Content uploaded by Susan Meacham
Author content
All content in this area was uploaded by Susan Meacham on Aug 27, 2015
Content may be subject to copyright.
~~~~~~~~~~~~~~~~~
-
-
-----
-
Effects
of
Boron
Supplementation
on
Bone
Mineral
Density
and
Dietary,
Blood,
and
Urinary
Calcium,
Phosphorus,
Magnesium,
and
Boron
in
Female
Athletes
Susan
L.
Meacham,
L.
Janette
Taper
and
Stella
L.
Volpe3
1
2~~~~~~~~~
'Department
of
Human
Nutrition,
Winthrop
University,
Rock
Hill,
South
Carolina;
2Department
of
Human
Nutrition
and
Foods,
Virginia
Polytechnic
Institute
and
State
University,
Blacksburg,
Virginia;
3Department
of
Nutritional
Sciences,
University
of
California,
Berkeley,
California
The
effects
of
boron
supplementation
on
blood
and
urinary
minerals
were
studied
in
female
college
students-i
7
athletes
and
1
1
sedentary
con-
trols-over
a
one-year
period.
The
athletes
had
lower
percent
body
fat
and
higher
aerobic
capacities
than
sedentary
controls.
Athletic
subjects
con-
sumned
more
boron
in
their
normal
diets
than
sedentary
subjects;
all
other
dietary
measures
were
similar
between
the
two
groups.
The
athletes
showed
a
slight
increase
in
bone
mineral
density,
whereas
the
sedentary
group
showed
a
slight
decrease.
Serum
phosphorus
concentrations
were
lower
in
boron-supplemented
subjects
than
in
subjects
receiving
placebos,
and
were
lower
at
the
end
of
the
study
period
than
during
baseline
analy-
sis.
Activity
depressed
changes
in
serum
phosphorus
in
boron-supplemented
subjects.
Serum
magnesium
concentrations
were
greatest
in
the
sedentary
controls
whose
diets
were
supplemented
with
boron,
and
increased
with
time
in
all
subjects.
A
group
x
supplement
interaction
was
observed
with
serum
magnesium;
exercise
in
boron-supplemented
subjects
lowered
serum
magnesium.
In
all
subjects,
calcium
excretion
increased
over
time;
in
boron-supplemented
subjects,
boron
excretion
increased
over
time.
In
all
subjects,
boron
supplementation
affected
serum
phosphorus
and
magnesium,
and
the
excretion
of urinary
boron.
-
Environ
Health
Perspect
102(Suppl
7):79-82
(1994)
Key
words:
boron
supplementation,
female
athletes,
serum
minerals,
urinary
mineral
excretion,
calcium,
phosphorus,
magnesium
Introduction
We
performed
a
study
to
ascertain
the
effects
of
boron
supplementation
on
bone
mineral
density
(BMD)
and
the
mineral
status
of
college
female
athletes.
The
results
of
selected
diet
and
mineral
analy-
ses
will
be
reported
here
(1,2).
Previous
studies
have
linked
vigorous
exercise
with
amenorrhea,
and
associated
this
condition
with
the
likelihood
of
increased
incidence
of
stress
fractures
and
increased
risk
of
osteoporosis,
particular-
ly
among
young
female
athletes.
Early
studies
by
Drinkwater
et
al.
(3),
Lindberg
et
al.
(4),
and
Marcus
et
al.
(5)
have
reported
that
the
BMD
of
the
verte-
brae
is
lower
in
amenorrheic
athletes
than
in
eumenorrheic
or
normal
subjects
(Table
1).
Researchers
at
the
United
States
Department
of
Agriculture,
Agricultural
This
paper
was
presented
at
the
International
Symposium
on
Health
Effects
of
Boron
and
Its
Compounds
held
16-17
September
1992
at
the
University
of
California,
Irvine,
California.
Address
correspondence
to
Dr.
Susan
L.
Meacham,
Department
of
Human
Nutrition,
Winthrop
University,
Rock
Hill,
SC
29733.
Telephone
(803)
323-2101.
Fax
(803)
323-2347.
E-mail
MeachamS@Winthrop.Edu
Research
Service
(USDA,
ARS),
Grand
Forks
Human
Nutrition
Research
Center
in
Grand
Forks,
ND
have
conducted
a
number
of
human
studies
on
boron
sup-
plementation
(6-9).
Their
findings
can
be
briefly
stated
as
follows:
a)
High
dietary
boron
lowered
urinary
calcium
and
total
calcium,
and
increased
blood-
ionized
calcium,
17P
estradiol,
and
testosterone.
b)
Blood-ionized
calcium
was
higher
and
serum
calcitonin
was
lower
with
boron
repletion
than
with
boron
depletion.
c)
The
researchers
con-
cluded
that
changes
seen
with
boron
sup-
plementation
induced
changes
in
post-
menopausal
women
consistent
with
the
prevention
of
calcium
and
bone
loss.
Thus,
boron
may
have
a
preventive
or
therapeutic
effect
that
helps
to
diminish
bone
mineral
loss
in
susceptible
popula-
tions.
Methodology
In
the
present
study,
28
female
students
were
recruited
from
Virginia
Polytechnic
Institute
and
State
University,
and
Ferrum
College.
The
subjects
were
screened
for
such
factors
as
oral
contraceptive
use
and
smoking.
The
selected
subjects
were
then
assigned
to
four
treatment
groups
in
a
sin-
gle-blind
study;
ten
athletic
subjects
received
the
boron
supplement,
while
seven
received
a
cornstarch
placebo,
and
six
sedentary
subjects
received
the
boron
sup-
plement,
while
five
received
the
placebo.
All
subjects
were
instructed
to
take
daily
either
a
boron
supplement,
3
mg/day
(Tri-
Boron,
Twin
Laboratories,
Inc.,
Table
1.
Amennorhea
and
bone
mineral
density.
Measurement
Assessment
Reference
Amenorrheic
=
1.12
DPA-L1-L4
Drinkwater
et
al.
1984
(3)
Eumenorrheic
=
1.30
Amenorrheic
=
1.08
DPA-vertebrae
Lindberg
et
al.
1984
(4)
Normal
=
1.2-1.6
Amenorrheic
=
1.51
CT-L1-L2
Marcus
et
al.
1985
(5)
Eumenorrheic
=
1.81
Abbreviations:
DPA,
dual
photon
absorptiometry;
CT,
calorimetric
test.
Environmental
Health
Perspectives
79
MEACHAM
ETAL.
Ronkonkoma,
NY),
or
a
placebo
(Revco
Pharmacy,
Blacksburg,
VA)
for
10
months.
Several
descriptive
tests
were
performed
to
assist
with
characterization
of
subjects.
A
physical
work
performance
test
(PWC170)
for
aerobic
capacity
was
performed
on
a
Monark
bicycle
ergometer
(Monark-
Cresent
AB,
Vargerg,
Sweden)
to
confirm
the
athletic
status
of
the
subjects.
Skinfold
measurements
of
the
triceps,
suprailiac,
and
thigh
were
used
to
determine
the
subjects'
body
fat
percentages
(10).
To
characterize
the
dietary
patterns
of
the
free-living
subjects,
duplicate
plates
were
collected
over
3
days
at
the
beginning
of
the
study.
The
macronutrient
composi-
tion
of
the
subjects'
diets
was
determined
from
aliquots
of
homogenized
composites
by
proximate
analysis.
Concentrations
of
several
selected
dietary
minerals
were
deter-
mined
by
either
atomic
absorptiometry
(AA),
colorimetry,
or
inductively
coupled
argon
plasma
spectroscopy
(ICAP).
Bone
mineral
density
analyses
were
per-
formed
on
a
LUNAR
DPIII
bone
density
absorptiometer
at
Montgomery
Regional
Hospital
in
Blacksburg,
VA.
Fasting
blood
samples
were
collected
in
mineral-free
tubes,
and
blood
plasma
and
serum
fractions
were
frozen in
polyethylene
tubes
for
subsequent
analysis.
Blood
miner-
als
were
determined
by
several
analytical
procedures.
Serum
total
calcium
was
deter-
mined
by
AA.
Also,
plasma
total,
and
nor-
malized
and
ionized
calcium
were
deter-
mined
immediately
following
collection,
using
a
NOVA
7
electrolyte
analyzer.
Blood
phosphorus
was
determined
colori-
metrically,
and
blood
magnesium
was
determined
by
AA.
Serum
boron
was
deter-
mined
by
ICAP,
applying
a
low-tempera-
ture,
Teflon
tube,
wet-ashing
digestion
procedure
(11).
Urinary
calcium
was
analyzed
by
AA
and
ICAP;
urinary
phosphorus
was
determined
colorimetrically
and
by
ICAP;
and
urinary
magnesium
was
determined
using
AA
and
ICAP.
Urinary
boron
was
determined
by
ICAP
following
the
digestion
procedures
described
by
Hunt
and
Schuler
(11).
Results
and
Discussion
The
athletic
and
sedentary
subjects
were
similar
in
age
and
body
weight
(Table
2).
However,
their
body
fat
and
aerobic
capac-
ities
differed
significantly
(p<0.05).
The
athletes'
average
percent
body
fat
was
20.6,
compared
to
the
sedentary
groups'
percent
body
fat
of
25.8
(p<0.05).
The
subjects'
physical
work
capacity,
determined
by
V02
max,
was
higher
in
the
athletic
groups
Table
2.
Age,
body
weight,
body
fat
and
V02max.
Athletes
Sedentary
Parameter
(n=
17)
(n=
11)
Age,
years
19.8
±
1.4a
20.3
±
1.1
Body
weight,
kg
61.8
±
9.1
59.6
±
10.5
Body
fat,
%b
20.6
±
5.6
25.8
±
6.5
V02max,
L02/minb
2.9
±
0.5
2.1
±
0.4
aValues
shown
are
means
±
standard
deviation.
bSignificant
difference
between
groups
(p<0.05).
(2.9)
than
in
the
sedentary
groups
(2.1)
(p<0.05).
The
athletic
and
sedentary
controls
did
not
differ
in
their
daily
consumption
of
total
kilocalories,
percent
protein,
fat,
and
carbohydrate
as
determined
by
proximate
analysis
(Table
3).
Also,
the
intake
of
mil-
ligrams
of
dietary
calcium,
phosphorus,
and
magnesium
did
not
differ
between
activity
groups.
However,
the
athletes'
dietary
boron
intake
levels
were
higher
than
those
for
sedentary
controls
when
expressed
as
either
total
mg
boron/day
or
as
mg
boron/kg
dry
matter/day
(p<0.05).
Macromineral
intakes
fell
markedly
below
the
RDAs
for
both
groups
(12).
Bone
mineral
density
determinations
of
the
lumbar
vertebrae
were
higher
for
ath-
letes
than
for
sedentary
subjects
(p<0.05)
(Table
4).
Also,
over
time,
the
differences
in
BMD
between
the
two
activity
groups
were
significant
(p<0.05);
BMD
increased
by
0.03
g/cm2
in
the
athletic
group
and
decreased
by
-0.005
g/cm2
in
the
seden-
tary
group.
However,
bone
mineral
densi-
ties
did
not
seem
to
be
influenced
directly
by
boron
supplementation.
Other
changes
over
time
were
observed
in
serum
total
calcium
determinations.
Concentrations
of
serum
calcium,
when
determined
by
AA,
and
plasma-ionized
cal-
cium,
when
determined
by
the
NOVA
7,
increased
significantly
over
time.
This
increase
over
time
was
not
seen
with
plas-
ma
total
calcium
determinations
from
the
electrolyte
analyzer
(Table
5).
In
contrast,
serum
phosphorus
decreased
over
time
among
all
subjects.
In
addition,
serum
phosphorus
was
significantly
lower
in
boron-supplemented
subjects
than
in
subjects
receiving
the
placebo
(p<0.05).
A
significant
group
x
supplement
interaction
was
also
observed
(p<0.05).
At
both
test-
ing
times,
the
sedentary
group
fed
the
boron
supplement
had
the
lowest
serum
phosphorus
levels;
the
sedentary
placebo
group
had
the
highest
serum
phosphorus
levels
at
both
testing
times
(p<0.05).
Activity
depressed
the
effect
of
boron
sup-
plementation
observed
in
sedentary
con-
trols
(p<0.05)
(Table
5).
Serum
magnesium
levels,
like
calcium
levels,
increased
over
time
among
all
sub-
jects
(p<0.05).
Also,
a
significant
group
x
supplement
interaction
was
observed-
sedentary
subjects
who
were
fed
boron
sup-
plements
exhibited
higher
serum
magne-
sium
levels
than
athletic
subjects
who
were
fed
the
boron
supplement.
Mean
serum
magnesium
concentration
was
significantly
Table
3.
Subjects'
daily
dietary
intakes
assessed
by
laboratory
analyses.
Athletes
Sedentary
(n=
17)
(n=
11)
RDAs
Energy,kcal
1468
±
503a
1417
±
584
Protein,
%
14.1
±
5.1
14.9
±
5.2
12
Fat,
%
28.7
±
7.7
30.8
±
6.3
(<30)
Carbohydrate,
%
57.3
±
9.8
54.3
±
6.3
55-60
Calcium,mg
650±
558
714±442
1200
Phosphorus,
mg
915
±
616
840
±
330
1200
Magnesium,
mg
103
±
107
73
±
30
280
Boron,
pgb
1.5
±
1.3
0.7
±
0.3
a
Values
shown
are
means
±
standard
deviation.
b
Significant
difference
between
groups
(p<0.05).
Table
4.
Bone
mineral
density
for
activity
and
supplement
groups.
Bone
mineral
density,
g/cm2
Athletes
Sedentary
Boron
Placebo
Boron
Placebo
Time (n=
10)
(n=
7)
(n=
6)
(n=
5)
MonthO
1.27±0.14a
1.3±0.08
1.25±011
1.19±0.11
Month
10
1.3
±0.16
1.34±0.09 1.26±0.13
1.17
0.11
Difference
b
+0.024
±
0.049
+0.036
±
0.038
+0.012
±
0.044
-0.024
±
0.029
a
Values
shown
are
means
±
standard
deviation.
b
Significantly
greater
for
the
athletes
than
for
the
sedentary
group
(p<0.05).
Environmental
Health
Perspectives
80
EFFECT
OF
BORON
ON
FEMALE
MINERAL
STATUS
Table
5.
Pre-
and
post-test
blood
calcium
(mg/
DI).
Athletes
Sedentary
Boron
Placebo
Boron
Placebo
Time
(n=
10)
(n=
7)
(n=
6)
(n=
5)
TCa(AA)a
1
8.9±0.4b
8.8±0.2
9.1
0.2
9.0±0.5
2'
f9.6±0.3
9.8±0.3
9.8±0.2
9.7±0.4
TCa
(NOVA)
1
g
9.4
±
0.5
9.7
±0.6
9.4
0.4
9.7
±0.2
2
9.6±0.2
9.7±0.3
9.5±0.2
9.4±0.1
iCa
(NOVA)
a
1
4.6
±
0.3
4.7
0.2
4.7
0.1
4.5
0.0
2
4.7
±
0.2
4.7
0.2
4.7
0.1
4.7
0.2
P
(C)
a.c.d
1
4.7
±
0.4
4.7
±0.6
4.3
±0.5
5.1
±0.3
2
3.9
±
0.2
4.0
0.2 3.6
1.0
4.0
0.3
Mg(AA)a,d.e
1
1.5±0.3
1.7±0.2
1.9±0.3
1.7±0.2
2
2.0±0.0
2.0±0.1
2.4±0.4
2.0±0.1
Abbreviations:
AA,
atomic
absorptiometry;
NOVA,
NOVA
7
eloctrolytic
analyzer;
C,
calorimetry.
a
Significant
time
effect
(p<0.05).
b
Values
shown
are
means
±
standard
deviation.
c
Significant
supplement
effect
(p<0.05).
d
Significant
group
x
supplement
effect
(p<0.05).
e
Significant
graup
effect
(p<0.05).
'Time
2,
tests
performed
at
the
end
of
the
study.
I
Time
1,
tests
performed
at
the
beginning
of
the
study.
Table
6.
Pre-
and
post-test
urinary
calcium
and
phosphorus.
Athletes
Sedentary
Boron
Placebo
Boron
Placebo
Time
(n=
10)
(n=
7)
(n=
6)
(n=
5)
Ca(mg/day)(AA)a
1
d
50+45b
72±50
62±37
47±31
2
e
106
40
89
27
74
41
53
16
P
(mg/day)
(C)
1
760
791
719
394
755
364
480
316
2
601
280
606
178
767
312
672
240
Mg
(mg/day)
(AA)
1
53
30
67
±
44
60
23
53
33
2
118±98
72±34
70±36
52±33
B(pg
/day)(ICP)ac
1
0.7±0.6
0.7±0.3
0.7±0.3
0.5±0.3
2
2.8
1.6
0.8±0.8
1.1
±1.1
0.7±0.6
Abbreviations:
AA,
atomic
absorptiometry;
C,
calorimetry;
ICP,
inductively
coupled
argon
plasma
spectroscopy.
a
Significant
time
effect
(p<0.05).
b
Values
shown
are
means
±
standard
deviation.
c
Significant
group,
supplement
and
time
x
supplement
effects
(p<0.05).
d
Time
1,
tests
performed
at
the
beginning
of
the
study.
e
Time
2,
tests
performed
at
the
end
of
the
study.
higher
(p<0.05)
in
athletes
than
in
seden-
tary
controls.
Boron
analysis
of
subjects'
serum
was
attempted.
Following
wet-ashing,
low-
temperature,
Teflon
tube
(WALTTT)
digestions,
serum
analysis
revealed
values
in
the
0.15
to
0.31
pg/ml
range.
Because
of
the
small
amount
of
the
sample
and
the
low
concentration
of
boron
typically
expected
in
serum,
the
validity
of
these
values
warrants
additional
study
to
improve
the
standard
error
rate.
Urinary
calcium,
like
serum
calcium,
increased
over
time
(p<0.05)
(Table
6).
Urinary
phosphorus
and
magnesium
did
not
reflect
any
significant
changes
(p<0.05)
over
time,
either
between
active
and
inactive
groups
or
with
supplementa-
tion
(Table
6).
Urinary
boron,
however,
differed
significantly
over
time,
both
between
high-
and
low-activity
groups
and
with
supplementation
(p<O.O5).
Boron
excretion
increased
over
time;
excretion
was
higher
among
athletes
than
sedentary
controls
and
was
greater
in
boron-supple-
mented
subjects
than
in
those
receiving
the
placebo
(p<0.05).
At
the
end
of
the
experimental
period,
a
time
x
supplement
interaction
was
observed
in
which
subjects
who
had
received
boron
supplementation
for
10
months
exhibited
greater
boron
excretion
levels
than
subjects
who
received
the
placebos
(p<0.05).
Here
is
a
brief
summary
of
the
present
findings:
a)
Athletic
subjects
and
controls
differed
in
aerobic
capacity
and
percent
body
fat,
but
not
in
age
or
body
weight.
b)
Dietary
intakes
between
groups
did
not
differ
for
selected
nutrients
measured,
except
for
dietary
boron,
which
was
higher
in
athletes.
c)
Serum
and
plasma
calcium
increased
over
time
in
all
subjects.
d)
Serum
phosphorus
decreased
over
time
and
decreased
with
boron
supplementa-
tion
in
sedentary
subjects.
e)
Serum
mag-
nesium
increased
over
time,
and
increased
with
boron
supplementation
in
sedentary
subjects.
f)
Urinary
calcium
increased
over
time,
but
urinary
phosphorus
and
magnesium
did
not
change.
g)
Urinary
boron
increased
over
time
with
dietary
boron
supplementation.
Conclusion
Our
1-year
study
of
athletic
and
sedentary
female
college
students
indicated
that
boron
affected
blood
phosphorus
and
magnesium,
an
effect
modified
by
exercise.
The
dietary
patterns
were
similar
between
athletes
and
sedentary
individuals
(except
for
dietary
boron),
and
all
subjects
had
intakes
below
the
RDA
for
calcium,
phos-
phorus,
magnesium,
and
total
energy.
Before
preventive
or
therapeutic
recom-
mendations
can
be
made
with
regard
to
boron
supplementation
for
bone
disorders,
the
effects
of
low-
and
unknown
boron
intakes
and
of
boron
supplementation
need
further
investigation
as
regards
bone
metabolism
in
humans.
Recommendations
for
Further
Research
Recommendations
for
future
research
include
studies
on:
a)
the
effects
of
various
boron
supplementation
levels
at
various
physical
work
capacities
on
dietary,
blood,
and
urinary
Ca,
P,
Mg,
and
calcitonin
and
parathyroid
hormone
levels;
b)
the
com-
parison
of
effects
of
varied
levels
of
boron
supplementation
on
males
and
females
of
various
ages,
physical
work
capacities
(PWCs),
and
hormonal
conditions;
c)
the
effects
of
interactions
of
various
dietary
intakes
of
Ca,
phosphorus,
Mg,
and
boron
on
bone,
blood,
and
urine
minerals;
d)
methods
of
analysis
for
Ca,
phosphorus,
Mg,
and
boron
in
food,
blood,
and
urine;
e)
the
mechanism
of
action
for
boron
(in
vitro
and
animal
studies)
as
it
relates
to
bone
metabolism.
Human
studies
characterizing
"normal"
physiologic
levels
of
boron
are
needed
at
this
time
to
advance
our
understanding
of
boron's
role
in
normal
human
nutri-
tion-our
ultimate
goal.
To
accomplish
this
goal,
reliable
analysis
techniques
and
digestion
procedures
for
study
of
dietary
boron
are
needed.
REFERENCES
1.
Darnton
SM.
Effects
of
boron
supple-
mentation
on
bone
mineral
density,
blood,
urinary
calcium,
magnesium,
phosphorus
and
urinary
boron
in
female
athletes.
Ph.D.
dissertation.
Blacksburg,
VA:Virginia
Polytechnic
Institute
and
State
University,
1991.
2.
Volpe-Snyder
S.
The
effects
of
boron
supplementation
on
bone
mineral
density
and
hormonal
status
in
college
female
athletes.
Ph.D.
dissertation.
Blacksburg,
Volume
102,
Supplement
7,
November
1994 81
DARNTON
ETAL.
Virginia:Virginia
Polytechnic
Institute
and
State
University,
1991.
3.
Drinkwater
BL,
Nilson
K,
Chestnut
CH,
Bremner
WT,
Shainholtz
S,
Southward
MB.
Bone
mineral
content
of
amenorrheic
and
eumenorrheic
women.
N
Engl
J
Med
311:277-281
(1984).
4.
Lindberg
J
S,
Fears
WB,
Hunt
MM,
Powell
MR,
Boll
D,
Wade
CE.
Exercise
induced
amenorrhea
and
bone
density.
Ann
Intern
Med
101:647-648
(1984).
5.
Marcus
R,
Cann
C,
Madvig
P,
Minkoff
P,
Goddard
M,
Bayer
M,
Martin
M,
Gaudiani
L,
Haskell
W,
Genant
H.
Menstrual
function
and
bone
in
elite
women
distance
runners:
endocrine
and
metabolic
features.
Ann
Intern
Med
102:158-163
(1985).
6.
Nielsen
FH,
Mullen
LM,
Gallagher
SK.
Effect
of
boron
depletion
and
repletion
on
blood
indicators
or
calcium
status
in
humans
fed
a
magnesium-low
diet.
J
Trace
Elem
Exp
Med
3:45-54
(1990).
7.
Nielsen
FH.
New
essential
trace
elements
for
the
life
sciences.
Biol
Trace
Elem
Res
26-7:599-611
(1990).
8.
Nielsen
FH,
Hunt
AM,
Gallagher
SK,
Hunt
JR,
Hunt
CD,
Johnson
LK.
Effects
of
dietary
boron,
aluminum
and
magnesium
on
serum
alkaline
phosphatase,
calcium
and
phosphorus
and
plas-
ma
cholesterol
in
postmenopausal
women.
In:
Trace
Elements
in
Man
and
Animals
6
(Hurley
LS,
Keen
CL,
L6nnerdal
B,
Rucker
RB,
eds).
New
York:Plenum,
1988;
187-188.
9.
Nielsen
FH,
Hunt
CD,
Mullen
LM,
Hunt
JR.
Effect
of
dietary
boron
on
mineral
estrogen
and
testosterone
metabolism
in
postmenopausal
women.
FASEB
J
1:394-397
(1987).
10.
Jackson
A,
Pollock
M,
Ward
A.
Generalized
equations
for
predicting
body
density
of
women.
Med
Sci
Sports
Exer
12:
175-182
1980).
11.
Hunt
CD,
Schuler
TR.
Open-vessel,
wet-ash,
low-temperature
digestion
of
biological
materials
for
inductively
coupled
argon
plas-
ma
spectroscopy
(ICAP)
analysis
of
boron
and
other
minerals.
J
Micronutr
Anal
6:161-174
(1989).
12.
National
Academy
of
Sciences.
Recommended
Dietary
Allowances,
10th
ed.
Washington:
National
Academy
of
Sciences,
1989.
82
Environmental
Health
Perspectives
