The Health Effects of Dioxins Furans and PCBs

Over a billion dollars has been spent on research to determine the extent to which dioxins, furans, and PCBs cause toxic reactions in humans. Nevertheless, conclusions about this issue are still tentative and controversial. Evidence about toxicity is derived from two sources:

• toxicological experiments on animals that have been deliberately exposed to the chemicals; and

• epidemiological studies of humans who have been accidentally exposed.

It is generally agreed that most PCBs are not acutely toxic to humans: The LD50 values of most congeners are large. In high doses, PCBs cause cancer in test animals, and consequently they are listed as a "probable human carcinogen" by the U.S. EPA. However, studies of humans exposed to them have produced inconsistent results. Most groups of people who have been exposed to relatively high concentrations of PCBs—e.g., as a result of their employment in electrical capacitor plants—have not experienced a higher overall death rate. (See, however, the comments about human cancer in Chapter 12.) The most common reaction to exposure to PCBs is chloracne, a biological response by humans to exposure to many types of organochlorine compounds.

Inadvertent PCB Poisonings

The most dramatic effects observed to human health from exposure to PCB mixtures occurred when two sets of people, one in Japan in 1968 and the other in Taiwan in 1979, unintentionally consumed PCBs that had accidentally been mixed with cooking oil. In the Japanese incident, and probably in the Taiwanese case as well, the PCBs had been used as a heat exchanger fluid in the deodorization process for the oil. Since the PCB-contaminated oils had been heated, their level of PCDF contamination was much greater than occurs in freshly prepared commercial PCBs. The thousands of Japanese and the Taiwanese people who consumed the contaminated oils suffered health effects far worse than has been found for workers at PCB manufacturing and handling plants, even though the resulting PCB levels in their bodies were about the same. From this difference, it has been concluded that the main toxic agents in the poisonings were the PCDFs and that they and dioxins were collectively responsible for about two-thirds of the health effects, with the PCBs themselves responsible for the remainder. Indeed, studies on laboratory animals indicate that the furans involved in these incidents are more than 500 times as toxic on a gram-for-gram basis than are pure PCBs. Cognitive development, as measured by IQ scores, of children born to the most highly exposed Taiwanese mothers—even if births occurred long after the consumption of the contaminated oil—was found to be significantly lower than that of their siblings born before the accident occurred and for children of unexposed mothers. Interestingly, children whose fathers but not their mothers had consumed the oil showed no detrimental effects. Further effects of this incident are discussed in Chapter 12 {Environmental Estrogens section).

An incident comparable to the Asian cooking oil poisonings occurred in early 1999 in Belgium. Several kilograms of a mixture of PCBs that had previously been heated to a high temperature—converting a tiny fraction of the PCBs to furans—were put into an 80,000-kg batch of animal fat, which was then mixed with animal feed and shipped to about 1000 farmers. Poultry producers subsequently noticed a sudden drop in egg production and in egg hatch-ability, and high concentrations of dioxins were subsequently found in chicken meat. The contaminated food was withdrawn from the market and destroyed.

Effects of in Utem Exposure to PCBs

Are humans who have been exposed to PCBs through their diets especially susceptible to reproductive problems? To answer this question, Sandra and Joseph Jacobson and their co-workers at Wayne State University in Detroit have spent more than two decades studying the offspring of people from the Lake Michigan area, including children whose mothers regularly eat fish from the lake and who, as a result, are expected to have elevated levels of PCBs in their bodies. They have discovered statistically significant differences in children born to women who have high levels of PCBs; these differences were present not only at birth but persisted to the age of at least eleven years.

The prenatal (i.e., prebirth) exposure of the infants to PCBs was determined by analyzing the blood of their umbilical cords for these chemicals after birth. Because analytical techniques in the early 1980s were not sufficiently sensitive to detect the amounts of PCB s in all the umbilical cord samples, exposure for many infants was estimated from the PCB levels in their mother's blood and breast milk. The postnatal exposure of the children was assessed by analyzing their mother's breast milk and also by analyzing blood samples from the children at the age of four years.

The Jacobsons discovered that, at birth, the children of mothers who had transmitted the highest amounts of PCBs to the children before birth had, on average, a slightly lower birth weight and a slightly smaller head circumference; they were also on average slightly more premature than those born to women who passed along lower amounts. The severity of these deficits was larger the greater their prenatal exposure to PCBs. When tested at the age of seven months, many of the affected children displayed small difficulties with visual recognition memory, again with the extent of the problem increasing with prenatal transmission of PCBs. At the age of four years, the lower body weight observed at birth for highly exposed infants still lingered. More serious was the observation that the four-year-olds displayed progressively lower scores on several tests of mental functioning (with respect to verbal and memory abilities) the greater their prenatal PCB exposure. A comparable study in Holland, using more modern analytical instrumentation, found similar results (see Figure 1 l-7a).

At eleven years of age, the effects of the prenatal PCBs were still apparent: the IQ scores of the part of the group that had been the most highly exposed before birth averaged 6 points below the others; the most affected mental functions were memory and attention span. However, prenatal PCB exposure at any but the highest levels did not appear to have affected the IQ of the eleven-year-olds (see Figure 11-7b).

Interestingly, at both four and eleven years of age, the children's total body content of PCBs, which is determined mainly from the breast milk they consumed as infants and to a lesser extent by the fish in their diet rather than from any prenatal transmission of the chemicals, was not the relevant factor in determining these physical and mental deficits. Rather, it was the smaller amounts of PCBs that had been transmitted from mother to fetus that were important. Thus PCBs appear to interfere with the proper prenatal development of the brain and with the mechanisms that ultimately determine physical size.

The study by the Jacobsons is one of the clearest examples available concerning the influence of toxic environmental chemicals on the health of human beings. It is important to realize that the highest levels of PCBs to which the children in this group were exposed before birth are not much greater than those to which the majority of unborn children in the general population were subjected a few decades ago. And while the chemicals did not produce gross birth defects in the children, they did result in small and consistent deficits of several kinds.

Health Effects Pcbs Table
<1.5 1.5-1.99 2.0- 2.49 2.5-2.99 >3.0 Sum PCB level in maternal plasma (pg/L)
Pcbs Effect Human Health

<0.5 0.5- 0.75- 1.00.749 0.99 1.249 Prenatal PCB exposure (/xg/g)

<0.5 0.5- 0.75- 1.00.749 0.99 1.249 Prenatal PCB exposure (/xg/g)

FIGURE 11-7 The effect on the intelligence of children of receiving PCBs prenatally: (a) overall cognitive abilities at 372 years of age, (b) full-scale IQ at 11 years. ISources: (a) S. Ratandin et al., "Effects of Environmental Exposure to Polychlorinated Biphenyls and Dioxins on Cognative Abilities in Dutch Children at 42 Months of Age/' Journal of Pediatrics 134 (1999): 33. (b) J. L. Jacobson et al., "A Benchmark Dose Analysis of Prenatal Exposure to Polychlorinated Biphenyls/' Environmental Health Perspectives 110(2002): 393.1

Studies of infants in North Carolina and in upper New York State have produced similar results to those found by the Jacobsons. Nevertheless, some doubt has been cast on the Jacobsons' results by an analysis that pointed out the difficulties in establishing the original in utero exposures. A more recent study in the Netherlands overcame these analytical difficulties, since the ability to detect very low levels of chemicals such as PCBs in blood serum has improved dramatically. The ongoing Dutch study has supported the Jacobsons' findings. It found that prenatal exposure to PCBs is more important than postnatal exposure and produces lower birth and growth weight and lesser cognitive abilities in young children, although the latter deficit was found for all but the least exposed fraction of children, rather than only the most highly exposed fraction. Negative effects on the IQs of three-year-old children by postnatal exposure to PCBs, as measured by blood levels, has recently been discovered by researchers in Germany. Subtle cognitive deficits have also been found in children in northern Quebec whose PCB concentrations are high due to the long-range air transport and subsequent deposition and entry into the food chain of the compounds (as discussed in Chapter 12). Dutch researchers also found that PCBs and dioxins transmitted to babies during gestation and via breast milk weakened their immune systems, contributing to more infections in the first few years of life.

The Toxicity Patterns of Dioxins, Furans, and PCBs

Research has shown that single doses of 2,3,7,8-TCDD administered to pregnant laboratory animals cause reproductive effects in their offspring. These results have raised some alarm about the potential effects of dioxins on human reproduction. In this connection, many scientists are worried about the dangers posed by environmental chemicals such as dioxins, furans, PCBs, and other organochlorines that can affect sex hormones, as discussed in Chapter 12.

Test results from studies of animals indicate that the acute toxicity of dioxins, furans, and PCBs depends to an extraordinary degree on the extent and pattern of chlorine substitution. The following generalization can be made; the very toxic dioxins are those with four beta chlorine atoms, and few if any alpha chlorines (see diagram on page 472 for definitions of the alpha and beta positions). Thus the most toxic is 2,3,7,8-TCDD, which has the maximum number (four) of beta chlorines and no alpha chlorines.

Dioxin congeners that have three beta chlorines, but no (or only one) alpha chlorine, are appreciably toxic, but less so than the 2,3,7,8 compound. Fully chlorinated dioxin, that is, octachlorodibenzo-p-dioxin (OCDD), has a very low toxicity since all the alpha positions are also occupied by chlorine.

Similarly, mono- and dichloro dioxins are usually not considered highly toxic, even if the chlorines are present in beta positions.


Predict the order of relative toxicities of the following three dioxin congeners, given that, for systems not too dissimilar to TCDD, the presence of an alpha chlorine reduces the toxicity less than does the absence of a beta chlorine;




The toxicity pattern for furans is similar though not identical to that for dioxins in that the most toxic congeners have chlorines in all the beta positions. However, the most toxic furan, the 2,3,4,7,8 congener, does have one chlorine atom in an alpha position.

According to animal tests, the most acutely toxic PCBs are those having no chlorine atoms (or at most one) in the positions that are ortho to the carbons that join the rings, that is, on the 2, 2', 6, and 6' carbons. Without ortho chlorines, the two benzene rings can easily adopt an almost coplanar configuration, and rotation about the C—C bond joining the rings is rapid. However, because of the large size of chlorine atoms, they get in each other's way if they are present in both ortho positions on the same side of the two rings; this interaction forces the rings to twist away from each other, preventing such rings from adopting the coplanar geometry:



Consequently, PCB molecules with chlorines at three or four of the ortho positions cannot adopt a coplanar geometry.

If the rings are not kept from coplanarity by interference between chlorine atoms, and if the hydrogen atoms at certain meta and para carbons are replaced by chlorine, then the PCB molecule can readily attain a coplanar geometry that is similar in size and shape to 2,3,7,8-TCDD. Such PCB molecules are found to be highly toxic. Apparently 2,3,7,8-TCDD and other molecules of its size and shape readily fit into the same cavity in a specific biological receptor; the complex of the molecule and the receptor can pass through cell membranes and thereby initiate toxic action. By comparing molecular models, it is not difficult to see, e.g., that the most toxic PCB, namely 3,3',4,4',5'-pentachlorobiphenyl, is almost the same size and shape as 2,3,7,8-TCDD. It is believed that some of the toxicity of the cooking-oil incidents arose from coplanar PCBs.

Only a very small fraction of commercial PCB mixtures corresponds to coplanar PCBs having no ortho chlorines. Although individually less toxic, PCBs with one ortho chlorine, and with chlorines in both para and at least one meta position, contribute substantially to the overall toxicity of PCB mixtures since they are far more prevalent than those having no ortho chlorines.

In humans, the more highly chlorinated furans, dioxins, and PCBs are stored in fatty tissues and are neither readily metabolized nor excreted. This persistence is a consequence of their structure: Few of them contain hydrogen atoms on adjacent pairs of carbons at which hydroxyl groups, OH, can readily be added in the biochemical reactions that are necessary for their elimination. In contrast, those compounds with few chlorines always contain one or more such adjacent pairs of hydrogens and tend to be excreted after hydrox-ylation, rather than stored for a long time.

The TEQ Scale

Since most organisms, including humans, have a mixture of many dioxins, furans, and PCBs stored in their body fat, and since all these compounds act qualitatively in the same way, it is useful to have a measure of the net toxicity of the mixture. To this end, scientists often report concentrations of these organochlorines in terms of the equivalent amount of 2,3,7,8-TCDD that, if present alone, would produce the same toxic effect. An international toxicity equivalency factor, or TEQ, has been devised that rates each dioxin, furan, and PCB congener's toxicity relative to that of 2,3,7,8-TCDD, which is arbitrarily assigned a value of 1.0. Recently, polybrominated biphenyls have also been added to this scale.

A summary of the TEQ values for some of the more toxic dioxins, furans, and PCBs is given in Table 11-1. As an example, consider an individual who ingests 30 pg (picograms) of 2,3,7,8-TCDD, 60 pg of 1,2,3,7,8-PCDF, and

TABLE 11-1

Toxicity Equivalence Factors (TEQ) for Some Important Dioxins, Furans, and PCBs

Dioxin or Furan or PCB

Toxicity Equivalency Factor

2,3,7, 8-Tetrachlorodibenzo-p-dioxin 1,2,3,7,8-Pentachlorodibenzo-p-dioxin l,2,3,4,6,7,8-Heptachlorodibenzo-f)-dioxin Octachlorodibenzo-p-dioxin

2,3,7,8-Tetrachlorodibenzofuran 2,3,4,7,8-Pentachlorodibenzofuran 1,2,3,7,8-Pentachlorodibenzofuran 1,2,3,4,7,8 - Hexachlorodibenzofuran 1,2,3,6,7,8 - Hexachlorodibenzofuran 2,3,4,6,7,8-Hexachlorodibenzofuran Octachlorodibenzofuran

3,3',4,4',5-Pentachlorobiphenyl 3,3',4,4',5,5'-Hexachlorobiphenyl

0.01 0.001

0.01 0.001

200 pg of OCDD. Since the TEQ factors for these three substances are, respectively, 1.0, 0.05, and 0.001, the intake is equivalent to

(30 pg X 1.0) + (60 pg X 0.05) + (200 pg X 0.001) = 33.2 pg

Thus, even though a total of 290 pg of dioxins and furans were ingested by this person, the mixture is equivalent in its toxicity to an intake of only 33.2 pg of 2,3,7,8-TCDD.


Using the TEQ values in Table 11-1, calculate the number of equivalent picograms of 2,3,7,8-TCDD that corresponds to an intake of a mixture of 24 pg of l,2,3,7,8,9-hexachlorodibenzo-f>-dioxin, 52 pg of 2,3,4,7,8-pentachlorodibenzofuran, and 200 pg of octachlorodibenzofuran.

The TEQ values for environmental samples are sometimes reported in the media as if they represent the concentration of 2,3,7,8-TCDD itself However, this compound often is not even the dominant contributor to the TEQ toxicity. As discussed previously, combustion of organic matter produces relatively few toxic dioxins; the TEQ from such sources is often dominated by penta- and hexachlorinated furans. Similarly, the TEQ arising from the use of chlorination in bleaching paper is dominated by toxicity from tetrachlorinated furans.

Dioxins, Furans, and PCBs in Food

About 95% of human exposure to dioxins and furans arises from the presence of the compounds in food. A bar graph showing the TEQ values for contamination of various types of foods purchased in U.S. supermarkets in the 1990s is shown in Figure 11-8. Notice that fresh-water fish contained the highest levels of both PCB and furan toxicity. Recently, average TEQ levels of several ppt have been found for farmed salmon; these high levels originate with the fishmeal and fish oil the growing salmon had been fed. However, the TEQ levels in young chickens and turkeys and in hogs sold in the United States has declined significantly—by 20 to 80%—in recent years. The composite vegan diet (i.e., all vegetable, fruit, and grain, with no animal products at all) has a very low TEQ compared to that from animal-based components.

FIGURE 11-8 TEQ values for foods collected from U.S. supermarkets. [Source: A. Schecter et al., "Levels of Dioxins, Dibenzofu-rans, PCB and DDE Congeners in Pooled Food Samples Collected in 1995 at Supermarkets Across the United States/' Chemosphere 34 '1997>: 1437.'





Dioxin Levels Foods


A 2003 report by the U.S. Institute of Medicine recommended that girls should reduce their consumption of animal products in order to reduce the amount of dioxins that build up in their body fat and that could subsequently affect any children they might have in the future.


Given that the average TEQ of animal-based foods was about 0.4 pg of TCDD equivalent per gram when the data in Figure 11-8 were collected, and that the LD50 for 2,3,7,8-TCDD is about 0.001 mg/kg body weight, what mass of animal-based food would you have had to consume to ingest a fatal dose of it?

In the mid-1980s, the average total concentration of all dioxins and furans in the fat tissue of adult North Americans was about 1000 ppt. However, because highly chlorinated and therefore less toxic congeners predominated, the TEQ value was much less: about 40 ppt of 2,3,7,8-TCDD equivalent. By the 1990s, the TEQ in human fat had fallen to about 15 ppt. The historical maximum TEQ levels, about 75 ppt, were reached in the 1970s. The variation with time of stored dioxins and furans fits a model in which the daily TEQ dose amounted to about 0.5 pg/kg body weight in the early decades of the twentieth century, rose to over 6 pg/kg in the 1940s to 1970s, and has now declined again to 0.5 pg/kg.

The average North American adult's body contains about 15 kg of fat, so his or her total body burden of 2,3,7,8-TCDD equivalents now amounts to about 0.2 /iig. Given that the average residence time t,avg of dioxins and furans in the human body is about seven years, and using the relationship from Chapter 6 connecting £avg to the total amount C and the input rate R, i.e.,

then the average human rate of intake of 2,3,7,8-TCDD equivalents is calculated to be

This value, which corresponds to about 1 pg per kilogram body weight per day, is close to the intake estimated over the last few years from the model discussed previously.

Dioxins as Probable Human Carcinogens

Although there is little argument as to the relative acute toxicities of various dioxin and furan congeners, their absolute risk to humans is very controversial. The amount of 2,3,7,8-TCDD per kilogram body weight required to kill a guinea pig is extraordinarily small—about 1 fig—making it the most toxic synthetic chemical known for that species. However, the LD50 required to kill many other types of animals is hundreds or thousands of times this amount—e.g., the LOD50 for hamsters 1200 fig/kg; for frogs, 1000 /¿g/kg; for rabbits and mice, 115 g/kg; for monkeys, 70 /¿g/kg; and for dogs, it may be as low as 30 /¿g/kg. A 2000 EPA report on dioxin suggests that humans fall in the middle range for acute susceptibility to dioxins; humans given doses of 100 /¿g/kg suffered no apparent ill effects beyond chloroacne.

Another widespread exposure of humans to 2,3,7,8-TCDD occurred in the early 1970s in and around Times Beach, Missouri. Waste oil containing PCBs and 2,3,7,8-TCDD from the manufacture of 2,4,5-trichlorophenol was used for dust control on gravel roads. Some horses died due to exposure in an area where the dioxin contamination was particularly high, and some children became ill. A decade later, widespread contamination of the soil in the town was discovered. In 1997, over 200,000 tonnes of soil from this town and 26 other affected sites in eastern Missouri that had soil 2,3,7,8-TCDD levels of 30-200 ppb were excavated and incinerated in order to remediate the problem. (The town of Times Beach was abandoned in 1982 due in part to a severe flash flood.) Although exposure to the chemical seems to have negatively affected their immune systems, a major study in 1986 did not find evidence of increased disease prevalence in the exposed group of Times Beach residents. Less formal studies and anecdotal evidence, however, indicate problems with seizures and congenital abnormalities, and so the issue remains controversial.

Scientists are more concerned about the long-term effects of exposure to dioxins than about their acute toxicities. The Seveso study discussed previously was the first to show an increased rate of cancer among people exposed accidentally to TCDD. A recent analysis of the health of American workers who were employed in industries that produced chemicals contaminated with 2,3,7,8-TCDD indicates that exposure to it at relatively high levels may cause cancer. The current theory concerning the action of dioxin predicts that there should be a threshold below which no toxic effects will occur, and recent studies of workers exposed to 2,3,7,8-TCDD supports this hypothesis.

In animal studies, a threshold of about 1000 pg (i.e., 1 ng) of 2,3,7,8-TCDD equivalent per kilogram of body weight per day is observed with respect to the cancer-causing ability of dioxins and furans. In determining the maximum tolerable human exposure to such compounds, many governments apply a safety factor of 100, resulting in a guideline for maximum exposure of 10 pg/kg/day averaged over a lifetime. Currently, the average American ingests only about one-tenth this amount from animal fats in his or her food supply. Exposure levels near the guideline limit are expected for persons consuming large amounts of fish that have elevated dioxin and furan levels.

The U.S. EPA draft report of 2000 concerning the health risks of dioxins concluded that 2,3,7,8-TCDD is a (known) human carcinogen—although this characterization was a point of controversy among members of the expert committee that reviewed the report—and that the mixtures of dioxins to which people are exposed is a "likely human carcinogen." The International Agency for Research on Cancer of the World Health Organization had previously classified TCDD as a known human carcinogen. Experiments had indicated that 2,3,7,8-TCDD was the most potent multisite carcinogen known in test animals. The EPA estimates that the most sensitive and most highly exposed Americans stand at least a 1-in-1000 chance of developing cancer from dioxin. The report notes that noncancer effects of dioxin are at least as important as cancer.

However, there is currently no clear indication that the level of cancer has risen in the general American population due to dioxins, though this could well be due to the inability to relate effects to exposure at current levels. In addition to cancer, the report concludes that dioxins adversely affect the endocrine and immune systems and the development of fetuses, as will be discussed in Chapter 12.

For furans, direct evidence of human susceptibility is available from the incidents of PCB-contaminated cooking-oil consumption mentioned above. The most common symptoms observed in these groups were chloracne and other skin problems. Unusual pigmentation occurred in the skin of babies bom to some of the mothers who had been exposed. The children also often had low birth weight and experienced a rather high infant mortality rate. Children who directly consumed the oil showed retarded growth and abnormal tooth development. Many of the victims also reported aching or numbness in various parts of their bodies and frequent bronchial problems. Other than chloracne, such symptoms are not observed in workers who are occupationally exposed to PCBs and whose body burdens of PCBs are comparable to those who consumed the contaminated oil, but in which the PCDF concentration is orders of magnitude lower. However, some children of these workers had mild cases of the less serious problems seen in the poisoned group.

Ukrainian president Viktor Yushchenko was apparently the recent victim of deliberate dioxin poisoning. Since September 2004, he has suffered from ulcers in his stomach and intestines, problems with his liver and spleen, and disfiguring facial cysts that have left him looking far older than he is (Figure 11-9).

Human Exposure to Dioxins, Furans, and PCBs

There continues to be vigorous debate in scientific, industrial, and medical communities regarding the environmental dangers of dioxins, furans, and PCBs. In one camp are those who feel that the dangers from these chemicals have been wildly overstated in the media and by some special-interest groups. They point to the very low concentrations of these substances that exist in the environment, to the possibility that there is a threshold below which

Yushchenko Before After

FIGURE 11-9 Ukranian president Viktor Yushchenko before and after he was poisoned with dioxin. (Source: AP Photo/Efrem Lukatsky.)

these compounds have no effect on human health, to the lack of known human fatalities resulting from them, and to the enormous economic costs associated with instituting further controls and cleanup measures. At the other extreme are persons who point to the substantial biomagnification and high toxicity per molecule of these substances and to their presence in almost all environments. They consider the detrimental effects such as cancer and birth deformities caused by these chemicals in wildlife to be "warning canaries" that signal potential ill effects in humans. Discovering where the truth lies between these opposing viewpoints presents a challenge even for environmental science students, to say nothing of the public at large!

Continue reading here: Introduction

Was this article helpful?

0 0


  • ted banks
    Why the toxicity of dioxins and furans depends on the extent and pattern of chlorine substitution?
    5 years ago