United States
Environmental Protection
Agency
Health Effects
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S1-85/009 Aug. 1985
x°/EPA Project Summary
Assessment of Cadmium
Exposure and Toxicity
Risk in an American
Vegetarian Population
Rodney E. Willard
It has been postulated that non-
vegetarians may be exposed to less
cadmium than vegetarians because of
the cadmium-poor meat in their diet.
This study attempts to test this possi-
bility by measuring the cadmium expo-
sure and accumulation in a population
subgroup that includes many vegetari-
ans. The conclusions are: 1. no statisti-
cal difference in cadmium exposure
rate is demonstrated between the non-
vegetarians and lacto-ovo-vegetarians
of this study; 2. the cadmium exposure
rate of the small group of pure vegetar-
ians studied is in the direction antici-
pated (higher than nonvegetarians) but
is not large enough to achieve statisti-
cal significance; 3. cadmium accumula-
tion in the critical organ (kidney) of sub-
jects coming to autopsy is well below
the range associated with renal injury;
and 4. quantitative methods for mea-
suring total urine protein are not as sat-
isfactory for assessing tubular protein-
uria as those developed to measure the
small proteins that normally appear in
the glomerular filtrate.
This Project Summary was devel-
oped by EPA's Health Effects Research
Laboratory, Research Triangle Park, NC,
to announce key findings of the re-
search project that is fully documented
in a separate report of the same title
(see Project Report ordering informa-
tion at back).
Introduction
Only 3% of the cadmium used in the
United States is recycled. The remain-
der is dispersed into the biosphere
where it may contaminate our drinking
water and enter the food chain. Esti-
mates based on autopsy data and envi-
ronmental analysis have led to the fear
that lifetime exposure and accumula-
tion may be nearing a critical toxic level
in the general population. Schroeder's
animal experiments and several au-
topsy series have suggested a relation
between unrecognized cadmium accu-
mulation and hypertension. Up to 20%
of the total body burden in smokers
may be related to absorption from
tobacco-born cadmium. Analysis of
foods from cadmium-contaminated re-
gions shows variation in cadmium con-
tent. Cereals and other foods of plant
origin contain higher concentrations
than most foods of animal origin. Re-
gardless of whether cadmium enters
the body via the gastrointestinal tract
from food and water or the respiratory
tract via smoke or dust, the physiologic
biochemical pathways result in its even-
tual binding to metallothionein. This
5000 dalton protein has the greatest
binding affinity for cadmium of any pro-
tein studied thus far and easily carries
the cadmium with it through the
glomerular basement membrane into
the renal tubular filtrate. The normal
tubular function for conservation of
amino acids and proteins results in the
removal of cadmium-metallothionein
complex from the filtrate into the tubu-
lar mucosa. The cadmium remains se-
questered in the renal cortex for the rest
of the life of the individual. When a suf-
-------
ficient concentration has accumulated,
renal tubular dysfunction with cad-
miumuria, beta-2-microglobuluria, and
Fanconi syndrome become apparent.
Additional accumulation leads to de-
struction of the renal cortex. The
scarred contracted end stage kidney is
indistinguishable from that of glomeru-
lonephritis, pylonephritis, or hyperten-
sive renal disease.
It has been postulated that non-
vegetarians may be exposed to less
cadmium than vegetarians because of
the cadmium-poor meat in their diet.
The food animal is exposed to the same
water and plant sources as the human
and the cadmium accumulates in his
kidneys by similar physiologic mecha-
nisms as in the human. As a conse-
quence the skeletal muscle is depleted
of its cadmium. Steak is cadmium-poor;
sweet breads, especially kidneys and
liver, are not. Food derived from skeletal
muscle will dilute the cadmium ob-
tained from the cadmium-accumulating
food plants.
This study attempts to test this possi-
bility by measuring the cadmium expo-
sure and accumulation in a population
subgroup that includes many vegetari-
ans. The study group is drawn from
Southern California Seventh-Day Ad-
ventists (SDA) already enrolled in the
ongoing Adventist Health Study.
Seventh-Day Adventists are a small
Protestant denomination with about
three million members worldwide. Ap-
proximately 100,000 live in California.
Church standards require that members
abstain from the use of tobacco and al-
coholic beverages. Previous studies
have shown that over 98% of church
members conform to these standards.
The church also recommends other
practices affecting life style that are not
used as criteria for membership. A veg-
etarian diet is urged as the ideal. The
use of unclean meats as biblically de-
fined (pork, shellfish, etc.) is proscribed.
Use of caffeine-contained beverages
(coffee, tea, colas) and strong condi-
ments (such as black pepper) is discour-
aged. Whole grains, nuts, vegetables
and fruits as major dietary elements are
recommended. A food industry special-
izing in protein products derived from
plant sources has developed. Although
these dietary practices have been
taught for over 100 years, members
vary widely in actual practice. A very
few are pure vegetarians or vegans
(PV). About half are lacto-ovo-
vegetarians (LV), who use eggs and
dairy products but abstain from meat,
poultry and fish. The remaining non-
vegetarians (NV) use clean meats and
fish in their diet. About 17% drink one or
more cups of coffee per day. Epidemio-
logic studies among this group began a
quarter of a century ago in the content
of unhealthful effects of tobacco usage.
These led to the prospective Adventist
Health Study begun in the 1970's. Ap-
proximately half of the California Ad-
ventist population are enrolled. They
completed life style questionnaires and
respond periodically to health status in-
quiries. Hospital record and death cer-
tificate information supplements health
data obtained directly from the partici-
pants.
The public concern for identifying and
minimizing environmental health risks
mandates a monitoring of exposure
rates. Following index population
groups that have higher than average
exposure to cadmium may be useful in
providing warning of general popula-
tion health risk.
Participants
Live Subjects
From the computerized records of the
Adventist Health Study, 50 pairs con-
sisting of one SDA lacto-ovo-vegetarian
and one SDA non-vegetarian were gen-
erated. The members of each pair were
matched by age, sex, marital status, ed-
ucation, menopausal status, and occu-
pational group. Since cadmium accu-
mulates over the life span of the
individual at a biological half life of ap-
proximately 20 years, these subjects
were selected from the 40- to 60-year-
old age group. Another matching was
attempted to develop a similiar pairing
with 50 pure vegetarians. Because these
vegans are such a small fraction of
those enrolled in the Adventist Health
Study, the age requirements were re-
laxed to increase the possibility of re-
cruiting a complete group. These sub-
jects were identified only by a
transformed identification number. The
ID numbers of the subjects that
matched were passed back to a com-
puter routine that mailed letters ex-
plaining the study to subjects. Each sub-
ject was randomly assigned to one of
the two dieticians for the rest of the
study. The dietician was given a name,
address, and phone number. Any fur-
ther information the dietician received
about the subject came directly from the
subject. If a subject declined to partici-
pate, none of our staff learned anything
further about the subject, even whether
or not he was part of the Adventist
Health Study.
Five to seven days after the introduc-
tory letter was mailed, the dietician
called the participant. She briefly ex-
plained the study, answered any ques-
tions, and made an appointment for the
first home visit. If the subject agreed to
participate, a letter confirming the ap-
pointment and an introductory manual
briefly explaining specimen collection
was mailed. At the first interview, a writ-
ten consent was obtained and a detailed
explanation of the collection procedure
made both verbally and in the form of
an extensive instruction manual.
When a subject declined to partici-
pate, a computer routine was used in an
attempt to replace him. If he was a veg-
etarian subject the computer searched
the files for another match. If this was
unsuccessful, the dietician assigned to
contact the matching subject was in-
structed to cancel the appointment for
home visit with the matching subject,
and expressed our regrets. If another
match was found, and home visits with
the original subject had not been sched-
uled, they were postponed until a few
days after the new subject had received
the introductory letter.
Once the first home visit had been
completed, all subjects were carried
through the completion of sample col-
lection, regardless of withdrawals of
matched subjects. The first visit was
conducted by a research nutritionist.
Any questions at this point were an-
swered. Written consent was obtained
from the study subject. Then the subject
was asked to recall food intake in the
previous 24 hours, using models and
following a prod question sheet. The
purpose of the unannounced recall was
to determine if the subjects changed
their diets while completing the three-
day diary. Seventh-Day Adventists in
particular may have a number of ideas
about an appropriate diet which they
may not be following exactly. The
knowledge that their diet is being ana-
lyzed may cause them to consciously or
unconsciously change their eating.
A health history (exclusion criteria)
questionnaire was administered by the
nutritionist. This questionnaire included
inquiries about a number of characteris-
tics that were the basis for further exclu-
sions. These included osteomas, major
changes in diet in the last two years,
and use of a number of drugs, including
cancer chemotherapeutic agents and.
lipid lowering agents. The use of estro-'
-------
gens was recorded, but not used as
grounds for exclusion.
The subject was instructed in the col-
lection of urine and feces. A 2.5-liter
container and a 1-liter widemouthed
plastic jar in an airline bag were pro-
vided for 24-hour urine collected on the
second day of feces collection. This ex-
cluded the first voiding of that day and
included the first voiding of the follow-
ing day. The container provided con-
tained a preservative (HC1). The fecal
samples included every stool during
three days. Plastic bags were provided
to hang in the toilet, or a port-a-potty
made available if so desired. After defe-
cation, the bag was removed, deposited
in another plastic bag and the outer bag
closed, and placed in a freezer chest
packed with dry ice. This outer bag was
labeled with the subject's identification
number, the date, whether contami-
nated with urine, and a sequential num-
ber to identify the bowel movement.
Subjects were instructed to urinate be-
fore defecating to minimize urinary con-
tamination of the fecal sample. The
plastic bag was hung in the toilet in
such a way that there was some space
in front of the bag for males to urinate.
In case of females, a small amount of
urine might fall into the bag. There also
was space beind the bag to drop paper
into the toilet. A small checklist was
used to identify any stools with urinary
contamination. Employed subjects
were provided with another freezer
chest at their place of employment if
they felt it likely they would defecate at
work. This was labeled "Biological Ma-
terials for Scientific Research—Do Not
Disturb or Remove." Our staff contacted
the employer to deliver and pick up the
chest.
All subjects were asked to collect food
samples. All foods and beverages (ex-
cept water) consumed were weighed on
a gram scale. During the three days of
fecal collection, the subject was asked
to place about 2 ounces of food or bev-
erage other than water, in a small plas-
tic container labeled with the names of
the food. Dressings, sauces, etc., added
at the table were collected separately,
as these may not be mixed homoge-
neously on the plate with the foods with
which they are eaten. A number of small
containers, each containing 2 ounces of
a different food, could be placed in a
larger container labeled for the data and
the meal. The food samples were refrig-
erated, but not frozen. Vitamin pills and
other dietary supplements were col-
lected in exact duplication of the
amount used by the subject, and a
record of the nutritional supplement's
composition obtained by the dietician.
Each subject was given forms to
record his complete dietary intake dur-
ing the three days of fecal collection. He
was instructed to record all food and
beverages consumed, and to save
recipes of homemade foods.
The subject was given a day within
the following next week to begin his di-
etary record and sample collection.
Each collection period included one
weekend day; that is, it was Thursday,
Friday, and Saturday, or Sunday, Mon-
day, and Tuesday.
The 24-hour recall was coded by the
dietician for the dietary analysis the
same day if possible.
On the second day of the collection
period, the same dietician phoned the
subject to review the procedures and
answer any questions. This allowed dif-
ficulties to be handled early. Specific
questions were asked to ascertain com-
prehension and compliance.
On the workday following the sample
collection (that is, a Monday or a
Wednesday), the dietician retrieved the
freezer chest containing the fecal and
urine samples, another containing the
food samples, and when used, a third
containing samples from the subject's
place of employment.
The dietician reviewed the three-day
diary with the subject and edited the
diary as necessary.
The subject's height, weight, triceps
skinfold and blood pressure were mea-
sured.
The dietician checked each food cup
for sample adequacy and completeness
and reviewed the urine and fecal sam-
ples collected for completeness.
The same day the dietary diary was
coded whenever possible.
The dietary interviewer made an ap-
pointment for a venipuncturist to draw
blood, and explain the 12-hourfast. This
appointment was made as soon as pos-
sible after the period of food collection,
within no more than seven days. It was
not made during the food collection pe-
riod, or on the day following, to prevent
the fast from interrupting the subject's
usual eating habits.
The venipuncturist visited the sub-
ject's home, usually early in the morn-
ing to draw the blood samples for the
biochemical profiling. Those subjects
living within easy driving distance from
the medical center were scheduled to
have the blood work drawn in the out-
patient venipuncture station.
Autopsy Subjects
Kidney, liver, pancreas and hair sam-
ples collected by cooperating patholo-
gists in Southern California Seventh-
Day Adventist hospitals were obtained
from subjects coming to autopsy. Life
style and health data were obtained
from hospital charts, next-of-kin, and
from the Adventist Health Study data
base.
Conclusions
1. No statistical difference in cadmium
exposure rate is demonstrated be-
tween the non-vegetarians and lacto-
ovo-vegetarians of this study.
2. The cadmium exposure rate of the
small group of pure vegetarians
studied is in the direction anticipated
(higher than non-vegetarians), but is
not large enough to achieve statisti-
cal significance.
3. Cadmium accumulation in the cntial
organ (kidney) of subjects coming to
autopsy is well below the range as-
sociated with renal injury.
4. Quantitative methods for measuring
total urine protein are not as satisfac-
tory for assessing tubular proteinuna
as those developed to measure the
small proteins that normally appear
in the glomerula filtrate.
Discussion
Anticipated differences in cadmium
exposure rates between vegetarians
and non-vegetarians are dependent
upon the replacement of cadmium-rich
calories from cadmium-accumulating
foods by cadmium-poor calories. For
this difference to be apparent, the food
of the vegetarian must contain more
cadmium than that of the non-
vegetarian. The ideal situation for dem-
onstrating the difference is in the case
of a single major food source for each
group. This was the case for the Japa-
nese eating rice grown in cadmium-
contaminated water. But for the situa-
tion where there are a variety of foods,
the difference may be masked. The
non-vegetarian may be getting higher
than background exposure by including
liver, kidney and shellfish in this diet.
The vegetarian who obtains the major
portion of his food from the national
food distribution system averages his
exposure by the mixing of foods grown
in cadmium-poor districts with those of
contaminated areas. Another difficulty
in demonstrating differences is in the
classification of the subjects. In the
strictest sense there are no pure vege-
3
-------
tarians unless they grow all the food
they eat. Processed foods usually in-
clude milk or egg solids and other
non-vegetable additives and anyone de-
pending upon the prepared food indus-
try is getting more than he realizes. Self
assignment to pure vegetarian, lacto-
ovo-vegetarian or non-vegetarian
groups may represent the person's
ideal for his dietary practice, rather than
his invariant practice. Of those pure
vegetarians in the Adventist Health.
Study we contacted, most used some
dairy products occasionally. Even the
non-vegetarian group may be subject to
cultural influence from its Adventist
background to eat less meat than the
general population does.
The lack of difference between the
non-vegetarian group and the lacto-
ovo-vegetarian group in this study is
probably due to several factors. In spite
of attempts to control confounding fac-
tors by matching subjects by age, sex,
occupational group, etc., there still re-
main many differences (genetic, geo-
graphic origins, past changes in life-
style) to mask diet-induced differences.
Both groups are subject to environmen-
tal contamination from tobacco smoke
and other erratic sources. Both replace a
portion of the higher cadmium content
plant foods in their diet by cadmium-
poor foods—dairy products for the veg-
etarians and skeletal muscle meat cuts
for the non-vegetarians. It is unfortu-
nate that we could not recruit a larger
group of pure vegetarians to reduce the
dietary overlap.
The subjects in the autopsy study are
not easily categorized into well-defined
dietary groups. A few were lifelong
lacto-ovo-vegetarians but many be-
came Adventists later in life and had
made significant life style and dietary
changes. Several had been heavy
cigarette smokers. Thus, the probable
rate of exposure had varied significantly
at different times of their lives. Even so,
the total renal cortex accumulation re-
mained below the risk threshold of 200
micrograms cadmium per gram of wet
tissue.
The concentration of cadmium in
flour used in making meat analogues is
in the range reported by other studies.
The highest value was from a sample
obtained from a partially open flour bin
in a market where it may have been con-
taminated by the environment.
Urine cadmium concentration was at
the low end of the working curve of the
instrument. Its measurement was facili-
tated by the concentration achieved by
the extraction step.
. Urine protein methods are tuned to
measure albumin. In the typical clinical
setting, the physician is most interested
in proteinuria as an indicator of
glomerular dysfunction. Since the
major protein appearing in the urine as
a consequence of glomerula injury is
albumin, such methods meet the clini-
cal need. They vary widely in their re-
sponse to other proteins. The smaller
proteins normally pass through the
glomerular basement membrane, but
are removed form the glomerular fil-
trate by the tubular epithelium. In the
setting of tubular dysfunction or injury
these small proteins replace albumin as
the constituents of greatest clinical in-
terest, but most methods grossly under-
estimate or fail to detect them. In our
study, this methodological shortcoming
is evidenced by the lack of correlation
between beta-2-microglobulin excre-
tion and total protein excretion. Tubular
proteinuria must be looked for specifi-
cally by urine protein electrophoresis or
by specific assay of one or more of the
small protein species.
R. E. Willard is with Loma Linda University Medical Center, Loma Linda, CA
92354.
Norman Kowal is the EPA Project Officer (see below).
The complete report, entitled "Assessment of Cadmium Exposure and Toxicity
Risk in an American Vegetarian Population," {Order No. PB 85-211 340/AS;
Cost: $10.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S1-85/009
•frU.S. GOVERNMENT PRINTING OFFICE:1985—559-016/2
------- |