Content uploaded by Charles H Muller
Author content
All content in this area was uploaded by Charles H Muller
Content may be subject to copyright.
Occupational
and
Environmental
Medicine
1996;53:411-416
Semen
quality
of
men
employed
at
a
lead
smelter
Bruce
H
Alexander,
Harvey
Checkoway,
Chris
van
Netten,
Charles
H
Muller,
Timothy
G
Ewers,
Joel
D
Kaufman,
Beth
A
Mueller,
Thomas
L
Vaughan,
Elaine
M
Faustman
University
of
Washington,
Department
of
Epidemiology
B
H
Alexander
H
Checkoway
B
A
Mueller
T
L
Vaughan
University
of
Washington,
Department
of
Environmental
Health
H
Checkoway
T
G
Ewers
J
D
Kaufman
E
M
Faustman
University
of
British
Columbia,
Department
of
Health
Care
and
Epidemiology
C
van
Netten
University
of
Washington,
Department
of
Urology
C
H
Muller
University
of
Washington,
Child
Development
and
Mental
Retardation
Center
E
M
Faustman
Correspondence
to:
Dr
Elaine
M
Faustman,
Department
of
Environmental
Health,
Box
357234,
University
of
Washington,
Seattle,
WA,
98195,
USA.
Accepted
16
January
1996
Abstract
Objective-To
evaluate
the
effects
of
recent
and
long
term
occupational
lead
exposure
on
indicators
of
male
reproduc-
tive
health.
Methods-In
a
cross
sectional
study
of
male
employees
of
a
lead
smelter
(n
=
2469),
blood
samples
were
obtained
from
152
workers
including
119
who
also
pro-
vided
semen
samples.
Semen
analysis
and
serum
concentrations
of
testosterone,
fol-
licle
stimulating
hormone,
and
luteinising
hormone
were
used
as
indicators
of
reproductive
health.
Semen
and
hormone
variables
were
examined
in
relation
to
measures
of
current
and
long
term
body
lead
burden
estimated
from
current
blood
lead
concentrations
and
historical
blood
lead
monitoring
data.
Results-For
current
blood
lead
concen-
tration
groups
of
<
15,
15-24,
25-39,
>
40
ugldl,
the
geometric
mean
sperm
concen-
trations
were,
respectively,
79-1,
56-5,
62-7,
and
44-4
million
cells/ml
and
geo-
metric
mean
total
sperm
counts
were
186,
153,
137,
and
89
million
cells
(P
for
trend
0-04).
Compared
with
workers
with
blood
lead
concentrations
less
than
15
ug/dl,
workers
with
current
blood
lead
concen-
trations
of
40
ug/dl
or
more
had
an
increased
risk
of
below
normal
sperm
concentration
(odds
ratio
(OR)
8-2,
95%
confidence
interval
(95%
CI)
1-2-57.9)
and
total
sperm
count
(OR
2-6,
95%
CI
0-4-15-7),
based
on
World
Health
Organisation
standards.
Independent
of
current
lead
exposure,
sperm
concentra-
tion,
total
sperm
count,
and
total
motile
sperm
count
were
inversely
related
to
measures
of
long
term
lead
exposure.
No
association
was
found
between
lead
expo-
sure
and
measures
of
sperm
motility,
sperm
morphology,
or
serum
concentra-
tions
of
reproductive
hormones.
Conclusions-Blood
lead
concentrations
below
the
currently
accepted
worker
pro-
tection
criteria
seem
to
adversely
affect
spermatogenesis.
(Occup
Environ
Med
1996;53:411-416)
Keywords:
lead;
reproductive
health;
semen
quality
Injury
to
the
male
and
female
reproductive
systems
is
a
recognized
consequence
of
lead
poisoning.'
2It
is
unclear,
however,
whether
blood
lead
concentrations
below
current
regu-
latory
thresholds
harm
the
reproductive
sys-
tem.
The
United
States
Occupational
Safety
and
Health
Administration3
requires
that
a
person
with
a
single
blood
lead
measurement
of
60
jig/dl
or
greater,
or
periodic
measure-
ments
over
50
jug/dl
leaves
work
on
medical
grounds.
Whether
these
standards
adequately
protect
reproductive
health
has
implications
for
workers
in
occupations
where
lead
expo-
sure
is
common.4
Lead
has
been
linked
to
damage
or
dysfunc-
tion
of
the
male
reproductive
system
in
both
animal5
and
human
studies.5"'5
Occupational
exposure
to
lead
has
been
associated
with
impaired
semen
quality,8-1'
an
increased
fre-
quency
of
spontaneous
abortion
and
other
adverse
outcomes
of
pregnancies
fathered
by
workers
exposed
to
lead,'2-14
and
possible
decreases
in
fertility
rates.'5
Plausible
mecha-
nisms
for
impaired
spermatogenesis
are
a
direct
effect
on
testicular
function
or
an
effect
mediated
by
hormonal
imbalances.7
9
16-18
The
scientific
literature
associating
occupational
lead
exposure
with
impaired
reproductive
health
in
men
primarily
consists
of
studies
involving
populations
with
blood
lead
concen-
trations
in
excess
of
60
pg/dl.8-'0
Furthermore,
no
information
is
available
on
the
effects
of
long
term
exposure
to
lead
on
male
reproduc-
tive
health.
This
study
was
conducted
to
examine
the
effects
of
current
and
long
term
occupational
lead
exposures
on
semen
quality
and
repro-
ductive
hormone
concentrations.
Material
and
methods
STUDY
SETTING
The
study
was
conducted
in
June
and
July,
1993
at
the
Cominco
smelter
in
Trail,
British
Columbia.
Smelting
of
lead
ores
began
in
1899
and
primarily
produces
lead
and
zinc.
Modifications
of
the
smelting
and
refining
processes,
environmental
controls,
and
the
use
of
personal
protective
equipment
have
reduced
exposure
of
the
workers
over
time,
but
the
potential
for
excessive
exposure
still
exists.
Individual
blood
lead
concentrations
as
high
as
90, 85,
and
79
yg/dl
were
recorded
in
the
workforce
in
1970, 1980,
and
1990,
respectively.
The
mean
blood
lead
concentra-
tions
from
a
sample
of
monitored
workers
were
42, 41,
and
38
yg/dl,
respectively.
STUDY
DESIGN
The
study
was
a
cross
sectional
survey
com-
paring
variables
of
semen
quality
and
concen-
trations
of
reproductive
hormones
in
serum
41
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from
Alexander,
Checkoway,
van
Netten,
Muller,
Ewers,
Kaufman,
et
al
among
men
employed
at
the
smelter.
All
male
workers
employed
on
1
May,
1993
(n
=
2123)
and
workers
laid
off
in
the
preceding
year
(n
=
346)
were
recruited
by
postal
ques-
tionnaire.
To
encourage
participation
the
study
was
discussed
in
meetings
with
small
groups
of
workers,
and
a
follow
up
postcard
and
two
repeated
mailings
of
the
questionnaire
were
sent
to
non-responders.
The
question-
naire
asked
for
data
on
reproductive
history,
sociodemographic
characteristics,
and
per-
sonal
habits,
such
as
tobacco
and
alcohol
use,
and
invited
workers
to
donate
blood
and
semen
samples.
SAMPLE
COLLECTION
AND
ANALYSIS
A
field
laboratory
was
established
in
the
nurse's
station
at
the
smelter.
Informed
con-
sent
forms,
previously
approved
by
the
University
of
Washington
Human
Subjects
Committee,
describing
study
purposes,
risks,
and
benefits,
were
given
to
participants.
Semen
sample
donors
were
provided
with
sample
collection
containers,
instructions
on
sample
collection
and
transport,
and
a
brief
questionnaire
to
record
information
pertaining
to
the
analysis
of
the
semen
sample,
including
time
of
collection,
if
any
sample
was
spilled,
period
of
abstinence
from
ejaculation,
history
of
reproductive
tract
infections,
fevers,
and
use
of
hot
tubs
and
saunas
in
the
previous
three
months.
Participants
were
requested
to
abstain
from
ejaculation
for
48
hours
before
the
sample
collection.
Semen
samples
were
collected
at
home
or
on
site,
delivered
to
the
field
labora-
tory
within
one hour
of
collection,
and
incu-
bated
at
370C
until
liquification
was
complete.
Complete
semen
analysis,
including
sperm
concentration,
total
sperm
count,
percentage
motility,
and
percentage
rapid
and
linear
motility,
was
completed
within
one
hour
of
sample
collection
according
to
the
World
Health
Organisation
(WHO)
protocol.'9
Semen
smears
were
made
and
later
stained
with
Bryan-Leishman
stain
for
assessment
by
one
trained
technician
of
sperm
morphology
with
the
strict
morphology
criteria
described
by
Kruger
et
al.20
Computer
assisted
sperm
analysis
tech-
niques21
were
used
to
relate
measures
of
lead
exposure
to
objective
measures
of
the
motility
characteristics
of
motile
sperm
in
semen.
With
phase
contrast
videomicrography
the
images
of
at
least
100
motile
sperm
were
recorded
and
subsequently
analysed
with
a
Hamilton-
Thorne
IVOS
analyser,
software
version
10-H,
at
60
frames
per
second.
The
temperature
of
the
slide
and
sample
was
maintained
at
370C
during
the
recording
process.
The
mean
val-
ues
for
measures
of
sperm
motility,2'
the
veloc-
ity
of
the
average
path
(jm/s),
the
straight
line
velocity
(gu/s),
the
curvilinear
velocity
(pm/s),
the
amplitude
of
lateral
head
deflection
(pm),
and
percentage
linearity
(straight
line
veloc-
ity/curvilinear
velocity)
were
computed
for
each
man
as
an
estimate
of
the
motility
char-
acteristics
of
the
motile
sperm
(velocity
of
average
path
>
5
4um/s)
in
the
sample.
Two
venous
blood
samples
were
drawn
by
the
company
occupational
health
nurse.
One
sample
was
collected
in
an
EDTA
coated
tube
and
refrigerated
until
analysed
for
blood
lead
with
graphite
furnace
atomic
absorption
spec-
trophotometry.2
Serum
was
collected
from
the
second
sample
for
measurement
of
con-
centrations
of
testosterone,
follicle
stimulating
hormone
(FSH),
and
luteinising
hormone
(LH)
by
radioimmunoassay.
Serum
samples
were
stored
at
-
20'C
in
the
field
and
at
-
80'C
thereafter.
Aliquots
of
blood
were
frozen
for
analysis
of
other
metals
potentially
toxic
to
the
male
reproductive
system
that
may
exist
in
the
smelter
environment,
including
zinc,
cadmium,
arsenic,
and
copper,
by
induc-
tively
coupled
plasma
mass
spectrometry
(ICP-MS).
Current
lead
burden
was
estimated
from
the
measured
blood
lead
concentrations.
Information
on
past
lead
exposure
was
obtained
from
company
work
history
and
blood
lead
monitoring
records
for
all
study
participants.
The
lead
biomonitoring
pro-
gramme
has
relied
on
blood
lead
surveillance
since
1980.
This
programme
targets
workers
in
the
highest
lead
exposure
areas,
but
reaches
all
work
areas
at
the
smelter.
Each
worker's
mean
blood
lead
concentration
during
the
10
years
preceding
the
study
was
used
as
the
pri-
mary
index
of
long
term
lead
exposure.
Peak
blood
lead
concentrations
during
the
preced-
ing
10
years
and
cumulative
years
of
employ-
ment
in
the
highest
lead
exposure
jobs
were
also
evaluated
as
measures
of
long
term
lead
exposure.
Summarised
work
history
and
blood
lead
monitoring
data
were
available
for
all
actively
employed
workers
and
were
used
to
compare
study
participants
with
all
employees
of
the
smelter.
DATA
ANALYSIS
The
study
participants
were
divided
into
four
blood
lead
concentration
categories,
<
15,
15-24,
25-39,
>
40
jig/dl,
for
both
current
and
long
term
blood
lead
concentrations.
The
standard
semen
quality
values
did
not
follow
a
normal
distribution
and
had
increasing
vari-
ability
at
the
higher
exposure
concentrations.
To
describe
these
distributions
and
their
rela-
tive
variation,
geometric
means
and
geometric
SDs
were
calculated
for
each
of
the
blood
lead
concentration
groups.
A
least
squares
regres-
sion
model
was
fitted
to
the
natural
logarithm
of
each
semen
quality
variable
to
analyse
linear
trends
with current
and
long
term
blood
lead
concentrations.24
Covariates
evaluated
in
these
models
were
age,
alcohol
consumption,
smok-
ing,
the
presence
of
other
metals
in
the
blood,
and
abstinence
before
sample
collection.
The
prevalence
of
below
normal
values
for
each
semen
quality
variable
was
determined
by
the
WHO
standard
values'9
for
evaluation
of
infertility.
Prevalence
odds
ratios
(OR)
and
associated
95%
confidence
intervals
(95%
CIs)25
were
calculated
to
estimate
the
risk
of
having
a
below
normal
value
for
each
semen
variable.
The
importance
of
potentially
con-
founding
variables,
including
age,
smoking,
alcohol
consumption,
the
concentrations
of
other
metals
in
the
blood,
and
abstinence
before
sample
collection
were
evaluated
with
412
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from
Semen
quality
of
men
employed
at
a
lead
smelter
Table
1
Employment
and
demographic
characteristics
of
eligible
study
participants
and
all
workers
employed
at
the
beginning
of
the
study
Study
participants
Non-semen
donors
Semen
All
current
Donors
All
Non-vasectomised
Employees*
Number
119
810
466
2121
Age
(mean)
39
7
42-2
42-2
42-3
Blood
lead
concentration
(ug/dl,
mean)
28-7
28-3
27.4
28-7
Current
work
area
(%):
Lead
smelter
27-7
18-5
16-1
22-9
Refinery
14-3
7-7
7-7
7-7
Maintenance
and
support
14-3
17
9
18
2 14
4
Zinc
production
17-6
22-0
21-5
24-0
Administration
or
technical
16-8
13-9
15-5
13-8
Other
7-5
200
210
17
2
Years
at
smelter
(mean)
17-0
18-5
18-2
Marital
status
(%):
Single
(never
married)
5
0
6-4
9
9
Ever
married
92-5
93-1
89-9
Didnotrespond
2-5
0.5
0-2
Current
smokers
(%):
t
Yes
13-4
23-0
25-4
Did
not
respond
37-8
30-0
28-5
>
10
Alcoholic
beverages
a
week
(%)
17-6
19-3
18-6
Want
children
in
the
future
(%)
19-3
79
12
9
No
pregnancy
after
trying
for
>
ly(%)§
16-7
11-2
12-9
Sought
fertility
consultation
(%)§
10-2
97
11-7
*Summarised
records
were
available
for
2121
of
2123
current
employees.
iIncludes
only
blood
lead
concentrations
from
workers
included
in
the
blood
lead
monitoring
system
in
1993
before
the
study.
tNot
available.
SEver
married
men.
Table
2
Characteristics
of
workers
who
donated
semen
samples
by
current
blood
lead
concentration
Blood
lead
concentration
(,ugldl)
<
15
15-24
25-39
>
40
Number
32
46
29
12
Blood
lead*
(ug/dl,
mean)
10-2
19-2
30-6
47-3
Age
(mean)
41-4
38-2
40
7
41-8
Years
of
service
(mean)
16-4
15-7
18-4
19-8
Current
smokers
(%):
Yes
12-5
13-0
6-9
33-3
Did
not
respond
46-9
39-1
34-5
16-7
>
10
Alcoholic
beverages
a
week(%)
15-6
10-9
24-1
33-3
No
pregnancy
after
trying
>
ly
(%)t
18-5
18-6
15-4
8-3
Sought
fertility
consultation
(%)t
14
8
9-3
7-7
8-3
*Detection
limit
=
5
,g/dl.
tEver
married
men.
logistic
regression.26
Covariates
were
included
in
the
model
if
the
adjusted
OR
differed
from
the
crude
OR
by
10%
or
more,
and
if
a
similar
change
occurred
with
other
covariates
in
the
model.
Means
and
covariate
adjusted
means
for
the
computer
assisted
sperm
analysis
motility
vari-
ables
and
serum
testosterone,
LH,
and
FSH
were
compared
by
analysis
of
variance
and
analysis
of
covariance24;
adjusted
means
are
presented
where
appropriate.
A
least
squares
regression
model
was
fitted
to
analyse
linear
trends.
Prevalences
of
abnormally
low
testos-
terone
or
high
LH
and
FSH,
based
on
normal
laboratory
values,
are
also
presented.
Results
Questionnaires
were
returned
by
929
of
the
2469
eligible
male
employees.
Of
these
152
donated
blood
samples,
including
1
19
workers
who
also
provided
semen
samples.
Compared
with
the
non-vasectomised workers
who
returned
the
questionnaire,
but
did
not
pro-
vide
semen
samples
(n
=
466),
semen
sample
donors
were
slightly
younger
and
had
fewer
years
of
service
at
the
smelter
(table
1).
Compared
with
non-donors,
fewer
semen
donors
were
current
smokers
or
consumed
10
or
more
alcoholic
beverages
a
week.
A
greater
proportion
of
the
semen
sample
donors
than
non-donors
reported
wanting
children
in
the
future
and
had
been
unsuccessful
in
producing
a
pregnancy
after
trying
for
at
least
one
year.
The
workers
who
returned
questionnaires
were
quite
similar
to
all
currently
employed
male
workers
with
available
work
history
records
(n
=
2121)
for
age,
mean
blood
lead
concentrations
from
surveillance
in
the
previ-
ous
six
months,
and
current
work
area
(table
1).
The
semen
donors,
however,
were
more
likely
to
have
participated
in
the
blood
lead
surveillance
in
the
previous
six
months
than
workers
with
available
work
histories
(66-4%
v
46-2%).
The
semen
donors
were
younger
than
Table
3
Geometric
means
(SD)
of
standard
variables
of
semen
and
arithmetic
means
SDs
of
computer
assisted
sperm
analysis
measures
of
motility
by
current
blood
lead
concentration
Blood
lead
concentration
(figldl)
<
15
15-24
25-39
>
40
Pvalue*
Number
32
46
29
12
Standard
semen
quality:
Sperm
concentration
(million
cells/ml)t
79-1
56-5
62-7
44.4
0-11
(GSD)t
2-7
3-1
2-5
4
8
Total
sperm
(million
cells)t
186-0
153-0
137-0
89
1
0-04
(GSD)t
3-2
3-7
2-5
6
9
Motile
sperm
(%)
458
56-4
56-3
36.8
0-16
(GSD)t
1
7
1-3
1
3
3
4
Rapid
and
linear
motile
sperm
(%)t
13.1
19
4
18
8
15
9
0
97
(GSD)4
3-4
3
1
2-8
4-9
Strict
normal
morphology
(%)t
17
7
18
4
17
3
16
2
0
81
(GSD)*
2
5 2-3
2-2
2-8
Computer
assisted
sperm
analysis:
Velocity
of
the
average
path
(um/s)§
70
0
70
3
69
5
70-8
0
90
(SD)
2
2
1-8
2-2
3-6
Straight
line
velocity
(um/s)§
59
2
60-5 59-7
61
4
0-67
(SD)
2
2
1-8
2-3
3-7
Curvilinear
velocity
(um/s)§
117
7
116
8
115
0
110-4
0-34
(SD)
3-3
2-7
3.3
5-4
Amplitude
of
lateral
head
deflection
(um)§
4-7
4-6
4-5
4
0
0
05
(SD)
0-2
0
1
0-2
0-2
Linearity
(%)§
50-2
51
8 52-0
55-0
0-87
(SD)
1
3
1.1
1-3
2-1
*From
least
squares
regression
model.
tGeometric
mean.
tGSD
=
geometric
SD.
§Age
adjusted
mean.
413
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from
Alexander,
Checkoway,
van
Netten,
Muller,
Ewers,
Kaufman,
et
al
Table
4
Adjusted
ORs
for
variables
of
semen
quality
that
were
below
World
Health
Organisation
normal
levels
by
current
blood
lead
concentration
groups
Blood
lead
concentration
(ugldl)
<
15
15-24
25-39
>
40
Reference
OR*
(950%
CI)
OR*
(95%
CI)
OR*
(95%
CI)
Sperm
concentration
(million
cells/ml)
<
20t
1
4-3
(0-8,
23-3)
2-1
(0-3,
14-5)
8-2
(1-2,
57-9)
Total
sperm
(million
cells)
<
40t
1
1-7
(0-4,
7-3)
1-0
(0-2,
5
6)
2-6
(0-4,
15-7)
Strict
normal
morphology
(%)
<14t
1
1.3
(04,40)
1-6(0-5,5-3)
1-6(0-3,8-4)
Motile
sperm
(%)
<
50t
1
0-4
(0-2,
1-0)
0-5
(0-2,
1-5)
0-8
(0-2,
3-1)
Rapid
linear
motile
sperm
(%)
<
25t
1
0-5
(0-2,
1-2)
0-6
(0-2,
1-8)
0-4
(0-1,
1-7)
*Adjusted
for
age
group
(
<
35,
36-45,
>
46),
smoking
status
(yes,
no,
missing).
t1Minimum
normal
level
as
established
by
the
World
Health
Organisation.
1tKruger
strict
morphology
normal
level.
the
overall
working
population
and
were
more
likely
to
work
in
the
smelter
or
the
lead
refinery.
The
mean
blood
lead
concentration
for
semen
donors
from
the
sample
drawn
for
the
study
was
22-4
yg/dl
with
a
range
of
5
(detec-
tion
limit)
to
58
yg/dl.
Table
2
shows
the
dis-
tribution
of
study
participants
by
current
blood
lead
concentration.
The
mean
age
was
similar
among
the
groups
of
differing
blood
lead
concentrations,
but
the
highest
blood
lead
concentration
group
had,
on
average,
more
years
of
service
at
the
smelter,
a
larger
propor-
tion
of
current
smokers,
and
were
more
likely
to
consume
10
or
more
alcoholic
drinks
a
week.
A
history
of
difficulty
in
producing
a
pregnancy
and
having sought
medical
advice
for
fertility
problems
were
more
common
among
men
with
current
blood
lead
concen-
trations
less
than
25
yg/dl.
The
geometric
mean
sperm
concentration
and
total
sperm
count
were
inversely
related
to
current
blood
lead
concentration,
with
the
largest
effects
detected
among
men
with
blood
lead
concentrations
of
40
ig/dl
or
more
(table
3).
The
linear
relation
between
blood
lead
concentration
and
sperm
count
was
not
altered
after
covariant
adjustment
for
age,
smoking,
alcohol
use,
period
of
abstinence,
and
blood
concentrations
of
other
metals.
Compared
with
those
with
current
blood
lead
concentrations
of
less
than
15
,g/dl,
workers
with
more
than
40
gg/dl
had
an
increased
risk
of
having
a
below
WHO
normal
sperm
con-
centration
(adjusted
OR
8-2,
95%
CI
1-2-57-9)
and
total
sperm
count
(adjusted
OR
2-6,
95%
CI
0A4-15-7,
table
4).
Controlling
for
the
effects
of
alcohol
consumption,
pres-
ence
of
other
metals
in
the
blood,
zinc,
and
cadmium
in
particular,
period
of
abstinence
before
sample
collection,
history
of
reproduc-
tive
difficulties,
the
use
of
hot
tubs
and
saunas,
and
history
of
reproductive
tract
infection
did
not
decrease
the
OR
point
estimates.
The
percentage
of
sperm
meeting
the
strict
normal
morphology
criteria
did
not
vary
with
blood
lead
concentration
(table
3).
Only
a
slightly
increased
risk
of
a
below
normal
mor-
phology
was
found
in
the
group
with
the
high-
est
blood
lead
concentration
(OR
1-6,
95%
CI
0-3-8-4,
table
4).
The
estimates
of
percentage
of
motile
sperm
and
percentage
of
rapid
and
linear
sperm
showed
no
consistent
associations
between
increasing
lead
exposure
and
poor
sperm
motility.
The
mean
percentage
motile
sperm
was
lowest
in
the
group
with
40
yug/dl
blood
lead
concentration;
however,
there
was
no
trend
for
mean
percentage
of
rapid
and
lin-
ear
motile
sperm
(table
3).
Moreover,
the
OR
for
below
WHO
normal
values
for
these
vari-
ables
were
less
than
1-0
in
the
groups
exposed
to
lead
(table
4).
The
lack
of
association
between
blood
lead
concentration
and
sperm
motility
was
confirmed
by
the
results
of
the
computer
assisted
sperm
analysis
for
motility
(table
3),
although
the
amplitude
of
lateral
head
deflection
did
decrease
with
increasing
blood
lead
concentrations.
When
classified
by
long
term
exposure
to
lead,
calculated
from
the
mean
blood
lead
concentrations
of
the
preceding
10
years,
20,
28,
45,
and
26
men
had
10
year
averages
of
<
15,
15-24,
25-39,
and
>
40
gg/dl,
respec-
tively.
To
evaluate
the
effects
of
long
term
lead
Table
5
Geometric
mean
and
geometric
SD
of
selected
variables
of
semen
quality
by
mean
blood
lead
concentration
of
the
preceding
10
years
for
workers
with
current
blood
lead
concentration
below
40
Blood
lead
concentration
(ugdl)
<
15
15-24
25-39 4
0
P
value"
Number
20
28
44
15
Mean
chronic
blood
lead
concentration
(ug/dl)
10-3
20-7
31-5
47-9
Sperm
concentration
(million
cells/mil)
Geometric
mean
72-4 72-0
61-8
49-4
0-29
(GSD)
2-8
2-8
2-7
3-3
Total
sperm
(million
cells)
Geometric
mean
201-0
188-0 144-0
108-0
0-07
(GSD)
2-9
3-1
3-3
3-6
Strict
normal
morphology
(%)
Geometric
mean
16-7
19-4
16-8
20-3
0-78
(GSD)
2-4
2-2
2-4
2-4
*From
least
squares
regression
model.
414
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from
Semen
quality
of
men
employed
at
a
lead
smelter
Table
6
Mean
(SD)
and
percentage
with
abnormal
clinical
concentrations
of
testosterone,
LH,
and
FSH
by
current
blood
lead
concentration
Blood
lead
concentration
(ugldl)
<
15
15-24
25-39
40
Pvalue*
Number
36
53
41
22
Mean
blood
lead~ug/dl
10
1
19-4
31-0
45-5
Testosterone
(nmol/l)
Mean
15-5
18-2
16-1
15
2
049
SD
6-2
5-7
6-1
4-5
<
9-7
nmol/l
(%)t
8-3
5-7
9-8
13-6
Luteinising
hormone
(JU/1):
Mean
4-6
4-5
4-2
4
9
0-46
SD
1-6
1-5
1
1
2-4
>
13-9
IU/l(%)
00
00
00
0.0
Follicle
stimulating
hormone
(IU/1):
Mean
5-1
5.0
4-4
5-8
0
49
SD
2-5
2-8
1-6
5.3
>
12-2
IU/
(%)t
0.0
1.9
0.0
9
9
*From
least
squares
regression
model.
tBelow
laboratory
normal.
tAbove
laboratory
normal.
exposure
independently
of
current
high
blood
lead
concentrations
workers
with
current
blood
lead
concentrations
of
40
yg/dl
(n
=
12)
were
excluded.
The
prevalence
of
workers
with
below
normal
variables
of
semen
quality
was
similar
across
the
long
term
exposure
groups.
The
geometric
mean
sperm
concen-
tration,
and
total
sperm
count,
however,
showed
consistent
inverse
trends
with
long
term
blood
lead
concentration
(table
5).
Sperm
morphology
and
all
measures
of
sperm
motility
were
not
associated
with
long
term
lead
exposure.
We
found
minimal
differences
in
mean
serum
testosterone,
LH,
and
FSH
across
the
concentrations
of
either
current
or
long
term
lead
exposure
(table
6).
Controlling
for
the
effects
of
age,
smoking,
alcohol,
and
other
metals
did
not
change
these
findings.
Furthermore,
these
results
did
not
change
materially
when
stratified
by
total
sperm
count
(<
40
v
>
40
million
cells/ejaculate).
Discussion
The
inverse
associations
found
between
cur-
rent
blood
lead
concentration
and
total
sperm
count
and
sperm
concentration
is
further
evi-
dence
that
lead
exposure
has
a
direct
effect
on
spermatogenesis.
The
effect
was
most
pro-
nounced
among
men
with
current
blood
lead
concentrations
over
40
gg/dl.
Moreover,
we
found
an
association
between
long
term
lead
exposure
and
a
decrease
in
sperm
number
that
is
apparently
independent
of
current
lead
bur-
den.
Neither
current
nor
long
term
exposure
to
lead
was
consistently
related
to
indicators
of
sperm
morphology,
sperm
motility,
or
repro-
ductive
hormone
concentrations.
The
interpretation
of
our
results,
with
refer-
ence
to
previous
research,
must
consider
the
inherent
limitations
and
potential
advantages
of
this
study.
Similar
to
most
studies
requiring
voluntary
donation
of
semen
samples,
the
overall
participation
rate
was
low.
Fewer
than
40%
of
the
eligible
workers
returned
the
ques-
tionnaire,
and
only
21%
of
the
non-vasec-
tomised
men
who
returned
the
questionnaire
volunteered
to
donate
semen
samples.
However,
workers
who
returned
question-
naires
were
representative
of
all
eligible
sub-
jects
for
age
and
blood
lead
concentrations,
and
semen
donors
and
study
participants
who
did
not donate
semen
samples
had
similar
demographic
and
exposure
profiles.
Semen
sample
donors
were
more
likely
to
work
in
the
lead
smelter
and
were
thus
more
often
subject
to
blood
lead
surveillance.
Cross
sectional
studies
of
semen
quality
can
be
subject
to
selection
bias
if
men
with
histo-
ries
of
reproductive
difficulties
are
overrepre-
sented
among
semen
sample
donors,
especially
if
self
selection
is
related
to
exposure
level.
In
this
study
the
self
reported
prevalence
of
periods
of
subfertility
was
somewhat
higher
among
the
semen
donors
than
other
non-
vasectomised
workers.
The
excess
of
subfertil-
ity,
however,
was
limited
to
men
with
current
blood
lead
concentrations
below
25
pg/dl.
Thus,
a
selection
bias
based
on
self
perceived
infertility
would
most
likely
result
in
underesti-
mates
of
the
true
associations
between
lead
exposure
and
semen
quality.
Logistical
considerations
of
this
cross
sec-
tional
study
permitted
the
collection
of
only
one
semen
and
one
blood
sample
from
each
participant.
The
validity
of
a
semen
analysis
improves
when
repeated
samples
are
taken,
although
a
single
measurement
is
generally
adequate
to
characterise
reliably
the
lowest
and
highest
tails
of
the
distributions.27
The
pulsatile
release
of
testosterone,
LH,
and
FSH
requires
sequential
sampling
over
a
period
of
several
hours
for
diagnostic
assessment
in
an
individual
man.
Nevertheless,
a
single
sample,
preferably
drawn
at
the
same
time
of
day,
is
considered
adequate
to
detect
a
population
shift
in
cross
sectional
studies.28
The
samples
drawn
for
semen
donors
were
drawn
in
the
morning,
usually
between
6
and
8
am.
The
samples
for
the
other
blood
donors
were
scat-
tered
throughout
the
day.
Nevertheless,
the
results
of
the
hormone
analysis
were
essen-
tially
the
same
when
limited
to
semen
donors.
A
notable
advantage
of
this
study
is
the
rela-
tive
stability
of
the
workforce
and
the
availabil-
ity
of
historical
employment
and
blood
lead
monitoring
data.
Work
history
and
blood
lead
data
permitted
comparisons
of
participants
and
non-participants
and
an
evaluation
of
the
effects
of
long
term
exposure
on
variables
of
semen
quality,
which
until
now
has
not
been
reported.
Measurement
of
other
metals
allowed
control
of
potential
confounding
from
other
reproductive
toxicants
that
might
be
encountered
in
the
same
environment.
This
study
provides
further
evidence
that
spermatogenesis
may
be
impaired
when
blood
lead
concentrations
exceed
40
yg/dl.
Of
par-
ticular
interest
was
the
increased
risk
of
below
normal
sperm
concentrations
and
counts
which
were
related
to
subfertility.19
Similar
results
have
been
reported
by
Lancrajan
et
al,8
who
found
an
increase
in
the
prevalence
of
altered
spermatogenesis
related
to
exposure
among
battery
factory
workers
with
group
mean
blood
lead
concentrations
from
41-75
jig/dl.
Studies
of
other
populations
of
men
exposed
to
higher
lead
concentrations
than
in
our
study
have
consistently
reported
similar
adverse
effects
on
spermatogenesis.9-10
Conversely,
a
study
of
battery
factory
workers
415
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from
Alexander,
Checkoway,
van
Netten,
Muller,
Ewers,
Kaufman,
et
al
in
Sweden
reported
no
differences
in
variables
of
semen
quality
between
workers
with
mean
lead
concentrations
of
45
pug/dl
and
22
yg/dl."1
Independent
of
current
blood
lead
concen-
trations,
sperm
count
and
concentration
were
lower
among
workers
with
long
term
increases
in
blood
lead
concentrations.
This
finding
may
indicate
that
recovery
from
the
effects
of
recent
lead
exposure
is
not
rapid
or
that
men
with
historically
high
blood
lead
concentra-
tions
may
be
more
susceptible
to
lower
expo-
sures.
The
few
workers
with
current
high
exposures
and
the
high
correlation
between
current
and
long
term
lead
concentrations,
however,
did
not
allow
for
the
separation
of
effects
from
these
exposures.
In
contrast
to
previous
research,8
10
we
found
no
relation
between
lead
exposure
and
sperm
morphology
and
motility.
It
is
notewor-
thy,
however,
that
in
the
studies
reporting
an
association
between
lead
exposure
and
sperm
motility
and
morphology,
the
blood
lead
con-
centrations
of
the
lowest
exposure
groups
were
similar
to
those
of
the
highest
exposure
group
in
this
study.
Moderate
increases
in
serum
FSH
and
LH17
18
and
decreased
serum
testosterone'7
have
been
attributed,
albeit
inconsistently,9
16
to
occupational
exposure
to
lead.
A
low
serum
testosterone
or
increases
in
the
concentrations
of
the
gonadotropins
would
indicate
a
hor-
monal
mechanism
for
the
suppression
of
sperm
production
found.
The
lack
of
such
an
association
implies
that
lead
directly
affects
testicular
function.
The
results
of
this
study
show
that
measur-
able
impairment
of
spermatogenesis
is
associ-
ated
with
current
and
long
term
blood
lead
concentrations
below
those
requiring
a
man
to
leave
work
for
medical
reasons,
as
specified
by
the
United
States
Occupational
Safety
and
Health
Administration.3
Assessment
of
the
ultimate
consequences
of
moderate
lead
con-
centrations
(25-50
gg/dl)
on
fertility
and
peri-
natal
outcomes
will
require
in
depth
follow
up
studies
of
cohorts
with
occupational
and
envi-
ronmental
exposures
in
that
range.
We
are
grateful
to
the
membership
of
United
Steel
Workers
Locals
480
and
9705
and
to
Cominco
Metals
for
supporting
this
study
and
for
providing
a
great
deal
of
assistance.
Special
thanks
are
also
due
to
Tom
Wynn,
Peter
Fulcher,
and
Graham
Kenyon
for
help
with
the
organisation
of
the
study,
and
to
Linda
Coxall
for
generous
help
with
blood
sample
collection.
We
also
thank
Jesara
Schroeder
and
Genna
Ratiu
for
their
assistance
in
the
field
laboratory,
Karin
Yeatts
for
compiling
historical
blood
lead
monitoring
data,
and
Iris
Nielsen
and
Changpu
Yu
for
sperm
morphology
assessments
and
laboratory
coordination.
This
research
was
sponsored
by
United
States
National
Institutes
of
Health
Grant
ES04696,
United
States
National
Institute
for
Occupational
Safety
and
Health
Grant
R03-OH02966,
and
the
British
Columbia
Workers'
Compen-
sation
Board.
BHA
was
supported
by
NIEHS
Training
Grant
T32-ES07262.
Hormone
analyses
were
performed
in
the
labo-
ratory
of
Dr
William
J
Bremner
with
support
from
National
Institutes
of
Health
Population
Center
Grant
P50-HD12629.
Agency
for
Toxic
Substances
and
Disease
(ATSDR).
Toxicological
profile
for
lead.
Washington:
US
Department
of
Health
and
Human
Services,
US
Government
Printing
Office,
1990.
2
Lockitch
G.
Perspectives
on
lead
toxicity.
Clin
Biochem
1993;26:371-81.
3
Office
of
the
Federal
Register.
Code
of
federal
regulations:
occupational
safety
and
health
standards.
Sub-part
Z.:
toxic
and
hazardous
substances-lead.
Washington,
DC:
Office
of
the
Federal
Register,
National
Archives
and
Records
Administration,
1985.
(29
CFR
1910-1025.)
4
Centers
for
Disease
Control.
Surveillance
for
occupational
lead
exposure-United
States,
1987.
Morbidity
and
Mortality
Weekly
Report
1989;38:687-94.
5
Winder
C.
Reproductive
and
chromosomal
effects
of
occu-
pational
exposure
to
lead
in
the
male.
Reprod
Toxicol
1989;3:221-33.
6
US
Congress,
Office
of
Technology
Assessment.
Reproductive
health
hazards
in
the
workplace.
Washington:
US
Government
Printing
Office,
1985.
(OTA-BA-266.)
7
Cullen
MR,
Kayne
RD,
Robins
JM.
Endocrine
and
repro-
ductive
dysfunction
in
men
associated
with
occupational
inorganic
lead
intoxication.
Arch
Environ
Health
1984;
39:421-40.
8
Lancranjan
I,
Popescu
HI,
Gavenesu
0,
Klapsh
I,
Serbanescu
M.
Reproductive
ability
of
workmen
occupa-
tionally
exposed
to
lead.
Arch
Environ
Health
1975;30:
396-401.
9
Assennato
G,
Claudio
P,
Molinini
R,
Candela
RG,
Baser
ME,
Altamura
BM,
Giorgino,
R.
Sperm
count
suppres-
sion
without
endocrine
dysfunction
in
lead-exposed
men.
Arch
Environ
Health
1986;41:387-90.
10
Lerda
D.
Study
of
sperm
characteristics
in
persons
occupa-
tionally
exposed
to
lead.
Am
Ind
Med
1992;22:567-71.
11
Wildt
K,
Eliasson
R,
Berlin
M.
Effects
of
occupational
exposure
to
lead
on
sperm
and
semen.
In:
Clarkson
TW,
Nordberg
GF,
Sager
PR,
eds.
Reproductive
and
develop-
mental
toxicity
of
metals.
New
York:
Plenum
Press,
1983.
12
Lindbohm
MJ,
Sallemen
M,
Anttila
A,
Taskinen
H,
Hemminki
K.
Paternal
occupational
lead
exposure
and
spontaneous
abortion.
Scand
Y
Work
Environ
Health
1991;17:95-103.
13
Nordstrom
S,
Beckman
L,
Nordenson
I.
Occupational
and
environmental
risks
in
and
around
a
smelter
in
northern
Sweden.
V:
spontaneous
abortion
among
female
employ-
ees
and
decreased
birth
weight
in
their
offspring.
Hereditas
1979;90:291-6.
14
Beckman
L,
Nordstrom
S.
Occupational
and
environmental
risks
in
and
around
a
smelter
in
northern
Sweden.
IX:
fetal
mortality
among
wives
of
smelter
workers.
Hereditas
1982;97:1-7.
15
Gennart
JP,
Buchet
JP,
Roels
H,
Ghyselen
P,
Ceulemans
E,
Lauwerys
R.
Fertility
of
male
workers
exposed
to
cad-
mium,
lead
or
manganese.
Am
J
Epidemiol
1992;135:
1208-19.
16
Gustafson
A,
Hedner
P,
Schutz
A,
Skerfving
S.
Occupational
lead
exposure
and
pituitary
function.
Int
Arch
Occup
Environ
Health
1989;61:277-81.
17
Rodamilans
M,
Osaba
MJM,
To-Figueras
J,
Rivera
F,
Marques
M,
Perez
P,
Corbella
J.
Lead
toxicity
on
endocrine
testicular
function
in
an
occupationally
exposed
population.
Human
Toxicol
1988;7:125-8.
18
Ng
TP,
Goh
HH,
Ng
YL,
Ong
HY,
Ong
CN,
Chia
KS,
et
al.
Male
endocrine
functions
in
workers
with
moderate
exposure
to
lead.
BrJ
Ind
Med
199
1;48:485-91.
19
World
Health
Organization.
WHO
laboratory
manualfor
the
examination
of
human
semen
and
semen-cervical
mucus
interaction.
Cambridge:
Cambridge
University
Press,
1992.
20
Kruger
TF,
Menkveld
R,
Stander
FSH,
Lombard
CJ,
Van
der
Merwe
JP,
van
Zyl
JA,
Smith
K.
Sperm
morphologic
features
as
a
prognostic
factor
in
in-vitro
fertilization.
Fertil
Sterd
1986;46:118-23.
21
Boyers
SP,
Katz
DF,
Davis
RO.
Automatic
semen
analysis.
Current
Problems
in
Obstetrics,
Gynecology
and
Fertility
1989;12:165-200.
22
Henry
JB.
Clinical
diagnosis
and
management
by
laboratory
methods,
16th
ed.
Philadelphia:
WB
Saunders,
1979:
513-5.
23
Matsumoto
AM,
Paulsen
CA,
Hopper
BR,
Reber
RW,
Bremner
WH.
Human
chorionic
gonadotropin
and
tes-
ticular
function:
stimulation
of
testosterone,
testosterone
precursors,
and
sperm
production
despite
high
E2
levels.
Y
Clin
Endocrinol
Metab
1983;56:720-8.
24
Fisher
LD,
van
Belle
G.
Biostatistics.
A
methodology
for
the
health
sciences.
New
York:
John
Wiley,
1993.
25
Rothman
KJ.
Modern
epidemiology.
Boston:
Little-Brown,
1986.
26
Breslow
NE,
Day
NE.
Statistical
methods
in
cancer
research.
Vol
1.
The
analysis
of
case-control
studies.
Lyon,
France:
International
Agency
for
Research
on
Cancer,
1980.
27
Schrader
SM,
Turner
TW,
Breitenstein
Mj,
Simon
SD.
Longitudinal
study
of
semen
quality
of
unexposed
workers
I:
study
overview.
Reprod
Toxicol
1988;2:183-90.
28
Schrader
SM,
Turner
TW,
Breitenstein
Mj,
Simon
SD.
Measuring
male
reproductive
hormones
for
occupational
field
studies.
J
Occup
Med
1993;35:57
4-6
416
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from
doi: 10.1136/oem.53.6.411
1996 53: 411-416Occup Environ Med
B H Alexander, H Checkoway, C van Netten, et al.
smelter.
Semen quality of men employed at a lead
http://oem.bmj.com/content/53/6/411
Updated information and services can be found at:
These include:
References
http://oem.bmj.com/content/53/6/411#related-urls
Article cited in:
service
Email alerting
the box at the top right corner of the online article.
Receive free email alerts when new articles cite this article. Sign up in
Notes
http://group.bmj.com/group/rights-licensing/permissions
To request permissions go to:
http://journals.bmj.com/cgi/reprintform
To order reprints go to:
http://group.bmj.com/subscribe/
To subscribe to BMJ go to:
group.bmj.com on July 15, 2011 - Published by oem.bmj.comDownloaded from
