By Chris
Masterjohn
Presented with permission from
the Weston A. Price Foundation, www.westonaprice.org
where this article originally appeared.
Table of Contents
The research of Dr. Weston
A. Price, documented in his classic volume Nutrition
and Physical Degeneration, demonstrated the
absolute necessity of certain fatty animal foods
for good health. However, a challenging argument
against eating animal foods--especially animal fat--arises
from vegetarian circles. This argument focuses on
a class of chemicals called dioxins, and suggests
that in the modern world, overburdened by pollutants,
these fat-soluble chemicals accumulate specifically
in the fatty tissue of animal products, making a
vegetarian--even vegan--diet a necessity for those
living in the modern world.
For example, one vegetarian website argues that
"nearly 95 percent of our dioxin exposure comes
in the concentrated form of red meat, fish, and
dairy products, because when we eat animal products,
the dioxin that animals have built up in their bodies
is absorbed into our own," and that eating
dioxin-laced animal products will make us vulnerable
to "a wide range of effects, including cancer,
depressed immune response, nervous system disorders,
miscarriages, and birth deformities."1
The same argument appears in environmentalist
circles as well. For example, the Pennsylvania-based
environmental organization ActionPA's "Dioxin
Homepage" argues that "[t]he best way
to avoid dioxin exposure is to reduce or eliminate
your consumption of meat and dairy products by adopting
a vegan diet."2
Thus, this argument for vegetarianism essentially
builds on a series of three points:
- Dioxins are potent human carcinogens, endocrine
disruptors, reproductive disruptors and immune
disruptors;
- Animal products are uniquely high in dioxins;
- Avoiding the harmful effects of dioxins is
primarily dependent upon minimizing dioxin intake,
and therefore avoiding animal products.
The assertion that dioxins accumulate specifically
in animal products is simplistic and inaccurate,
and in fact a diet rich in pastured animal products
provides protective nutrients, especially vitamin
A, that directly oppose the toxic actions of dioxins
in animal experiments, while a diet rich in most
plant fats provides compounds that enhance the actions
of dioxin. The argument that we should avoid animal
products because of their dioxin concentration is
thus no less flawed than the argument that we should
avoid animal products because they contain saturated
fat and cholesterol.
Dioxins:
Some Background
The prototypical dioxin compound is 2,3,7,8
tetrachlorodibenzo-p-dioxin, abbreviated
as "TCDD." The word "dioxin,"
however, refers more broadly to dioxin-like
compounds from three classes: polychlorinated
dibenzo-p-dioxins (PCDDs, including
TCDD), polychlorinated dibenzofurans (PCDFs),
and polychlorinated biphenyls (PCBs). Not
all PCDDs, PCDFs, and PCBs are considered
dioxins. Only 17 out of 210 PCDDs and PCDFs
are considered dioxin-like, and only 11 out
of 209 PCBs are considered dioxin-like. The
precise positioning pattern of chlorine atoms
on the molecule determines whether or not
it is dioxin-like, and it is important not
to confuse the PCBs classified as dioxins
with other PCBs that are believed to be toxic
through non-dioxin-like mechanisms.3
The relative toxicity of dioxins is expressed
in relation to the toxicity of TCDD, the most
potent dioxin. A "toxicity equivalency
factor" (TEF) relates the degree of toxicity
of a specific PCDD, PCDF or PCB to the toxicity
of the prototypical TCDD, and the TEF is then
multiplied by the number of molecules of that
particular dioxin compound in a food to yield
a "toxicity equivalent quantity"
(TEQ). The sum of TEQs from all dioxin compounds
within a given foodstuff estimates the presumed
degree of toxicity contained within that foodstuff.3
A higher amount of TEQs doesn't necessarily
mean that there is a greater absolute quantity
of dioxins in the food, since the TEQ gives
greater weight to the more potent dioxins.
So, a food with a smaller total amount of
dioxins but a more potent specific compound
could have a higher TEQ value than a food
with a higher quantity of dioxins but less
potent specific types of dioxins. However,
the TEQ is not necessarily an indicator of
how toxic the food is, simply because some
dioxins, such as PCBs, also have toxicity
that is non-dioxin-like. |
Exposure to Dioxins
Although 95 percent of human exposure to dioxins
is believed to come from food,3 this
fact deceptively overestimates the impact of foodbased
dioxins, because industrially exposed populations
have been exposed to 10-1000 times higher concentrations
of dioxin than the general population.4
And even at these high exposures the evidence of
dioxin-induced harm is inconclusive at best.
Since the 1970s, after an historical peak in the
1950s and 1960s, sources of dioxins released into
the environment have changed, and the levels have
dramatically declined,4 due to government
regulations and to the advancement of technology.
The US and other countries have banned the use of
pesticides and herbicides such as 2,4,5-trichlorophenoxyacetic
acid and hexachlorophene, the production of which
was once a primary source of dioxin contamination.
Alternatives to the bleaching of paper with free
chlorine have further reduced or eliminated dioxin
production. The dioxin contribution of municipal
and medical waste incineration has decreased by
over 90 percent because of technological advances
in waste disposal.5
Open barrel burning of trash is now the primary
source of dioxin released through human agency,
while modern incinerators make a comparatively negligible
contribution. Certain metal refining processes also
lead to dioxin generation. The other major contributors
are natural, including volcanoes5 and
forest fires.6
Human body burdens of TCDD, the most potent dioxin,
in the US have decreased 10-fold, and total dioxin
TEQs have decreased 4-fold to 5-fold between 1972
and 1999. Given the typical half-life of dioxins
in the body, this means that dioxin exposure during
this period has decreased by a full 95 percent!6
Similar observations have been found in other countries.
For example, dioxin concentration in the breast
milk of Japanese mothers declined by 87 percent
between 1974 and 1998.7 Dioxin intake
declined about 90 percent in the Netherlands between
1978 and 1999,8 and in Finland dioxin
exposure declined 50 percent over the course of
the 1990s alone.9
We will never know exactly what level of dioxins
Price's healthy primitives or other premodern societies
were exposed to. However, since natural sources
of dioxins like volcanoes and, more significantly,
forest fires, are now primary sources of dioxins,
and since pre-modern populations would be expected
to have additional exposure through the direct inhalation
of fumes from the incineration of heating and cooking
materials (living, for example, in thatched houses
without chimneys, as Price described the primitive
Gaelics), as well as the use of incinerated materials
as soil fertilizer (such as slash-and-burn techniques
or the use of smoke-impregnated thatch as a fertilizer,
both described by Dr. Price), it is not unreasonable
to conclude that we are now approaching a level
of dioxin exposure similar to that of pre-industrial
populations.
Even by conservative estimates, no one in the
US is currently consuming a level of dioxins that
would be expected to exert physiological harm. The
World Health Organization (WHO) developed what is
called a "tolerable daily intake" (TDI)
for dioxins based on the intake levels that produce
decreased sperm count, immune suppression and genital
malformations in the offspring of exposed rats,
and neurobehavioral effects and endometriosis in
the offspring of exposed monkeys.4 However,
since the WHO's TDI is supposed to assume the greatest
degree of sensitivity, in order to yield the safest
and most conservative estimate, the harm done to
male rats exposed during gestation is the primary
basis for the TDI.6
Using this estimate, taken from the most sensitive
individual rats, the WHO then added a "safety
factor" of 10 to yield a TDI of 2 picograms
(pg) TEQ per kilogram of body weight.4
(A picogram is a trillionth of a gram or a billionth
of a milligram.) This means that a human whose intake
of dioxins meets the WHO's TDI is consuming only
one-tenth of the concentration required to yield,
after a lifetime of exposure, body burdens with
concentrations that were required to produce the
minimum physiological effect not in the most sensitive
adult or child rat, but in the most sensitive rat
during gestation, the critical period where
a developing organism would be much more sensitive
than at any other time.
According to a 2005 study covering the years 1999
through 2002, only 1 percent of 2-year-olds in the
United States exceeded the TDI in 1999 and 2000,
and this excess of the TDI was very small. The risk
to children is probably overestimated since the
TDI is based on body weight alone and does not take
into account the fact that children have higher
fecal excretion rates of dioxin, nor does it take
into account the fact that, since we are experiencing
a decline in dioxin exposure, current exposures
will overestimate the cumulative body burden that
will be reached over time. In 2001 and 2002, no
intakes at any age in the US were estimated
to exceed the TDI.6
A
Modern Threat?
Dioxins are not merely a modern industrial
phenomenon. Chlorinated organic compounds
are produced naturally, by biological and
abiotic means, have been found in coal samples
dating back 300 million years, and are produced
by cyanobacteria, which have existed for billions
of years.a
There are 4,519 known naturally ocurring
organohalogens, 2,320 of which are organochlorines.
Bleach, chlorine gas, and organochlorines
are naturally produced in the human body.
Brominated dioxins are produced biologically
by sponges as a defense mechanism, while chlorinated
dioxins are naturally produced by the decay
of plant matter in peat bogs, the incineration
of wood in forest fires, or in gases released
from volcanos.a
The smoke of fires--to which primitive peoples,
unlike moderns, were exposed on a daily basis--contains
between 10 and 40 nanograms of chlorinated
dioxins per gram of smoke.b A single
gram of smoke thus contains between 125 and
500 times the amount of dioxin that a 80 kg
adult consumes from food per day.
Wood naturally contains chloride compounds
that are oxidized under high heat, producing
chlorine that readily reacts with organic
compounds to form organochlorines, including
dioxins. Although chlorine released into the
atmosphere by industry makes a very small
contribution to the chlorine available for
this reaction, of the 4 million tons of methyl
chloride --the most abundant atmospheric form
of chlorine--produced each year, only 10,000
tons originate from industry.a
Therefore the healthy pre-modern groups that
Price studied, who thrived on diets rich in
animal products, probably consumed some level
of dioxins in their food, possibly rivaling
our own consumption. What has changed in the
modern era is not the introduction of chemical
pollutants, but the disappearance of protective
factors abundant in traditional diets -- which
have protected us from pollutants throughout
history--from the modern menu.
- Gribble, Gordon W., "Amazing Organohalogens,"
American Scientist Online, Vol
92 No.621 (2004) 342-349
- US EPA, "A Summary of the Emissions
Characterizations and Noncancer Respiratory
Effects of Wood Smoke." EPA-453/R-93-036,
as cited in Citizens for Safe Water Around
Badger, "Fact Sheet: Open Burning at
Ravenna Arsenal," http://www.cswab.org/ravenna.html
Accessed October 2, 2005.
|
Dioxin and Cancer:
"Sufficient Evidence" Not Required
Although the World Health Organization's (WHO)
International Agency for Research on Cancer (IARC)
designated TCDD (but not the other dioxins) as carcinogenic
to humans (Group 1) in 1997, TCDD does not
actually pass the test for carcinogenicity.
For decades, sufficient evidence of carcinogenicity
to humans was a necessary criterion for classification
of a substance as a Group 1 carcinogen. TCDD was
the second chemical whose classification utilized
the IARC's 1990 change of criteria, by which a substance
could be judged carcinogenic to humans even if ".
. . evidence in humans is less than sufficient
but there is sufficient evidence of carcinogenicity
in experimental animals and strong evidence in exposed
humans that the agent . . . acts through a relevant
mechanism of carcinogenicity." [emphasis
added]10
Dioxins do not initiate the transformation of
a normal cell to a cancerous cell in any species.
TCDD has been shown, however, to be a very powerful
promoter of cancers that are first initiated
by another carcinogen, and is thus considered a
"non-genotoxic carcinogen." For example,
one study using mouse fibroblasts in cell culture
found TCDD to enhance the carcinogenic effect of
N-methyl-N'-nitro-N-nitrosoguanidine
over 3-fold, and to enhance the carcinogenic effect
of 3-methylcholanthrene almost 4-fold. Yet in the
absence of these carcinogens, TCDD could not initiate
cancer foci even at doses 1000 times higher than
those used for its promoting effect and 1000 times
higher than the dose of the genotoxic carcinogens
used for initiation of cancer.11
The cancer-promoting effects of dioxin are not
consistent across species or across tissues. In
fact TCDD has been used to inhibit estrogen-dependent
breast cancer in rodent models and cultured human
cells, leading researchers to look into development
of anti-cancer drugs from this dioxin.12
The fact remains, however, that TCDD can be a powerful
cancer-promoter in certain tissues of certain species.
The mechanism by which TCDD exerts its toxic effects
is believed to be mediated by its binding to the
aryl hydrocarbon receptor (AhR), a receptor that
is also involved in mediating responses to polynuclear
aromatic hydrocarbons, combustion products and numerous
phytochemicals such as flavanoids and indole-3-carbinol.
Once bound, the TCDD-AhR complex then moves into
the nucleus, where it binds to the aryl hydrocarbon
receptor nuclear translocator (Arnt) protein. Finally,
this TCDD-AhR-Arnt complex then binds to DNA to
induce the expression of the cytochrome P-450 1A1
gene.5 Less is known about precisely
how this activation of the cytochrome P-450 system
leads to toxicity and carcinogenesis, but the toxic
effects of TCDD are usually correlated with its
activation of this system and appear to be dependent
upon it.
The WHO's argument for TCDD's inclusion as a Group
1 carcinogen despite its failure to fulfill the
criterion of sufficient evidence of carcinogenicity
in humans relies on the following reasoning:
- TCDD is a multi-site carcinogen in animals
acting through the AhR;
- The AhR is highly conserved across species
and acts in a similar way in humans as it does
in animals;
- Tissue concentrations in humans where epidemiological
studies have observed increased risk of cancer
are similar to those of rats exposed to carcinogenic
doses in the laboratory.13
Yet, although the AhR is highly conserved across
species, the carcinogenicity and toxicity of TCDD
is not. For example, the lethal dose of TCDD varies
5000-fold across species.5 Activation
of the AhR cannot be sufficient to induce cancer,
because the effect of indole-3-carbinol, a substance
found in cruciferous vegetables, is also mediated
by the AhR, yet is used to inhibit cancer.14
TCDD's use of inhibiting estrogen-dependent
breast cancer in rodent and human mammary cells
is also mediated through the AhR.12
In a major review on dioxins published in 2003,
Phillip Cole and his co-workers point out that TCDD
acts as a carcinogen in certain tissues in certain
species, not many tissues in one species. The variation
across species as to which tissues are vulnerable
to carcinogenicity is not an argument for multi-site
carcinogenicity in humans, but an argument against
generalizing from species to species. The types
of cancer induced in animals by TCDD "bear
little correspondence to those reported among humans
exposed to TCDD," and, while the tissue concentrations
of TCDD are similar in animals who develop cancer
and humans observed to have an increased risk of
cancer, "even the few positive epidemiological
studies of TCDD-exposed populations generally report
at most a minimal increase of total cancer, while
in rats the increase is much greater."13
The fact that activation of the AhR is not consistently
linked to cancer, that the response of animals varies
widely to TCDD, and that the types of cancer and
magnitude of increased risk observed in humans bear
little resemblance to the types of cancer and magnitude
of increased risk in rats is sufficient to refute
the WHO's inclusion of TCDD as a Group 1 carcinogen
by its own criteria.
A question remains: Do humans exposed to high
concentrations of dioxins really exhibit an increased
risk for cancer? The Cole review refutes this hypothesis,
showing that the WHO used its evidence selectively,
and that researchers failed to appropriately adjust
for exposure to other carcinogens.13
The WHO's argument rests on epidemiological evidence
from industrial and occupational exposure, populations
that have been exposed to 10-1000 times the concentrations
of TCDD compared to the general population.4
While admitting the absence of a strong case for
the elevation of any specific cancer, they have
compiled four major cohort studies to find a 40
percent increased risk for all cancers combined
for "highly exposed" workers, the definition
of which differed between studies.
Yet the WHO excluded from this compilation a study
by Kogevinas that the very same monograph referred
to as " . . . the largest overall cohort study
of [TCDD]-exposed workers," and which included
the data from the other four cohorts. The WHO argued
that it had to be excluded because it included individuals
with lower TCDD exposures; but, as Cole and his
colleagues point out, the data were reported separately
for those who were "highly exposed," and
those with lower exposure. Therefore the Kogevinas
study could--and should--have been included. The
Kogevinas study found a 20 percent increased risk
for all cancers with occupational dioxin exposure,
but those who were most highly exposed (20 or more
years of work experience) had an 8 percent decreased
risk of all cancers.13
The Seveso cohort study described the highest
exposure to TCDD ever documented in a population.
Seveso was the site of a 1976 accident at a chemical
manufacturing plant in which a dense cloud of TCDD
was released from a reactor in quantities measured
in kilograms over 10 square miles, necessitating
the evacuation of 600 homes.15 Yet follow-up
of the Seveso population reveals that "all-cause
and all-cancer mortality did not differ significantly
from those expected in any of the contaminated zones."13
Cole and his team also noted that in numerous
studies, confounding factors were not taken into
account:
- The NIOSH cohort study used smoking information
from industrial plants 1 and 2, where there was
no lung cancer elevation, but did not record smoking
data from plants 8 and 10, where lung cancer was
elevated, which they attributed to dioxin exposure.
- In the same study, two deaths from mesothelioma
could have reflected exposure to asbestos, and
workers were also exposed to the bladder carcinogen
4-amino-biphenyl, neither of which was taken into
account, while cancers in those exposed to these
known carcinogens were attributed to dioxin.
- In the Zober study, 35 of 37 cancer cases were
smokers, and 10 of 11 respiratory cancer cases
were smokers, yet cancer risk was assumed attributable
to TCDD.
- The one study (Ott and Zober) "with even
minimally adequate information on smoking"
found no statistically significant relationship
between respiratory cancer incidence or mortality
and TCDD.
- No attention was given to possible confounding
of socioeconomic class, "even though the
individuals most exposed to TCDD frequently are
from the less privileged socio-economic groups
that have high overall mortality, including mortality
from all cancer."13
All of the epidemiologic studies published before
1997 that were not included in the WHO's IARC Monograph,
found no association between TCDD exposure and increased
risk for cancer or mortality, including those by
Dalager and team concerning non-Hodgkin's lymphoma,
Watanabe and team concerning overall mortality,
Bullman and team concerning testicular cancer, and
Dalager and team concerning Hodgkin's disease.13
After the WHO's IARC classified TCDD as a Group
1 carcinogen in 1997, subsequent reviews and studies
began to rely on the IARC's interpretation of earlier
study results, rather than the study results themselves.
Of the follow-up studies since 1997, the data of
Steenland and team show that longer follow-up decreased
the magnitude of associations previously found in
the same cohort, and caused loss of statistical
significance; Pesatori and team found that non-cancer
mortality in Seveso--where the highest exposure
to dioxins ever documented occurred--did not differ
from that of the general population, and Ketchum
and team found 30 percent fewer deaths
from cancer in US Air Force veterans who were highly
exposed to dioxins.13
Thus, while TCDD is claimed to be a non-genotoxic
multi-site carcinogen, the evidence suggests that
the wide variation in responses to dioxins across
species prevents generalization to humans, and that
the failure of dioxin exposure to act as an independent
risk factor for cancer, even in human populations
exposed to concentrations 1000 times greater than
the general population, would contradict claims
of human carcinogenicity.
Dioxins
in Pastured Animal Products?
A review published in 1995 suggested that
pastured animal products would probably contain
higher dioxin concentrations because of a
higher rate of soil ingestion;3 however,
newer research has revealed the fact that
the primary sources of above-average dioxin
concentration in beef samples are feeding
troughs constructed with pentachlorophenol-treated
wood and the inclusion of incinerator waste
as a feed additive.6 Grass-fed
beef is not exposed to these sources of dioxins. |
Non-Cancer Effects of
Dioxins
Dioxins are responsible for a wide range of different
toxic effects in different species. Non-cancer effects
observed in wildlife exposed to high concentrations
of dioxins, experimentally induced in animals treated
with dioxins, and observed in humans exposed to
industrial concentrations of dioxins vary between
species and between types of exposure. Like dioxins'
carcinogenic effects, the non-cancer effects of
dioxins are believed to be primarily mediated by
their ability to bind to the aryl hydrocarbon receptor
(AhR).5
Seals fed dioxin-contaminated fish had depressed
blood levels of vitamin A and thyroid hormone, and
depressed natural killer cell and T-cell activity
(indicating immune suppression). Herring gulls have
been found with decreased liver stores of vitamin
A, but increased egg yolk vitamin A levels, when
exposed to dioxins, while great blue herons have
lower levels of vitamin A in their egg yolks. Exposed
cormorants experience decreased levels of free thyroid
hormone, herring gulls experience decreased vitamin
A, and common terns experience both decreased vitamin
A and thyroid hormone levels. In white suckerfish,
the AhR-mediated dioxin-like activity of PCBs was
associated with birth defects. Skin diseases (resembling
vitamin A-deficiency skin diseases) and increased
thyroid weight have been observed in response to
organochlorines, which include, but are not limited
to dioxins.16
One interesting experiment, demonstrating the variation
of dioxin toxicity between species, fed goiter-bearing
salmon exposed to high concentrations of dioxin-like
and non-dioxin-like PCBs in the wild to rats and
other salmon. Although every single one of the wild
salmon (previously transferred from the Pacific
to the Great Lakes) in the organochlorine-polluted
Great Lakes had an enlarged thyroid gland and a
high PCB body burden, the degree of thyroid enlargement
had no relation to PCB burden. When these PCB-laden
fish were fed to immature Coho salmon, the latter
did not develop any thyroid enlargement. Yet, when
the PCB-laden fish were fed to lab rodents, the
rodents developed goiter in direct proportion to
the dioxin-like activity induced by the dietary
PCBs. It has been hypothesized that the reason the
Great Lakes salmon developed goiter is because of
a goitrogenic factor of bacterial metabolism in
the Great Lakes, which has also proved goitrogenic
to humans, while to rodents, on the other hand,
the PCBs carried by the fish are the goitrogenic
factor.16
Reduced female fertility and reduced male sperm
production, as well as genital deformations, have
been induced by dioxin exposure in rodents. Dioxins
can cause calcium uptake in neurons of the rat hippocampus,
and have species- specific effects on gene expression
in the nervous system of zebrafish and rats, though
it is unknown how these effects may or may not result
in any type of neurotoxicity.5 TCDD,
the most potent dioxin, acts as an anti-estrogen
in rodent mammary and uterine tissues, as well as
human mammary cells, where it exerts anti-carcinogenic
effects.17
TCDD can induce wasting syndrome and death in
chickens and rodents, though its lethal dose varies
5000-fold across species. TCDD can induce cleft
palate and other deformities, reproductive failure
and liver damage in birds,18 endometriosis
in rhesus monkeys, growth of surgically induced
endometriotic cysts in rats and mice, and various
effects on metabolism and hormones in various species.4
In humans, the only conditions to which dioxins
have been conclusively linked are a type of skin
acne known as chloracne, and a temporary increase
in liver enzymes. Other non-cancer phenomena have
been associated with exposure to industrial concentrations
of dioxins in some human epidemiological studies,
but the evidence is inconclusive or contradicting.4
Effects on various thyroid-related hormones and
proteins were found in the Ranch Hand cohort and
the National Institute for Occupational Safety and
Health (NIOSH) cohort, but they were mostly weak
and non-significant, and not consistent between
studies. In one sub-section of the NIOSH cohort,
diabetes was not associated with TCDD, but the highest-exposed
group did have the highest rate of diabetes. The
NIOSH cohort as a whole found a negative correlation
between TCDD exposure and diabetes mortality, while
women, but not men, in zones A and B in Seveso had
a greater risk of diabetes mortality with greater
TCDD-exposure.4
Exposure to dioxins in breast milk was associated
with tooth enamel defects in one study, and alterations
in thyroid hormone levels have been associated with
prenatal dioxin exposure in children. In one study,
exposure to dioxins in breast milk was found to
have no effect on psychomotor outcome of infants
between three and seven months or after 18 months,
but was associated with depressed psychomotor skills
between seven and 18 months. Few studies have identified
statistically significant effects of industrial-level
dioxin exposure on spontaneous abortions, birth
weight,or birth defects. However, the most TCDD-contaminated
area of Seveso, which has the highest exposure of
a population to TCDD ever documented, found a modification
in the sex ratio in favor of females to be associated
with the total dioxin exposure of both parents.
This is an interesting finding, but there were only
13 couples and 15 children in this group, and the
association was only found in the highest-exposed
group and between the years 1977 and 1984.4
The failure of dioxins to be consistently and conclusively
correlated with cancer in humans, or non-cancer
effects in humans with the exception of chloracne
and temporary increases in liver enzymes, even at
industrial levels that exceed what the general population
encounters by up to a factor of 1000, should give
pause to those who advocate exchanging the proven
health-promoting diets of our ancestors for modern
vegetarian and vegan diets that do not provide the
same type of nutrition. Although dioxins can experimentally
induce a variety of endocrine-disrupting, immune-depressing
or anti-reproductive effects in animals, the effects
are generally species-specific and in a minority
of cases--such as the anti-estrogenic effects in
mammary and uterine tissues--apparently beneficial.
Even if we assume that the worst of these findings
can be generalized to humans, the fact that dioxin
exposure has declined 95 percent since the 1970s
and continues to decline, and the fact that no one
in the US is currently exposed to even one-tenth
of the dosage that has produced an abnormality in
the most sensitive gestational rat, should assure
us of the safety of consuming animal products. Moreover,
dioxins do not act in a vacuum, but their effects
are subject to the influence of many other physiological
and dietary factors, and it is a diet rich in traditionally
valued animal products that offers the most protection
against their effects.
Dioxins: It's Not Just about
the Meat
The first leg of the dioxin-based argument for
a vegetarian or vegan diet--that dioxins are potent
human carcinogens, endocrine disruptors, reproductive
disruptors, and immune disruptors--has been shown
above to be either false or irrelevant at the level
of dioxins currently consumed in the US. The second
leg of the argument--that animal products are uniquely
high in dioxins--likewise fails to sustain analysis.
While the most contaminated foods in some studies
have been animal products, other studies cite animal
products as among the least contaminated foods.
Variation between samples is usually much greater
than any variation between animal and vegetable
products, making any supposed trend inconsistent
at best.
One 2003 study actually measured the intake of
dioxins in humans, where fourteen subjects ate an
omnivorous diet for two weeks, then ate a vegan
diet for two weeks. Although exposure to dioxins
on a TEQ basis was higher during the omnivorous
phase than the vegan phase, the diets of some subjects
were actually comparable in total PCBs in the vegan
and omnivorous phases. The TEQ measurement weights
the relative dioxin-like toxicity of each dioxin-like
compound against the toxicity of TCDD, so that compounds
with less dioxin-like activity (meaning less toxicity
mediated by the aryl hydrocarbon receptor) will
contribute less per weight to the total TEQ. However,
since PCBs have non-dioxin-like toxicity, and since
dioxin-like PCBs measured in the study could be
indicators for the presence of non-dioxin-like PCBs,
it's possible that total PCB-related toxicity of
the vegan diets could have been comparable to that
of the omnivorous diets. Most significantly, even
on the higher-TEQ omnivorous diet, average TEQ intake
was 1.09 pg/kg bodyweight, which is only about half
of the WHO's tolerable daily intake (TDI), a hyperconservative
estimate of toxicity risk.19
A 1995 review of the significance of animal products
as sources of human exposure to dioxins claimed
that the "major food sources [of dioxins] seem
to be fat-containing animal products and some seafoods."3
Since data from the US was not available at the
time, the authors used data from Germany and the
Netherlands. Table 2 of this review, showing the
contribution of selected food products in pg TEQ
per day, shows no consistent role of animal products
in exposure to dioxins. For example, in the Netherlands,
leafy vegetables (4.4) contributed a quantity of
TEQs roughly equivalent to pork (4.2), and poultry
and eggs (4.8). In Germany, vegetable oils (3.8)
contributed only half as many TEQs as pork (7.6),
and only 20 percent as many as beef and veal (19),
while in the Netherlands, vegetable oils (14) contributed
3.3 times as many TEQs as pork (4.2) and 7 percent
more than beef (13). Not only did the contribution
of TEQs in the same type of food vary widely between
the two countries, but the relative contribution
of animal and vegetable products also varied widely.3
More recently, data from a wider range of countries
has become available, and the FDA provides data
from the years 2000-2003 on its website. These results
continually reaffirm the wide variation between
regions and samples and the lack of any consistent
trend between animal and vegetable products.
For example, in Finland, fish accounted for 94
percent of dioxin TEQ intake,9 while
in Canada fish only accounted for 3 percent of TEQ
intake.20 A Japanese study found fish
intake to be an independent predictor of blood dioxin
levels, while for all other animal products except
pork the correlations were insignificant. Eggs,
butter, cheese and pork were actually negatively
associated with dioxin levels. Yet the variation
of blood levels between regions was large, ranging
from 13 TEQ per gram of lipid to 21 TEQ per gram
of lipid. Despite the higher fish intake in Japan,
the median blood levels of dioxin were still lower
than those found in other industrialized countries,
especially from earlier studies, probably reflecting
both inter-regional differences and declining dioxin
levels in the environment.21
Clearly, however, it would be more beneficial to
look at the quantities in food rather than the contribution
of foods to intake, since the former is independent
of what the general population is eating. What we
are interested in is whether a diet rich in traditional
animal products is excessive in dioxins compared
to a vegetarian diet, not which foods people eating
a standard diet are getting their dioxins from.
Unfortunately, most of the studies available have
relatively small sample sizes which, when combined
with the large variance between the samples, make
the data relatively useless for establishing trends
among types of foods. For example, in Greece, five
samples of fish oil varied by a factor of six (meaning
the sample with the highest concentration had six
times more dioxins than the sample with the lowest
concentration); five samples of butter and seven
samples of farmed fish both varied by a factor of
five; eight samples of lamb varied by a factor of
three; three samples of poultry, three samples of
beef liver and three samples of rice all varied
by a factor of two; four samples of wild fish had
concentrations ranging from zero to amounts nearly
approaching the median for farmed fish.22
The FDA's data for the United States only uses
three samples for each item.23 Although
the report does not give information indicating
the variation between samples, we can infer that
there is considerable variation between samples
by comparing different specific food items within
the same type of food. For example, expressed in
pg TEQ per gram, ground beef contains 0.0425 while
a fast food quarter-pound burger contains 0.0, and
whole milk contains 0.0087 while half and half,
which should concentrate the dioxins in the butterfat,
contains 0.0.24
The FDA's report offers us another way to ascertain
the type of variation found among samples because
the FDA reports the data in three ways: the first,
where a non-detect is assumed to be equal to zero;
the second, where a non-detect is assumed to be
equal to half the limit of detection; and a third,
where a non-detect is assumed to be equal to the
limit of detection. If there weren't any non-detects,
we would expect all three figures to be the same.
If there is one or more non-detect, we should expect
a certain pattern: the second figure should be higher
than the first, and the difference between the third
and the first figures should be exactly double the
difference between the second and the first figures.
In fact, this pattern is nearly ubiquitous among
the items that showed any detectable contamination
with dioxins. This fact allows us to conclude that
some of the items contained at least one sample
with no detected dioxin, including milk, beef, lamb,
turkey, beef liver, butter and salmon.
The FDA's data show that some of the highest concentration
of dioxin TEQs are found in animal products, but
this finding hardly provides justification for adoption
of a vegan diet. Of the few whole foods actually
measured, pork loin, eggs, and shrimp were animal
products containing no detectable dioxin. Even the
highest-ranking animal products, such as butter,
salmon, lamb and beef had at least one non-detect
among three samples, which means that only one or
two samples out of three are responsible for the
high ranking, while one or two out of the three
samples did not contain any detectable dioxin at
all.
Table 1 shows selected items
from the FDA's report.24 Animal products
are scattered among the highest and lowest concentrations
of dioxin TEQs, with some plant foods containing
significant quantities of dioxins. Since the FDA's
report contained little in the way of vegan protein
foods, I estimated the dioxin concentration of tofu
using the FDA's figures for the dioxin concentration
of
vegetable oil (usually meaning soybean oil) and
the USDA's25 data for the fat content
of tofu. Every single item listed had at least one
sample in which no dioxin was detected.
Since the US FDA's data rely on only three samples,
and since the variation between samples appears
to be large, it is necessary to look at other data
to establish or refute a trend--and this data destroys
the argument that dioxins are found primarily in
animal foods.
Table 2 gives data from Finland.9
Although animal products have a tendency to be higher
on the list, vegetables have higher concentrations
than milk products, and cereal products have more
than a 4-fold concentration of dioxins compared
to milk products. Fish take the cake in Finland,
with 163 times the dioxin concentration compared
to their nearest competitor, fats. According to
this data, avoiding meat in favor of cereals would
have a negligible impact compared to avoiding fish
in favor of meats. Unfortunately the Finnish data
give us no information on vegan sources of protein.
Table 3 provides data from
the Netherlands.8 Since animal products
except fish were reported per gram of fat while
the other data were reported per kilogram of product,
I adjusted the animal product data by choosing at
random a specific item from the USDA's database25
to represent each category, and multiplying the
grams of fat per gram of product by the pg TEQ per
gram fat to yield the adjusted figures. The data
for fish and plant products were divided by 1000
to yield the adjusted figures. As in Finland, fish
in the Netherlands are considerably higher in dioxins
than any other food, having 10 times the dioxin
concentration as beef. Still, vegetables have about
a 50 percent greater concentration of dioxins than
whole milk, and roughly the same concentration as
pork.
The data from Greece,22 shown in Table
4, are particularly damning to the notion that
dioxins exist primarily in animal products. In this
compilation, since plant products were reported
on the basis of product weight while animal products
were reported on the basis of grams of fat, it was
necessary to choose a specific food product to represent
each category and adjust the animal product figures
in the same way as the previous table. It is therefore
possible that some of the items, such as "farmed
sockeye salmon," were not actually sampled
in Greece, but serve only as a model to adjust the
figures for the purpose of comparison.
Amazingly, the Greek study found, with the exception
of fish oil, which isn't consumed in significant
quantities, that rice was the most concentrated
source of dioxin TEQs! Like the FDA's data for the
US, the Greek study found animal products to be
distributed between both the highest and lowest
sources of dioxins, but this study actually found
plant products in general to be higher in concentration
than most of the animal products. For example, vegetables
had almost six times the dioxin concentration
of beef liver.
In all of these analyses, in most cases the wide
variation of dioxin concentrations between regions
and between individual samples is wider than the
variation between types of foods. Animal products
tend to be distributed randomly among the highest
and lowest concentrations, yielding no consistent
trend of dioxin accumulation in animal products.
In some cases, such as the Greek data, vegetable
products dominate the higher-concentration readings,
and animal products dominate the lower-concentration
readings. In the United States, certain animal products
like butter are found to have the highest concentrations,
but the presence of at least one sample of butter
out of three with no detectable dioxins and three
out of three samples of half and half with no detectable
dioxins makes it impossible to claim a consistent
connection between butterfat and dioxin.
Thus, the second leg of the dioxin-based argument
for vegetarianism, that animal products are uniquely
high in dioxins, crumbles to pieces when subjected
to critical analysis.
Table
1.
Dioxin Concentrations in Foodstuffs, United
States, 2003.
Data are reported assuming a non-detect is
equal to zero.
| FOOD ITEM |
PG TEQ DIOXIN/G
PRODUCT |
| Butter |
0.2847 |
| Salmon Steak/Fillet |
0.1918 |
| Lamb Chop |
0.0714 |
| Beef Roast,
Chuck |
0.0687 |
| Beef Loin |
0.0606 |
| Olive Oil |
0.0295 |
| Tomato Sauce |
0.0232 |
| Beef liver |
0.0207 |
| Roasted Turkey Breast
|
0.0171 |
| Vegetable Oil |
0.0150 |
| Whole Milk |
0.0087 |
| Margarine |
0.0015 |
| Chicken Leg, fried with
skin |
0.0014 |
| Tofu, Firm |
0.0006 |
| White Beans |
0.0001 |
| Sunflower Seeds, roasted
|
0.0000 |
| Peanuts, dry roasted |
0.0000 |
| Tuna, canned |
0.0000 |
| Shrimp, boiled |
0.0000 |
| Half and Half |
0.0000 |
| Eggs, scrambled or boiled
|
0.0000 |
| Pork Loin |
0.0000 |
|
Table
2.
Dioxin Concentration in Foodstuffs, Finland,
2004.
Data are reported assuming a non-detect
is equal tozero.
| FOOD ITEM |
PG DIOXIN TEQ/G
PRODUCT |
| Fish |
1.80000 |
| Fats |
0.01100 |
| Meat and Eggs |
0.00820 |
| Cereal Products |
0.00430 |
| Solid Milk Products |
0.00270 |
| Vegetables |
0.00120 |
| Liquid Milk Products |
0.000930 |
| Fruits and Berries |
0.000830 |
| Beverages |
0.000170 |
| Potato Products |
0.000031 |
|
Table
3.
Dioxin Concentration in Foodstuffs, The Netherlands,
1999.
Data are reported assuming a non-detect
is equal to zero.
| FOOD ITEM |
PG DIOXIN TEQ/G
PRODUCT |
| Fatty Fish |
3.1580 |
| Crustaceans |
1.3540 |
| Butter |
1.3305 |
| Lean Fish |
0.5930 |
| Cheese |
0.4253 |
| 80% Lean Ground Beef |
0.3654 |
| Chicken, Dark Meat |
0.3230 |
| Margarine |
0.3000 |
| Prepared Fish |
0.2670 |
| Vegetable Oil |
0.1800 |
| Pork Chop Loin |
0.0614 |
| Vegetables |
0.0600 |
| Whole Milk |
0.0410 |
| Nuts |
0.0130 |
|
Table
4.
Dioxin Concentration in Foodstuffs, Greece,
2002.
Data are reported assuming a non-detect
is equal to the limit of detection.
| FOOD ITEM |
PG DIOXIN TEQ/G
PRODUCT |
| Fish Oil |
1.010 |
| Rice |
0.900 |
| Butter |
0.640 |
| Fruit |
0.470 |
| Vegetable |
0.430 |
| Olive Oil |
0.300 |
| Lamb Shoulder |
0.110 |
| 80% Lean Hamburger |
0.098 |
| Beef Liver |
0.077 |
| Pork Chop |
0.051 |
| Farmed Sockeye Salmon
|
0.050 |
| Boiled Egg |
0.039 |
| Chicken Leg |
0.017 |
| Wild Sockeye Salmon |
0.013 |
|
Factors Affecting Dioxin
Toxicity
Although the first tenet of the dioxin-based argument
for vegetarianism, that dioxins are potent human
carcinogens, endocrine disruptors, reproductive
disruptors, and immune disruptors appears to be
false or irrelevant to humans at the levels at which
they are exposed, it is still sensible for us to
err on the side of caution and, ceteris paribus
(all things being equal), opt for a lower dioxin
intake over a higher one. However, the argument
for vegetarianism does not use the ceteris paribus
stipulation; rather, it argues that dioxin intake
be minimized regardless of other factors.
The third and final leg of the dioxin-based argument
for vegetarianism,-- that avoiding the harmful effects
of dioxins is primarily dependent upon minimizing
dioxin intake and therefore avoiding animal products--implicitly
assumes that dioxin toxicity is merely a function
of dioxin intake. On the contrary, a variety of
dietary and other factors influence dioxin uptake
from the intestines, excretion of dioxin, the half-life
of dioxin in the body and the toxicity of dioxin
at the cellular level. As it turns out, vegetarian
diets tend to be lower in protective nutrients and
higher in toxicity-enhancing compounds, whereas
a traditional diet is highest in protective nutrients
and lowest in toxicity-enhancing compounds.
Not all dioxin consumed in a food is actually absorbed.
One human study found widely varying intestinal
absorption rates, with a maximum of 63 percent.
The study did find that a higher-fat meal produced
a higher absorption rate; however, since protective
compounds are also fat-soluble, it shouldn't be
concluded that a lower-fat diet is preferable. The
older individuals in this study actually had a net
excretion of dioxins, excreting more dioxins
in the feces than was present in the food. Apparently,
dioxins are stored in the tissues when tissue levels
are lower than blood levels, and released from the
tissues when blood levels drop below tissue levels.
Since dioxin levels have decreased so dramatically
over the past few decades, older individuals who
experienced the high peaks in environmental dioxin
levels in earlier decades cannot eat high enough
concentrations in food to prevent automatic tissue
release and fecal excretion.26
Various vegetable fibers have been shown to increase
fecal excretion of dioxin in animals. Chlorophyll
compounds, especially copper chlorophyllin, were
shown in one study to be the most effective compounds,
increasing excretion rates by 144 percent over normal
when fed as 1 percent of the diet by weight.27
This might indicate a modest benefit of chlorophyll-rich
vegetables (which would supply a much lower concentration
of chlorophyll than used in the study), which could
be obtained from both a vegetarian and a meat-based
diet.
Once absorbed from the diet or from other forms
of exposure into the bloodstream, dioxins are stored
in fatty tissues and slowly detoxified and excreted
over long periods of time. The half-lives of dioxins
are not consistent between individuals, however.
Investigations into the halflives of dioxins in
industrially exposed persons reveal that the variation
between minimum and maximum half-lives in different
individuals is often several times greater than
the value of the median half-life. Kidney and thyroid
disorders may inhibit detoxification of dioxins,
though results for thyroid disorders are conflicting.
Persons with a higher percentage body fat have a
slower dioxin decay rate, while intermediate weight
loss can increase the decay rate by a factor of
2.5. For some unknown reason, smoking increases
the decay rate significantly, although when adjusted
for age and percent body fat the association becomes
lower and for TCDD it becomes non-significant. For
certain dioxins, however, smoking decreases the
half-life by up to 25 percent.28
The most important variations in diet that affect
the potential for toxic effects of dioxins are antioxidants
and factors that increase oxidative damage in the
body, such as polyunsaturated fatty acids. Among
the antioxidants, vitamin A has many other roles
independent of its antioxidant activity and deserves
special attention, since depletion of vitamin A
and interference with vitamin A metabolism is central
to the toxicity of dioxins.
Dioxin Toxicity and Vitamin
A
Although relatively little is known about which
factors tie the ability of dioxins to bind to the
aryl hydrocarbon receptor (AhR) and the subsequent
activation of the cytochrome P-450 system to their
toxicity, it is clear that one of the missing links
is vitamin A. Changes in vitamin A levels in wildlife
are correlated with dioxin exposure, and TCDD is
able to experimentally induce vitamin A depletion
as well as resistance to vitamin A signaling, which
is correlated with its toxic effects. Also, TCDD
and vitamin A have opposing actions in certain tissues,
and the addition of dietary vitamin A exerts a strong
protective effect against a wide range of TCDD-induced
effects.
The most consistent effects observed in wildlife
in response to dioxin exposure are changes in vitamin
A and thyroid hormone levels. Changes in liver or
plasma vitamin A concentrations have occurred in
captive harbor seals eating polluted fish, Great
Lakes herring gulls and tree swallows, great blue
herons and lake sturgeon of the St. Lawrence River,
common terns of Belgium and the Netherlands and
white suckerfish of Montreal. Typically, decreases
in liver or plasma vitamin A are observed, or signs
of increased mobilization of vitamin A from the
liver. In several of these cases, decreased levels
of thyroid hormone have also occurred, and in cormorants
of the Netherlands, a decrease in free thyroid hormone
was observed without changes in vitamin A.16
When rats were fed daily doses of dioxins roughly
equivalent to one million times more than humans
typically consume, major impacts on vitamin A and
thyroid hormone levels occurred. TCDD increased
blood levels of vitamin A by 21 percent, while all
other dioxins decreased blood levels. All of the
dioxins, including TCDD, depleted liver stores of
vitamin A by 60-80 percent. This was considered
a "very sensitive response" to dioxins,
since even the lowest dose, only 70,000 times the
equivalent of that which humans consume, produced
a statistically significant effect. A dose-dependent
reduction of thyroid hormone (T4) was
induced, yielding a 76 percent reduction at a dose
equivalent to two million times more than humans
typically consume, and still yielding a significant
50 percent decrease even in the group fed only 70,000
times more than humans typically consume.29
The above study found the effect of TCDD at reducing
thyroid hormone levels to be much less potent than
that of the other dioxins, while it actually raised
blood levels of vitamin A rather than lowering them.
This is probably because some of the other dioxins
produce metabolites that bind to transthyretin,
the protein that transports both vitamin A and thyroid
hormone in the blood. TCDD, however, does not have
this effect. The study found the WHO's TEQ concept
to have no predictive value with respect to these
effects. This calls into question whether vegan
diets that are lower in dioxin TEQs but comparable
in absolute quantities of dioxin-like PCBs are truly
lower in toxic elements.19
Dioxins, Vitamin A, and
Cancer
It appears that the capacity of dioxins to produce
both cancer and non-cancer toxicity relates to their
ability to deplete vitamin A reserves and oppose
the actions of vitamin A in the body. In cultured
human skin cells incubated with TCDD, TCDD induces
the expression of transforming growth factor alpha
(TGF-a) and decreases the expression of transforming
growth factor beta-2 (TGF-ß2), while incubation
with retinoic acid, the hormone form of vitamin
A, increases the expression of TGF-ß2. (TGF-a
increases cellular proliferation, while TGF-ß2
has the opposite effect.) Since excessive cellular
proliferation is a mechanism of cancer promotion--causing
cells to multiply before they are able to fix DNA
damage--this may explain part of the carcinogenic
potential of dioxins and the protective effect of
vitamin A.30
On the other hand, in human breast cancer cells
where dioxins inhibit cancer, vitamin A
enhances the anti-estrogenic effect of
dioxins. Both retinoic acid and TCDD inhibit breast
cancer in rodents by opposing the effects of estrogen.
In cultured human cells, TCDD and retinoic acid
inhibit estrogen-induced cell proliferation and
the synthesis of estrogen receptors, and the effectiveness
of each is enhanced when used together.31
Both the carcinogenic and non-carcinogenic toxicity
of dioxins are believed to stem from the ability
of dioxins to bind to the AhR and induce the formation
of the cytochrome P-450 system. A recent study showed
that vitamin A fed to rodents reduced the TCDD-induced
expression of cytochrome P-450 by 68 percent.32
Other studies also show vitamin A to be effective,
along with other antioxidants, in inhibiting the
free radical products that are induced by dioxins
and also believed to play a role in carcinogenesis,
as well as many other toxic effects, discussed below.
Dioxins, Vitamin A,
and Non-Cancer Toxicity
Many of the observed toxic effects of dioxins
resemble those of vitamin A deficiency. Table
5 shows selected effects of dioxins in various
species that are also widely accepted to be effects
of vitamin A deficiency. Diseases such as cancer
that are effectively treated with or prevented by
vitamin A but are not considered deficiencies of
vitamin A in standard literature are not included.
Many of the toxic effects induced by dioxins correlate
with vitamin A depletion. TCDD can result in impaired
growth and wasting disease, and in the guinea pig,
rat, mouse and hamster, a dose-response relationship
has been demonstrated between degree of vitamin
A depletion and degree of depressed weight gain.38
Decreased vitamin A stores have been found along
with hyperkeratotic skin diseases in elephant seals,
decreased fertility and suppressed immune function
in harbor seals, suppressed immune function in herring
gulls, and increased birth defects in white suckerfish
and lake sturgeon, all of which resemble the effects
of vitamin A deficiency and were associated with
exposure to dioxins or organochlorines in general.16
Thus, dioxins deplete vitamin A stores and are
associated with many effects that seem to mimic
vitamin A deficiency. But there is more to the story.
Although liver reserves are depleted when vitamin
A deficiency-like symptoms induced by dioxins arise,
these symptoms usually occur when there are still
significant tissue reserves remaining, whereas in
simple vitamin A deficiency, symptoms usually do
not occur until tissue reserves are almost entirely
depleted. Dioxins appear not only to deplete vitamin
A, but also to induce cellular resistance to retinoic
acid, which is the hormone form of vitamin A.39
Many effects of dioxins can be reversed by vitamin
A. Supplementation with vitamin A enabled 25 percent
of rats fed a lethal dose of TCDD to survive, while
supplementation with vitamin E enabled only 10 percent
to survive.40 Injection of vitamin A
into a fertile egg largely protected against the
increase in mortal
