples were obtained from each of 43 residences (14 each from
Queens and Riverhead and 15 from Bronx). Two samples were collected
from the front yard, and two from the back yard. All samples were taken
to a depth of two inches with a stainless steel auger. Ninty-nine
housedust samples (29 from Queens, 37 from Riverhead, and 33 from Bronx)
were obtained by collecting the contents of vacuum cleaner dust bags from
community homes.
Scalp Hair and Covariate Information
In March, 1971, letters giving information about the proposed trace
4
-------
metals study were sent to all active CHESS acute respiratory disease (ARD)
families in the three communities. These letters included for each family
member a return addressed postpaid envelope, with the individual member
identification label affixed to the back, and instructions for the
collection of the hair sample.
In the instructions it was stressed that the hair should be taken
from the next normal haircut or trim, and that as much hair as possible
should be collected. It was also stressed that there was no special need
to wash or shampoo the hair before the haircut or trim, because it would be
washed in the laboratory before analysis. Polyethylene bags for hair
samples were sent with the letters and had an identification label on the
back with each person's first name, family name and a space for the date
of the haircut. During a scheduled ARD survey phone call, after allowing
sufficient time for receipt of the letters, the color of each member's
hair and the location of the haircut (barbershop or home) were ascertained.
This information was then combined with the CHESS ARD background information
questionnaire obtained at the start of the ARD study to make a complete
covariate information file on each contributing family member. Collection
of hair samples was terminated in July 1971. The hair samples were then
stored from July 1971 until trace metal analyses were carried out in the
Spring of 1973.
Chemical Analysis
Dustfall samples were acid extracted and the metals determined by
atomic absorption spectrophotometry (13). Housedust samples were sieved
through a 0.5 mm screen for 5 minutes at 260 oscillations per minute on
a mechanical shaker, and extracted with 6N nitric acid at 50°C for 30
minutes. Soil samples were also sieved through a 0.5 mm screen. Twenty
5
-------
grams of soil was then extracted with 40 ml of IN HC1. Hair specimens
were washed with a detergent solution, rinsed and dried according to the
procedure of Harrison et al. (14). Digestion was achieved by oxygen
combustion for some elements and by dry ashing for others, as indicated in
Table 1. To prevent any losses when volatile elements were to be
analyzed, weighed portions of the washed and dried hair were prepared
by the Schoniger Flask technique.
Analytical methods for each metal are also shown in Table 1. Standard
laboratory quality control procedures were employed. In addition, recovery
of all 19 elements added to a housedust sample and to a hair sample were
evaluated using additions that were either twice the detection limit or
twice the endogenous level, whichever was larger. Recovery rates were
greater than 85 percent in all cases, and greater than 95 percent in most
of the cases.
Statistical Analysis
The first step taken in preparation for statistical analysis of the
hair data was the careful editing of the data for outliers, values so
far removed from the main body of readings as to warrant their removal
from the population for statistical analysis purposes.
A statistical procedure was developed for this process, so that all
subjective tendencies toward removal of any values were eliminated. In
this procedure the inherent population variability, as measured by the
standard deviation of the logs of the values, was estimated from the
central section of the sample. Limits were then obtained by taking +3
standard deviations from the mean of the logs of the sample. Histograms
of the data were carefully examined to insure the effectiveness of this
-------
procedure, and to verify that a large number of seemingly valid obser-
vations were not being eliminated. No such problems were ever encountered.
For the trace metals measured, the percent declared outliers varied from
0.2 percent for lithium up to 7.6 percent for selenium, with the median
percent rejected being 1.5 percent.
Examination of the trace element values revealed a consistent
tendency toward log-normality of the distribution (i.e. populations skewed
to the right), so that logs of all values were employed in the subsequent
statistical analyses. Standard statistical techniques of correlation and
multiple linear regression were then used to discover the effects and
interrelationships of all of the measured variables.
-------
SECTION 3
RESULTS
Study Population Characteristics
Over 3000 subjects participated in the CHESS-ARD study (13); 426
families, comprised of about 1900 total members, responded to the hair
study letter in some fashion. Family members who gave no hair, who
gave an insufficient quantity of hair, or who had incomplete covariate
information, were excluded from the study. Because responses were
available from very few nonwhite families, these persons were also
excluded. The resultant population consisted of 498 subjects with
complete information. The distribution of subjects by age, sex and
residence is shown in Table 2. The study population is seen to consist
of two distinct age groups. This is a result of the method of contact
in the ARD study, wherein only families having children in elementary
schools were selected. This population division presents no difficulty,
however, since subsequent analyses are made on children and adult
populations separately. The distribution of the final group of participants
with respect to smoking patterns and education for head of households
(Table 3) was similar to the original ARD population (13).
Reported hair color by sex for children (15 years of age and less)
and adults (over 15 years of age) showed significant differences in hair
color distribution between male adults and male children, but not between
female adults and female children (Table 4). The adult males showed a
higher percentage of black and grey hair than male children.
Sample Hair Trace Metal Characteristics
The 15 trace metals in this study all have typical log-normal type
8
-------
of distribution usually displayed by trace metals in hair (15,16).
Analytical hair values obtained generally agree with published values
(3,11,15,16).
The interrelationships of scalp hair trace metal levels were
examined separately for children (15 years of age or less) and adults
(over 15 years of age), since children are in a rapidly developing
stage of growth in contrast to adults who have reached a leveling
off in growth and development. Some adults also are likely to have
been exposed to some work-related sources of trace metals, and hence
may display a good degree more variability from person to person in
scalp hair trace metal burden than children.
It was found that adults have higher mean scalp hair levels than
children in 10 out of the 15 metals (Table 5).
Pairwise correlation coefficients were computed separately for the
adults and children groupings using the logs of scalp hair trace metals.
There were 68 significant correlations for adults and 85 for the children
out of the 120 correlation coefficients (Table 6). Selenium (Se) was
the only element to consistently show significant negative correlations.
In both adults and children Cd, Cu and Pb were highly inter-correlated.
The relation of hair trace metal levels to media indices of trace
metal exposure and to personal covariates are considered in a later
section of this report.
Pollution Media Trace Metal Characteristics
Arithmetic means of dustfall and geometric means of soil and
housedust trace metal concentrations by community are presented in
Tables 7, 8 and 9.
-------
A previous study of pollution data from these same sites within
these same communities established the fact that there were indeed
statistically significant differences in trace metal levels between
communities in dustfall, soil, and housedust for Cr, Cu, Mn, Ni, Pb
and Cd, except that chromium was not analytically determined in the
soil, and Cd was found to display no difference in the housedust levels
across communities (10). Bar graphs of the means for dustfall and
geometric means for housedust and soil by community are shown in Figures
1-3. These figures demonstrate the concentration gradients as well as
the similarities and differences in trace metal patterns across these
three media. Cd, Cr and Ni generally have much lower concentrations than
Cu, Pb, Mn and Zn. There is apparently a higher percent of Mn in the
soil than in housedust or dustfall, while dustfall seems to have relatively
less Ni and Cu than the other two environmental media. For most of the trace
metals the three media indices of exposure are quite similar, reflecting
increased exposure in moving from Riverhead to Queens to Bronx. In order to
more precisely assess the interrelationships of soil, housedust and dustfall,
correlations of the respective trace metals across these media were
computed (Table 10). Logs of the concentrations were used in this
computation in order to normalize the data and make significance tests
valid. One should keep in mind that all correlations with dustfall
metals involve the pairing of housedust and soil metals within a community
to a single dustfall value. However, this correlation produces the same
results as one would obtain by a simple linear regression of the media
metal on the dustfall metal. The housedust-soil correlation was
obtained by first averaging the four soil values to obtain a single
value per home, and using all houses for which both housedust and solid
10
-------
results were available In order to obtain the correlation coefficients.
All of the correlation coefficients were found to be highly significant
indicating a strong relationship between these measures of environmental
trace metal levels for the six metals available for analysis.
The use of soil as an environmental index of trace metal burdens
has several problems that have been mentioned in earlier reports (19,20,
21). The problems that must be considered are background levels in the
soil, possible gradients in soil concentrations due to roadway traffic,
and contamination from older homes on which a great deal of lead-based
paint had been used. From an analysis of the logs of soil trace metal
levels (Table 11), the strong relationship of dustfall levels to soil
levels is seen, as well as the great variability between homes within the
areas, and between the front and back yard measurements.
Hair Trace Metal Concentrations in Relation to Media Indices of Trace
Metal Exposure and to Personal Covariates
One must take into account as many influences on scalp hair trace
metal levels as possible in order to assess adequately its utility
as an epidemiological personal exposure index. The first step in this
process was the separate analysis of children (5!5 years old) and adults
(>J6 years old). Within each of these groups, the following possible
influences were assessed: age, sex, hair color, location of haircut
(home or barbershop), socioeconomic level (as measured by education of head
of household) and smoking patterns (in adults only). The relation
of environmental exposure to hair trace metal levels, after adjusting
for the effects due to these covariates, was then tested using a linear
multiple regression model. The logs of both the scalp hair trace metal
11
-------
levels and the environmental exposure trace metal levels were used in
the analysis to make the scalp hair values more closely fit a normal
distribution and to help insure the fit of a linear model to the data.
The results of the statistical tests when monthly average trace
metal dustfall rates were used are given in Table 12. Dustfall trace
metals had a significant effect on hair levels for Pb, Cr, and Ni in
children but only on hair levels of Pb in adults. All of these scalp
hair metals show a marked increase in concentration with the media
gradient (Table 13).
Housedust was used in place of dustfall as an environmental index
in the above models by computing the geometric mean of the housedust
trace metal readings obtained from contributing households within each
community. The results of this substitution were essentially identical
to those with dustfall in the model, except that housedust chromium was
not indicated as having a significant effect on scalp hair chromium in
children (Table 12).
By selecting the family members of households contributing housedust
and computing the correlation of these scalp hair trace metals and the
corresponding housedust trace metals, a more direct comparison of
environmental exposure and scalp hair levels of trace metals is possible,
although other covariates must of necessity be ignored. Significant
correlations found were for Pb (r=0.27) and Ni (r=0.32) in children
and for Cu (r=0.31) in adults. It is apparent that the housedust and
dustfall trace metal values compared above can be used interchangeably
as an exposure index.
12
-------
There are nine metals in this study for which the only pollution
index available is housedust: Ba, B, Fe, L, Hg, Se, Ag, Sn and V.
The linear model analysis for these metals using housedust as an environ-
mental index is shown in Table 12.
Selenium was found to show no differences in housedust trace metals
across communities, so that the effect of selenium in housedust on scalp
hair selenium levels could not be tested here. Ba, Pb, Hg and V were
found to have a significant relationship between housedust and scalp hair
levels for both adults and children, while Sn was significant for children
only.
Significant trends of scalp hair Pb, Mn and Ba with age were found
in children, while Cu and Ba in scalp hair were related to age in adults.
In children, scalp hair Pb decreased with age while Mn and Ba showed an
increase with age. In adults, both Ba and Cu decreased with age. A
separate linear model including an age-sex interaction term was also
examined in the belief that the concentration change with age might be
different between sexes. This proved to be true for adult scalp hair Cu
and Ba as well as for Ba, Pb and Mn in hair of children (Figures 4-8).
When examined separately by sex, the female adults were found to
have rapidly decreasing Cu scalp hair levels, while male levels were
fairly steady over age. These trends in Cu levels are in close agreement
with the results of Schroeder et al. (15). The increased hair Ba and Mn
values in females (but not males) at ages 12-13 is intriguing in reference
to puberty. Mn does have a role in reproduction (39), but Ba has
usually been considered nonessential (11). Ba is chemically related to
13
-------
calcium (Ca), and the higher concentration of Ba in female hair than in
male hair may be associated with a similar differential observed for Ca
(16,17,36,38).
For adults, significant differences in scalp hair trace metals between
sexes were noted for all metals except Cr, B, Fe, Se, Ag and V. For
children only Cr, B, Fe and Se were not significantly different by sex.
Females levels were always higher than male concentrations wherever
significant differences were found in children. In adults, females were
higher than males except for cadmium and lead.
Fe levels in scalp hair were found to be significantly related to
socioeconomic level as measured by education for both adults and children
with the scalp hair levels following a negative gradient with increasing
education level. That is, subjects with higher levels of education of
the head of household had lower levels of Fe in their scalp hair. Fe
in adult scalp hair was the only metal showing a significant relationship
to hair color, with brown and black hair colors having less Fe, while
gray hair had more.
Hair samples collected at barbershops as opposed to home collection
were significantly higher for adults in Cd, Pb, Ba, B and V while in
children Ni, Ba and V were higher. Selenium in children's hair collected
in barbershops were significantly lower than found in hair from children
who had home haircuts.
Smokers were found to have significantly higher hair Se and Fe.
Hair Pb was higher for smokers than for nonsmokers, but was not quite
statistically significant (p=0.06).
14
-------
SECTION 4
SUMMARY AND DISCUSSION
While other studies have Illustrated metal changes In hair when
exposure differences have been dramatic, this report indicates that hair
metal content can reflect exposure trends within a single metropolitan
area. Several personal covariates were found to Influence scalp trace
metal levels strongly and must be taken into consideration if scalp hair is
to be used as an environmental index.
In this study, the environmental exposure gradients of Pb, Ba, Hg and V
(as measured by dustfall or housedust) were significantly associated with
scalp hair trace metal levels in both adults and children, while the
environmental exposure gradients of Cr, Ni and Sn were reflected in
children's hair only. The seven other metals investigated (Cd, Cu, Mn, B,
Fe, Li and Ag) displayed no such significant associations. Housedust and
dustfall trace metal values proved to be usable interchangeably as an
exposure index for the metals that were measured in both media.
Sex was found to be the most important covariate in the study, being
significantly associated with 11 of the 15 trace metals examined in children
and 9 of the 15 trace metals in adults. Female scalp hair values were
higher than male values in all cases where differences were significant
except for Cd and Pb in adults. The reversal of the general sex trend for
Cd and Pb could be a reflection of work-related exposure of males to these
nonessential metals, although this hypothesis was not examined in this
study. Adult males reportedly have more kidney Cd than adult females
(69,70) but Pb did not differ by sex in 33 tissues (70). However, blood
Pb is reported (9) to be lower in women than in men in similar environments.
15
-------
Other investigators have also found sex differences as observed in this
study, with females most often higher than males (3,11,15-18,25,36). The
tendency for female hair to be higher than male hair in several metals
may be related to a higher inorganic content for female hair on the
average (49). Additional studies may also investigate whether the sex
difference might be explained by other factors such as hair length or
shampoo frequency.
Scalp hair Pb values decrease rapidly with age in children. This
decrease is probably a result of a proclivity to pica as well as a higher
respiratory rate in the very young. Mn and Ba begin increasing in
children's scalp hair around age 8; Ba displays peak values in women at age
11-12 and 31-35 while Mn does not change with age in adults. The causes
of these age trends in Ba and Mn are not clearly understood. Cu was found
to rapidly decrease with age in adult females. Other reports have shown
a decreasing trend in hair copper with age (15,25,50).
Hair color and education of head of household both reflected a
relationship only to scalp hair Fe. The hair color differences in Fe were
significant only for adults. Over 90 percent of the hair in this study
was either blond or brown, so other hair colors may not have been adequately
sampled for the determination of any differences in trace metal content.
Other investigators have reported differences with hair color (11,15,39).
The increased concentrations of Cd, Pb, Ni, Ba, B and V found for hair
samples collected in a barbershop or beauty shop as compared to collection in
the home may represent contamination. The concentration of Se, however, was
found to be lower in hair collected in barbershops or beauty shops compared
to the amounts found in children's hair from home haircuts.
16
-------
Smokers were found to have higher Pb, Fe and Se scalp hair levels than
nonsmokers in this study. Although these metals have been reported to be
in cigarettes, the authors are unaware of any published studies in which
the relationship between hair levels and smoking have been examined,
although blood Pb values have been found to be higher in smokers than in
nonsmokers (9).
The hair element intercorrelations (Table 6) show agreement with
several element intercorrelations reported for other human tissues.
Schroeder et al. (52) reported that Cr was significantly correlated with
Ba, Mn and Ni in 14 to 18 of the 29 human tissues examined. Ba and Mn
correlated in 15 of 28 tissues in another report (53). In our data, hair
Ba, Cr, Mn and Ni were significantly intercorrelated in both children and
adults. As found for a majority of human tissues (54), hair showed a
significant correlation between Cu and Fe and (in adults) between Cr and
Mn. Cr and Pb did not correlate significantly in hair (Table 6) nor
in 20 out of 29 other human tissues (52). Although exceptions to these
trends can be found between the data, hair does not appear to be unique
among body tissues in its compositional variations.
The highly significant correlations in hair between Cd, Cu and Pb
have been found by other investigators (16-18). As in another study (17),
only hair Se, a nonmetal, showed a negative correlation with certain hair
metals. The Fe - V correlation in hair is interesting in view of evidence
that V, like Fe, is bound to transferrin in blood (65) and that hemoglobin
and tissue respiration are decreased in Vanadium toxicity (72).
Hair growth in humans is a complex cyclic process involving non-growing
and growing phases of variable duration for each fiber plus differences in
17
-------
growth rate depending on age and sex. In addition scalp hair grows faster
and is retained longer in summer than in winter (51).
Metal concentrations are known to differ in hair from the same
individual and even at different locations along single fibers or fiber
bundles (4,22,23,30-38,40-47,50). These hair trace metal variations have
been considered by investigators as evidence for
(a) external contamination
(b) body intake variations
(c) seasonal trends
The relatively large number (7.6%) of scalp hair Se values rejected
as outliers may be related to the use of special shampoos containing Se
(16). Exogenous Se on hair was shown by Bate (23) to be resistant to
removal by a variety of washing procedures.
Complexity in interpretation of hair trace metal data is increased by
evidence indicating that physiological uptake by hair of trace elements is
not a simple growth phenomenon (36,44). For this study, hair was sampled
without restriction as to location on the scalp or distance from the scalp.
Consequently, effects observed in the covariate data could possibly be
related to sampling technique such as differences in distance from the scalp
between male and female samples. However, area differences would not be
affected by these sampling techniques since variations in the sampling
were randomized across all areas.
Laboratory washing of hair before metal analyses has been a point of
contention (2,3,14,22-32,34-38). A recent report (36) indicates that the
binding of certain metals to hair is not as strong as has been previously
assumed by many investigators. After comparison testing of five techniques,
18
-------
the washing procedure described by Harrison et al. (14) was selected for
use in this study. The detergent used in this procedure was demonstrated
to be similar in effect to commercial shampoos, which agrees with the
findings of another investigator for non-ionic detergents (24). In an
earlier EPA hair study (1), a metal chelating agent (EDTA) was included
in the hair washing procedure, but use of this reagent was not recommended
in a follow-up report (2). Other investigators have subsequently concurred
with the opinion that EDTA should not be used in washing hair before
analyses for metals (33,34).
The relationship of hair content to tissue content and metabolic
status are separate and complex issues that should not be confused with
exposure relationships.
Some evidence indicates that hair trace element content can reflect
whole body content (38,55,58,67) or content in specific tissues (3,38,40,
43-45,50,57,58,64,68). When hair content does not reflect other tissue
values (15,38,66,71), hair can reflect the metabolic or health status (38,
43,47,64,68) while the blood and other tissue values do not (9,38,47,56,59-64,
66).
The absence of a demonstratable relationship between hair content
and media values for some elements in this study is not definitive, but
may simply indicate that:
(a) the exposure difference was not sufficient for a correlation
effect on hair trace metal levels to be observed or
(b) that the media indices employed were not representative
of the overall metal exposure for the population sampled.
Future hair trace element content reports from other geographic areas
will help to clarify the utility of scalp hair as a community exposure monitor.
19
-------
ro
o
SCALE TO LEFT FOR Cd, Cr, AND Ni;
SCALE TO RIGHT FOR Cu, Pb, Mn, AND Zn mg/m2/month
UI
f
4
r
S
z 0.5
«£
0
^
o
0.1
0
rr
Ni
Cu
MHM;
Pb
Mn
Zn
MMM
Cd
RIVERHEAO
Ni
Cf
Cu
Pb
MMEM
Zn
Mi
^^
QUEENS
^L_
Ui
Cr
Cd
III
^MEBI
Zn
Pb
MM
Mn
-
20.0
f*
e
10.0*
|
f+
£
d
2.0
n
BRONX "
FIGURE 1. TRACE METAL LEVELS BY ELEMENT AND COMMUNITY FOR DUSTFALL
-------
SCALE TO LEFT FOR Cd, Cr, AND Ni;
SCALE TO RIGHT FOR Cu, Pb, Mn AND Zn yg/g
«^^M
Cr
__
_ Cd
^^^
t
Pb
MMM
Zn
Mn
MMB>
^i
Cr
Cd
Ni
ill
IM^M
Zn
Pb
Pll
Ml
^^^
a
Zn
^M
Pb
Cr
Cd
Cu
Mn
~
111
100
50
10
1000
500
100
RIVERHEAD
QUEENS
BRONX
FIGURE 2. TRACE METAL LEVELS BY ELEMENT AND COMMUNITY FOR HOUSEDUST
-------
SCALE TO LEFT FOR B, Li, Hg, Ag AND Sn
SCALE TO RIGHT FOR Ba, Fe X TO-2, Se X TO3, AND V yg/g
30
ro
ro
OL
C
Ul
o
"S, 20
01
10
SnBape
Li
Hg.
Se
Ba
Li
Hg Sn
Fe
RIVERHF.AD
QUEENS
Li
Ba
Sn
Hg
Aq
Fe
Se
300
200
100
2*
o
<
X .
0>
I
o
X
-------
ro
GO
20
10
SCALE TO LEFT FOR Cd AND Ni;
SCALE TO RIGHT FOR Cu, Pb, Mn AND Zn Mg/g
Pb
Ni
Cu
Zn
NI
Cd
n
Cu
RIVERHEAO
QUEENS
Pb
1300
Zn
200
100
£
BRONX
25
0
FIGURE 3. TRACE METAL LEVELS BY ELEMENT AND COMMUNITY FOR SOIL
-------
ro
0-1
2-3
4-5
6-7 8-9
CHILDREN'S AGE IN YEARS
10-11
12-13
14-15
FIGURE 4. GEOMETRIC MEAN SCALP HAIR Pb FOR CHILDREN BY AGE AND SEX
-------
2.0
ro
CJl
1.6
I L2-
0.8
0.4
0-1
2-3
4-5
6-7
8-9
10-11
12-13
14-15
CHILDREN'S AGE IN YEARS
FIGURE 5. GEOMETRIC MEAN SCALP HAIR Mn FOR CHILDREN BY AGE AND SEX
-------
ro
0-1
4-5 fi-7 8-9
CHILDREN'S AGE IN YEARS
10-11
12-13
14-15
FIGURE 6. GEOMETRIC MEAN SCALP HAIR Ba FOR CHILDREN BY AGE AND SEX
-------
ro
26-30 31-35 3640 41-45 46-60
ADULT'S AGE IN YEARS
ONLY 1 MALE AND 1 FEMALE IN THIS CATEGORY,
NO ACCEPTABLE MEAN VALUE IS AVAILABLE.
FIGURE 7. GEOMETRIC MEAN SCALP HAIR Ba FOR ADULTS BY AGE AND SEX
-------
ro
oo
55
49.5
44jO
38.5
33.0
ge
-27.5
"
22.0
16.5
11.0
5.5
\
16-20
21-25*
26-30
ONLY 1 HALE AND 1 FEMALE HERE IN THIS CATEGORY,
SO NO ACCEPTABLE MEAN VALUE IS AVAILABLE.
31-35 36-40
ADULT AGE IN YEARS
41-45
4640
60*
FIGURE 8. GEOMETRIC MEAN SCALP HAIR Cu FOR ADULTS BY AGE AND SEX
-------
TABLE 1. SAMPLE PREPARATION AND ANALYTICAL METHODS
Metals Preparation Analysis
Cd, Pb Oxygen combustion AA aspiration
Cu, Mn, Zn, Fe Oxygen combustion AA aspiration
As, Hg, Se Oxygen combustion AA on vapor
Li Oxygen combustion Flame photometry
Ag Oxygen combustion ES
Ba, Be, B, Cr Dry ashing ES
Co, Ni, V, Sn Dry ashing ES
Remarks:
1. Managanese in hair was evaluated from ES data when detection from AA
was found to be inadequate.
2. AA = atomic absorption
ES = emission spectroscopy
29
-------
TABLE 2. NUMBER OF PARTICIPANTS BY AGE, SEX, AND AREA OF RESIDENCE
0-5
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
51 +
Total
Riverhead
Mai
5
41
31
7
0
2
7
12
8
6
1
120
es Females
15
25
9
3
1
9
9
10
7
1
2
91
211
Queens
Males
16
27
17
2
0
7
10
10
7
0
2
98
Females
6
20
4
1
2
8
9
8
6
2
3
69
167
Bronx
Males
8
26
8
1
1
0
2
6
7
1
3
63
Females
7
17
5
1
0
5
10
6
6
0
0
57
120
30
-------
TABLE 3. DEMOGRAPHIC CHARACTERISTICS OF ADULTS BY AREA OF RESIDENCE
Smoking Patterns (%)
Never
Ex
Current
River head
37.6
16.5
45.9
Queens
44.2
24.7
31.2
Bronx
36.7
30.6
32.7
Education of Head of Household (%)
< High School 15.3 10.4 26.5
High School 48.2 50.6 44.9
> High School 36.5 39.0 28.6
31
-------
TABLE 4. HAIR COLOR BY AGE CATEGORY AND SEX
Adults
Brown
Blond
Red
Black
Grey & White
Unknown
Male
67
10
2
15
7
1
Female
76
17
2
8
4
2
Overall
143
27
4
23
11
3
Children
Male
122
43
6
6
0
2
Female
76
21
3
6
0
2
Overall
198
64
9
12
0
4
Total
102
109
211
179
108
287
32
-------
TABLE 5. TRACE METAL LEVELS IN HUMAN HAIR
IN CHILDREN* AND ADULTS*
Children
Barium
Boron
Cadmium
Chromium
Copper
Iron
Lead
Lithium
Manganese
Mercury
Nickel
Selenium
Silver
Tin
Vanadium
Adults
Barium
Boron
Cadmium
Chromium
Copper
Iron
Lead
Lithium
Manganese
Mercury
Nickel
Selenium
Silver
Tin
Vanadium
No. of
obs.
267
265
281
261
279
282
284
277
267
280
265
260
266
265
267
197
197
201
192
204
202
207
206
197
203
194
188
198
191
193
geo. mean
0.762
0.881
0.88
0.56
12.11
20.83
13.47
0.044
0.56
0.672
0.51
0.320
0.205
0.561
0.250
1.41
0.981
0.76
0.62
18.25
22.30
12.21
0.056
0.95
0.774
0.74
0.303
0.165
0.785
0.182
min.
0.054
0.030
0.14
0.076
1.01
2.70
2.12
0.009
0.05
0.048
0.036
0.025
0.007
0.034
0.010
0.121
0.037
0.08
0.06
2.22
3.60
1.96
0.009
0.07
0.050
0.045
0.025
0.007
0.048
0.009
max.
9.29
22.00
6.90
4.80
144.0
152.00
100.0
0.300
12.0
11.30
11.0
1.65
6.20
8.30
2.90
29.00
25.00
8.73
5.30
184.00
177.00
155.00
0.228
11.0
14.00
11.0
1.58
4.30
12.00
2.20
+1 geo.
0.271
.263
0.42
0.26
4.97
10.52
6.07
0.022
0.22
0.236
0.20
0.158
0.066
0.231
0.085
0.419
0.297
0.33
0.26
7.28
10.46
5.13
0.025
0.34
0.276
0.27
0.140
0.049
0.283
0.056
std. dev.
- 2.135
- 2.956
- 1.85
- 1.22
- 29.48
- 41.24
- 29.91
- 0.088
- 1.45
- 1.914
- 1 .30
- 0.645
- 0.637
- 1.361
- 0.738
- 4.77
- 3.23
- 1.74
- 1.46
- 45.75
- 47.53
- 29.07
- 0.083
- 2.67
- 2.17
- 2.07
- 0.653
Off fm f*
.550
- 2.17
- 0.588
* Children defined as ages 0 through 15, adults as ages greater than 15.
measurements in
All
33
-------
, TABLE 6
A SIGNIFICANCE TABLE FOR HAIR ELEMENT/ELEMENT CORRELATIONS
Pb
Cd
Cu
Hg
Li
Se
Fe
Ba
B
Cr
Ni
Ag
V
Sn
Mn
Zn
Pb
Cd
Cu
Hg
Li
Se
Fe
Ba
B
Cr
f
Ni
Ag
V
Sn
Mn
Zn
+ indicates significant positive correlation and
- indicates significant negative correlation, where significant
means p<0.05.
34
-------
TABLE 7. TRACE METALS IN DUSTFALL BY AREA*
No. of months
observed
Minimum
Maximum
Mean
Cadmium
Riverhead
Queens
Bronx
Chromium
Riverhead
Queens
Bronx
Copper
Riverhead
Queens
Bronx
Lead
Riverhead
Queens
Bronx
Manganese
Riverhead
Queens
Bronx
Nickel
Riverhead
Queens
Bronx
20
22
22
4
5
7
8
8
8
20
22
22
8
8
8
8
8
8
0.0
0.01
0.0
0.02
0.0
0.06
0.32
0.69
1.85
0.30
1.10
2.20
0.05
0.53
0.68
0.0
0.03
0.05
0.12
0.23
0.22
0.11
0.38
0.60
13.41
11.12
40.68
4.38
23.28
30.99
0.70
2.11
4.57
0.96
0.78
1.45
0.047
0.095
0.103
0.048
0.156
0.371
3.338
3.216
19.009
2.005
10.119
16.367
0.346
1.179
2.860
0.212
0.384
0.850
* Levels in mg/m2/month.
35
-------
TABLE 8. TRACE METALS IN HOUSEDUST BY AREA*
Cadmium
Riverhead
Queens
Bronx
Chromium
Riverhead
Queens
Bronx
Copper
Riverhead
Queens
Bronx
Lead
Riverhead
Queens
Bronx
Manganese
Riverhead
Queens
Bronx
Nickel
Riverhead
Queens
Bronx
Barium
Riverhead
Queens
Bronx
No. of Samples
37
29
33
37
29
33
37
29
33
37
29
33
37
29
33
37
29
33
37
29
33
Minimum
0.1
4.0
4.5
4.0
12.0
28.0
23.8
50.4
72.5
42.6
188.0
124.0
27.9
16.3
37.0
1.0
2.0
18.0
10.0
32.0
75.0
Maximum Geo. Mean
32.0
34.0
118.0
400.0
210.0
230.0
2250.0
900.0
1250.0
1630.0
3180.0
2930.0
158.0
455.0
275.0
95.0
200.0
250.0
430.0
4000.0
13000.0
7.6
13.3
14.6
31.7
42.3
45.0
109.3
196.9
233.7
278.9
629.0
766.3
79.9
131.0
153.4
14.6
26.1
42.0
65.2
137.6
312.4
(continued)
* Levels in yg/g
36
-------
TABLE 8. (CONTINUED)
Boron
Rlverhead
Queens
Bronx
Iron
Riverhead
Queens
Bronx
Lithium
Riverhead
Queens
Bronx
Mercury
Riverhead
Queens
Bronx
Selenium
Riverhead
Queens
Bronx
Silver
Riverhead
Queens
Bronx
Tin
Riverhead
Queens
Bronx
Vanadium
Riverhead
Queens
Bronx
No. of Samples
37
29
33
37
29
33
37
29
33
37
29
34
37
29
33
37
29
33
37
29
33
37
29
33
Minimum
1.0
2.0
4.0
2320.0
6650.0
2500.0
0.9
2.5
1.3
0.4
0.5
0.5
0.005
0.005
0.005
0.1
0.1
0.1
2.0
2.0
2.0
1.0
1.0
3.0
Maximum
110.0
370.0
410.0
16900.0
92400.0
41000.0
5.8
8.7
10.1
9.1
116.0
19.8
0.255
0.234
0.252
7.0
7.0
38.0
270.0
76.0
230.0
170.0
77.0
90.0
Geo. Mean
14.1
28.5
30.8
6091.6
11195.4
11431.2
3.0
4.3
4.9
1.9
5.9
3.6
0.038
0.062
0.049
0.4
0.5
0.8
6.4
6.2
7.8
14.3
18.8
40.2
37
-------
TABLE 9. TRACE METALS IN SOIL BY AREA*
Cadmium
Riverhead
Queens
Bronx
Copper
Riverhead
Queens
Bronx
Lead
Riverhead
Queens
Bronx
Manganese
Riverhead
Queens
Bronx
Nickel
Riverhead
Queens
Bronx
No. of Samples
56
56
60
56
56
60
56
56
60
56
56
60
56
56
60
Minimum
0.04
0.20
0.11
0.8
6.9
3.9
4.8
33.2
37.0
11.8
28.8
91.5
0.4
1.2
2.5
Maximum
1.72
8.75
5.82
28.0
1020.0
405.0
407.0
1010.0
1660.0
132.0
269.0
283.0
2.7
37.4
25.2
Geo. Mean
0.27
0.89
1.28
5.8
29.3
37.4
32.6
150.5
298.2
34.6
113.1
161.0
1.2
4.3
8.4
*Levels in pg/g,
38
-------
TABLE 10. CORRELATION COEFFICIENTS OF LOGS OF MEDIA TRACE METAL LEVELS*
Cadmi urn
Chromium
Copper
Lead
Manganese
Nickel
Zinc
Dustfall-Housedust
(N=99)
0.37
0.23
0.30
0.56
0.53
0.40
0.51
Dustf all -Soil
(N=172)
0.64
-
0.43
0.73
0.82
0.80
0.72
Housedust-Soil
(N=36)
0.25
-
0.52
0.77
0.54
0.41
0.56
* All of the above correlation coefficients are significant at p <0.005 except
for chromium dustfall-housedust, with p = 0.02.
39
-------
TABLE 11. TESTS OF SIGNIFICANCE OF THE EFFECT OF
SELECTED FACTORS ON SOIL TRACE METAL LEVELS*
Cadmium
Copper
Lead
Manganese
Nickel
Zinc
Dustfall
Levels
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
Homes Within
Areas
<0.0001
0.28
<0.0001
<0.0001
0.0006
<0.0001
Front Yard-Back Yard
Differences
.0004
.99
.0059
.0162
.0029
.0705
* Values given are probabilities of no relationship of factors to soil
trace metal levels given the observed differences in soil trace metal
levels between factor categories. Logs of soil and dustfall levels
were used in the analysis.
40
-------
TABLE 12. TESTS OF SIGNIFICANCE OF THE EFFECT OF SELECTED FACTORS ON SCALP HAIR TRACE
METAL LEVELS, USING (1) DUSTFALL AND (2) HOUSEDUST AS A MEASURE OF ENVIRONMENTAL EXPOSURE
(1) Dustfall
Metal
Cd
Cu
Cr
Pb
Mn
Ni
Dust-
fall
_
_
.006
.005
_
.02
Age
_
_
.005
.02
Children
Sex Educ.
.0003
.0001
_
.003
.0001
.0001
Hair Haircut
Color Location
,
_
_
_ _
.04
Dust-
fall
-
_
.01
_
~
Age
.005
_
_
_
m
Sex
.0001
.0001
-
.02
.0001
.0001
Adults
Hair Haircut
Educ. Color Location
.004
.
.
.02
-
- *
Smoking
_
-
_
.06
.
(2) Housedust
Metal
Cd
Cu
Cr
Pb
Mn
Mi
Ba
B
Fe
Li
Hg
Se
Ag
Sn
V
House-
dust
_
_
.005
-
.06
.0001
.
-
.0001
Mo Test
_
.0001
.0001
Age
.005
.03;
.ooor
.
_
-
Children
Sex Educ.
.006
.0001
.003
.0001
.0001
.0001
.004
.01
.01
.0002
.0003
.0003
Hair Haircut
Color Location
_
_ _
_
_ _
.04
.03
_ _
-
.06
.02
_
_
.03
House-
dust
_
_
_
.01
-
.
.0006
_
_
-
.04
No Test
-
-
.0001
Age
_
.004
_
.
-
-
.01
_
_
-
-
-
-
.0001
-
Sex
.0002
.0001
_
.02
.0001
.0001
.0001
.
.
.003
.001
-
-
-
Adults
Hair Haircut
Educ. Color Location
.005
_
.
.02
.
_
.05
.04
.04
...
_
_
-
-
.03
Smoking
-
-
-
.06
-
-
-
-
.04
-
-
.02
-
-
-
Values given are the probability of the observed difference in sample mean levels between factor categories assuming
no difference in the original population. Only values of .06 or less are listed.
-------
TABLE 13. DUSTFALL, HOUSEDUST, AND SCALP HAIR MEAN TRACE METAL
LEVELS BY COMMUNITY FOR SCALP HAIR METALS WITH A
SIGNIFICANT ENVIRONMENTAL EXPOSURE EFFECT
Chromium
Riverhead
Queens
Bronx
Lead
Riverhead
Queens
Bronx
Nickel
Riverhead
Queens
Bronx
Barium
Riverhead
Queens
Bronx
Mercury
Riverhead
Queens
Bronx
Tin
Riverhead
Queens
Bronx
Vanadium
Riverhead
Queens
Bronx
Mean
Dustfall
(mg/m2/mo)
0.05
0.16
0.37
2.01
10.12
16.37
0.21
0.38
0.85
_
-
-
Geometric
Mean
Housedust
(ug/g)
31.7
42.3
45.0
278.9
629.0
766.3
14.6
26.1
42.0
65.2
137.6
312.4
1.9
5.9
3.6
6.4
6.2
7.8
14.3
18.8
40.2
Childrens1
Geometric Mean
Scalp Hair
(pg/g)
0.52
0.45
0.80
11.74
14.07
17.13
0.50
0.40
0.70
0.64
0.63
1.26
0.48
1.00
0.73
0.54
0.44
0.83
0.20
0.24
0.40
Adults'
Geometric Mean
Scalp Hair
(ug/g)
-
10.39
14.50
12.88
-
1.11
1.36
2.34
0.66
1.01
0.67
-
0.12
0.20
0.35
42
-------
TABLE 14. GEOMETRIC MEAN SCALP HAIR TRACE METAL CONCENTRATIONS
BY SEX FOR CHILDREN AND ADULTS*
Children
Metal
Cd
Cu
Cr*
Pb
Mn
N1
Ba
B*
Fe*
Li
Hg
Se*
Ag
Sn
V*
Males
0.77
9.92
0.52
11.86
0.43
0.39
0.54
0.81
18.9
0.04
0.59
0.31
0.18
0.47
0.20
Females
1.14
17.91
0.62
17.52
0.88
0.79
1.30
1.00
23.4
0.05
0.84
0.34
0.28
0.77
0.37
Adults
Males
0.96
13.87
0.57
13.95
0.64
0.47
0.82
0.90
19.8
0.04
0.58
0.35
0.18
0.54
0.18
Females
0.62
25.06
0.63
10.97
1.34
1.14
2.41
1.04
24.0
0.05
0.99
0.27
0.15
1.17
0.19
* No differences by sex for these metals at the 0.05 level of
significance.
* Levels given in
43
-------
REFERENCES
1. Hammer, D. I., Finklea, J. F., Hendricks, R. H., and Shy, C. M., Hair
trace metal levels and environmental exposure. Am. 0. Epidemiol.
93, 84, 1971.
2. Hammer, D. I., Finklea, J. F., Hendricks, R. H., Hinners, T. A.,
Riggan, W. B., and Shy, C. M., Trace metals in human hair as a
simple epidemiologic monitor of environmental exposure. In Trace
Substances in Environmental Health - V. D. D. Hemphill, Ed. Uni-
versity of Missouri Press, Columbia, Mo., 1972, pp 25-38.
3. Klevay, L. M., Hair as a biopsy material: III. Assessment of
environmental lead exposure. Arch. Environ. Health 26, 169, 1973.
4. Kopito, L., Byers, R. K., and Shwachman, H., Lead in hair of children
with chronic lead poisoning. New Eng. J. Med. 276, 949, 1967.
5. Yamaguchi, S., Matsumoto, H., Matsuo, S., Kaku, S., and Hoshide, M.,
Relationship between mercury content of hair and amount of fish
consumed. HSMHA Health Reports 86_, 904, 1971.
6. Nelson, N. et al., Hazards of mercury. Environ. Res. 4_, 20, 1971.
7. Riggan, W. B., Hammer, D. I., Finklea, J. F., Hasselblad, V., Sharp,
C. R., Burton, R. M., and Shy, C. M., CHESS, A Community Health and
Environmental Surveillance System. In Proceedings of the Sixth
Berkeley Symposium on Mathematical Statistics ana Probability
(Vol. 6), University of California Press, Berkeley, Calif., 1972.
8. Lewis, G. P., Jusko, W. J., Coughlin, L. L., and Hartz, S., Contribution
of cigarette smoking to cadmium accumulation in man. The Lancet
1, 291, 1972.
9. National Research Council, Airborne Lead in Perspective. National
Academy of Sciences, Washington, D. C., 1972, p. 62, 64, 68.
10. Pinkerton, C., Creason, J. P., Hammer, D. I., and Colucci, A. V.,
Multi-media indices of environmental trace metal exposure in
humans. In Trace Element Metabolism in Animals - 2, W. Hoekstra,
J. Suttie, H. Ganther, and W. Mertz, Eds. University Park Press,
Baltimore, Md., 1974, pp. 465-469.
11. Underwood, E. J., Trace Elements in Human and Animal Nutrition, 3rd
ed., Academic Press, New York, N. Y., 1971.
12. Hinners, T. A., Kent, J., Terrill, W., Heiderscheit, T., Burton, R.,
and Colucci, A. V., Metals in atmospheric particulates collected by
Hi-Vol sampler and by dustfall bucket. EPA In-House Technical
Report, June 1972.
44
-------
13. U. S. Environmental Protection Agency, CHESS measurement methods.
In Health Consequences of Sulfur Oxides: A report from CHESS,
1970-1971.U. S. EPA, Research Triangle Park, N. C., 1974.
14. Harrison, W. W., Yurachek, J. P., and Benson, C. A., The determination
of trace elements in human hair by atomic absorption spectroscopy.
Clin. Chem. Acta 23, 83, 1969.
15. Schroeder, H. A. and Nason, A. P., Trace metals in human hair. J.
Invest. Derm. 53_, 71, 1969.
16. Gordus, A. A., Maher, C. C., and Bird, 6. C., Human hair as an
indicator of trace metal environmental exposure. In Proceedings of
the First Annual NSF Trace Contaminants Conference. Oak Ridge
National Laboratory, 1973.
17. Gordus, A. A., Environmental aspects of trace metals on human hair,
report no. 2. NSF Grant No. GI-35116, 1973.
18. Petering, H. G., Yeager, D. W., and Witherup, S. 0., Trace metal
content of hair: II. Cadmium and lead of human hair in relation to
age and sex. Arch. Environ. Health 27., 327, 1973.
19. Creason, J. P., McNulty, 0., Heiderscheit, L. T., Swanson, D. H., and
Buechley, R. W., Roadside gradients in atmospheric concentrations of
cadmium, lead and zinc. In Trace Substances in Environmental Health -
V. D. D. Hemphill, Ed. University of Missouri Press, Columbia, Mo.,
T972, p. 129.
20. Lagerwerff, J. V., and Sprecht, A. W., Contamination of roadside soil
and vegetation with cadmium, lead, and zinc. Environ. Sci. and Tech.
4, 583, 1971.
21. Hinners, T. A., Kent, J. L., and Terrill, W. J., Soil metal variations.
Materials Research and Standards (ASTHM), June 1972, p. 46.
22. Hambidge, K. M., Hambidge, C., Jacobs, M., and Baum, J. D., Low
levels of zinc in hair, anorexia, poor growth, and hypogeusia in
children. Pediat. Res. 6, 868, 1972.
23. Bate, L. C., Absorption and elution of trace elements on human hair
Int. J. Appl. Radio. Isotop. V7_, 417, 1966.
24. Bate, L. C., The use of activation analysis 1n procedures for the
removal and characterization of the surface contaminants of hair.
J. Foren. Sci. 1J), 60, 1965.
25. Petering, H. G., Yeager, D. W., and Witherup, S. 0., Trace metal
content of hair: I. Zinc and copper content of human hair in
relation to age and sex. Arch. Environ. Health 23_, 202, 1971.
45
-------
26. Klevay, L. M., Hair as a biopsy material. Am. J. Clin. Nutr. 23,
3/7, 1970. ~~~
27 ' wlyJns 2'if Fratjamie! R: ? ""I. 0. A., Pories, W. J., and Strain,
W. H., Malversation in hair analyses. Am. J. Clin. Nutr. 24, 893,
I / / I
28' rnnJ?ldP'+*- M> 'r^*"*1 *"l M' "" ' and Jacobs M" A« Hair chromium
concentration: Effects of sample washing and external environments.
Am. J. Clin. Nutr. 25_, 384, 1972.
29. Eads, E. A., and Lambdin, C. E. , A survey of trace metals in human
hair. Environ. Res. 6_, 247, 1973.
3°" anHU|nlk> J"nGiS]aSOn' *?' Maes' P" ^Millan, D. K. , D'Auria, J.,
and Pate, B. D., The variation of trace element concentrations in
single human lead hairs. J. Radioanal. Chem. ]_5, 115, 1973.
31. Gionanoli-Jakubczak, T., Greenwood, M. R. , Smith, J. C., and Clarkson,
I. w., Determination of total and inorganic mercury in hair by
fameless atomic absoprtion and of methylmercury by gas chromatography.
1*1 in. inem.
-------
40. Hambidge, K. M., and Baum, J. D.t Hair chromium concentrations of
human newborn and changes during infancy. Am. J. Clin. Nutr. 25,
376, 1972. ~~
41. Hambidge, K. M., Chromium nutrition 1n the mother and the growing
child. In Newer Trace Elements in Nutrition. W. Mertz and W.
Cornatzer, Eds., Marcel Dekker, New York, N. Y., 1971, pp. 171-194.
42. Reinhold, J. G., Kfoury, G. A., Ghalambor, M. A., and Bennett, J. C.,
Zinc and copper concentrations in hair of Iranian villagers. Am.
J. Clin. Nutr. 18., 294, 1966.
43. Pories, W. J., Strain, W. H., Rob, C. G., Henzel, J. H., Hennessen,
J. A., and Plecha, F. R., Trace elements and wound healing. In
First Annual Conference on Trace Substances in Environmental Health.
D. D. Hemphill, Ed. University of Missouri Press, Columbia, Mo., 1967,
pp. 114-133.
44. Strain, W. H., Pories, W. J., Flynn, A., and Hill, 0. A., Trace
element nutriture and metabolism through head hair analysis. In
Trace Substances in Environmental Health - V. D. D. Hemphill, Ed.
University of Missouri Press, Columbia, Mo., 1972, pp. 383-397.
45. Klevay, L. M., Hair as a biopsy material: I. Assessment of zinc
nutriture. Am. J. Clin. Nutr. 23_, 284, 1970.
46. Prasad, A., Metabolism of zinc and its deficiency in human subjects.
In Zinc Metabolism. A. Prasad, Ed. Thomas, Springfield, 111., 1966,
p. Z57J!
47. Reinhold, J. G., Kfoury, G. A., and Thomas, T. A., Zinc, copper and
iron concentrations in hair and other tissues: Effects of low
zinc and low protein intakes in rats. J. Nutr. 92_, 173, 1967.
48. Hunt, W. F., Pinkerton, C., McNulty, 0., and Creason, J. P., A
study in trace element pollution of air in 77 midwestern cities. In
Trace Substances in Environmental Health - IV. D. D. Hemphill, Ed.
University of Missouri Press, Columbia, Mo., 1970.
49. Anke, M., and Schneider, H. J., untersuchungen "uber den Mineral -
stoffgehalt der Frauen - und Mannerhaare. Deut. Z. Verdau.
Stoff. 22_, 31, 1962.
50. Klevay, L. M., Hair as a biopsy material: II. Assessment of copper
nutriture. Am J. Clin. Nutr. 23_, 1194, 1970.
51. Flesch, P., Hair growth. In Physiology and Biochemistry of the Skin.
S. Rothman, Ed. University of Chicago Press, Chicago, 111., 1954,
pp. 601-661.
47
-------
52. Schroeder, H. A., Nason, A. P., and Tipton, I. H., Chromium deficiency
as a factor in atherosclerosis. J. Chron. Dis. 23, 123, 1970.
53. Schroeder, H. A., Balassa, J. J., and Tipton, I. H., Essential trace
?n lLin-,oan: Man9anese- A study in homeostasis. J. Chron. Dis.
19, 545, 1966.
54. Schroeder, H. A., Nason, A. P., Tipton, I. H., Balassa, J. J.,
Essential trace metals in man: Copper. J. Chron. Dis. J9., 1007, 1966.
55. Nordberg, G. F., and Nishiyama, K., Whole-body and hair retention of
cadmium in mice. Arch. Environ. Health 24_, 209, 1972.
56. Becker, W. M. , and Hoekstra, W. 6., Dialysis studies of liver zinc
in zinc deficient and control rats. J. Nutr. 94_, 455, 1968.
57. Strain, W H., Berliner, W. P., Lankau, C. A., McEvoy, R. K. , Pories,
W. J., and Greerilaw, R. H. , Retention of radioisotopes by hair, bone
and vascular tissue. J. Nucl. Med. 5_, 664, 1964.
59. Beritie, T., Lead concentration found in human blood in association
with lead colic. Arch. Environ. Health 23, 289, 1971.
60. Piscator, M., and Axelsson, B. , Serum proteins and kidney function
after exposure to cadmium. Arch. Environ. Health 21_, 604, 1970.
61. Winge, D. R., and Rojagopalan, K. V., Purification and some
properties of Cd-binding protein from rat liver. Arch. Biochem.
Biophy. 153. 755, 1972.
62. Parizek, J., Ostadalova, I., Kalouskova, J., Babicky, A., and Benes,
J., The detoxifying effects of selenium interreations between
compounds of selenium and certain metals. In Newer Trace Elements in
Nutrition. W. Mertz and W. Cornatzer, Eds., Marcel Dekker, New Vork,
N. Y. , 1 971 , p. 96.
63. Mertz, W., and Roginski, E. E., Chromium Metabolism: The glucose
tolerance factor. In Newer Trace Elements in Nutrition. W. Mertz
and W. Cornatzer, Eds. Marcel Dekker, New York, N. Y., 1971, pp. 134-
1 35.
64. Hambridge, K. M., Chromium nutrition in the mother and the growing
child. In Newer Trace Elements in Nutrition. W. Mertz and W.
Cornatzer, Eds. Marcel Dekker, New Vork, N. V., 1971, pp. 169-194.
65. Hopkins, L. L., and Mohr, H. E., The biological essentiality of
vanadium. In Newer Trace Elements in Nutrition. W. Mertz and
W. Cornatzer, Eds. Marcel Dekker, New York, N. Y., 1971, pp. 203-204.
48
-------
66. McBean, L. D., Mahloudji, M., Reinhold, J. G., and Halsted, J. A.,
Correlation of zinc concentrations in human plasma and hair.
Am. J. Clin. Nutr. 24, 506, 1971.
67. Hambidge, K. M., and Rogerson, D. 0., Comparison of hair chromium
levels of nulUparous and parous women. Am. J. Obstet. Gynec.
103, 320, 1969.
68. Strain, W. J., Steadman, L. T., Lankau, C. A., Berliner, W. P. and
Pories, W. J., Analysis of zinc levels in hair for the diagnosis
of zinc deficiency in man. J. Lab. Clin. Med. 68., 244, 1966.
69. Piscator, M., and Lind, B., Cadmium, zinc, copper, and lead in
human renal cortex. Arch. Environ. Health 24_, 426, 1972.
70. Hammer, D. I., Colucci, A. V., Hasselblad, V., Williams, M. E.,
and Pinkerton, C., Cadmium and lead in autopsy tissues. J. Occup.
Med. 15., 956, 1973.
71. Schroeder, H. A., and Tipton, I. H., The human body burden of lead.
Arch. Environ. Health 17., 965, 1968.
72. National Research Council, Vanadium. National Academy of Sciences,
Washington, D. C., 1974, p.TT
49
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO
EPA-600/1-78-037a
4 TITLE AND SUBTITLE
HUMAN SCALP HAIR: AN ENVIRONMENTAL EXPOSURE INDEX FOR
TRACE ELEMENTS. I. Fifteen Trace Elements in
New York, N.Y. (1971-72)
3 RECIPIENT'S ACCESSION-NO.
5 REPORT DATE
May iQ-
. PERTORMI
MING ORGANIZATION CODE
7 AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO
John P. Creason, Thomas A. Hinners, Joseph E. Bumgarne
and Cecil Pinkerton
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Health Effects Research Laboratory and Environmental
Monitoring and Support Laboratory
Office of Research and Development
Research Triangle Park, N.C. 27711
10. PROGRAM ELEMENT NO.
1AA601
11. CONTRACT/GRANT NO.
12 SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park. N.C. 27711
13 TYPE OF REPORT AND PERIOD COVERED
RTP.NC
14 SPONSORING AGENCY CODE
EPA 600/11
15 SUPPLEMF.NTARY NOTES
16 ABSTHACT
Previous studies have revealed that hair trace element concentrations can
reflect exposure in cases of frank poisoning and deficiency. Correlations have
been found also in some populations living in regions where metallurgic processes
are conducted.
This study reports significant correlations between hair barium, chromium, lead,
mercury, nickel, tin, and vanadium content and exposures (as measured by analyses
for the corresponding elements in dustfall or housedust) within a single
metropolitan area. Age, sex, hair color, and smoking habits were included in the
statistical evaluation. Several metals showed a tendency to increase and decrease
together in the hair specimens in agreement with trends reported for other human
tissues.
It is acknowledged that hair has the capacity to adsorb and to release trace
elements in certain situations. However, population studies can compensate for
confounding influences by (1) a randomizing effect, by (2) an averaging effect, and
(3) by statistical rejection of unrepresentative data values. The relationship of
hair content to (a) content in other tissues and to (b) metabolic status are
separate and complex issues that 'should not be confused with (c) exposure relation-
c h -i r\c
KF Y WORDS AND DOCUMENT ANALYSIS
DESCHIHTORF,
trace elements
hair
indexes (ratios)
environmental surveys
l> IDENTIFIERS/OPEN ENDED TERMS
C COSATI I'lcld/CfOUp
06, T, F
UIS TRIBUTION STATEMENT
RELEASE TO PUBLIC
19 SECURITY CLASS (This Report)
UNCLASSIFIED
21 NO. OF PAGES
59
20
Tins page)
22 PRICE
EPA Form 2220-1 (9-73)
50
------- |