Pesticides and Breastfeeding
LLLI Breastfeeding Reference Library & Database
From: LEAVEN, Vol. 30 No. 3, May-June 1994, pp. 37-40
Ed. Note: We provide articles from our publications from previous years for reference for our Leaders and members. Readers are cautioned to remember that research and medical information change over time.
Very few people any longer dispute the basic physiological and psychological superiority of human milk and breastfeeding. In this contaminated world in which we live, however, some wonder if human milk and breastfeeding can still be unequivocally recommended for every mother and baby? And what about the alternative for baby, infant formula? What steps can a woman take during pregnancy and lactation to minimize her, and consequently her developing baby's, exposure to pesticides and other organochlorine environmental toxins, such as DDT, DDE, PCBs and associated dioxins and furans, heavy metals such as lead, cadmium, and mercury, and others? Let's begin by taking a look at what we already know and what we must make our best educated guesses about for the future.
One well known and respected researcher in the contaminant field, Dr. Walter Rogan, works for the Epidemiology Branch of the National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina. Along with his colleagues, Dr. Rogan has spent fifteen years doing a prospective, ongoing study of North Carolina mothers and their children. It is interesting to see how his research has progressed and his viewpoint has changed over the years.
In 1980 in the New England Journal of Medicine, Dr. Rogan reviewed what was then known about "Pollutants in Breast Milk," namely PCBs and the pesticides DDT, dieldrin, heptachlor, and chlordane. He began by reiterating that these chlorinated pesticides are lipid soluble, gradually accumulating in the general population's body fat over long-term exposure and being concentrated in the lipids of human milk. As a matter of fact, lactation is the main route of elimination of these persistent chemicals from the female body.
He also pointed out that it's very difficult to accurately quantify the amounts of these organohalides in breast milk because levels vary widely from the beginning of a feed to the end, as well as throughout the day in any one sample of breast milk from any one woman. The findings of different research studies are also difficult to compare depending on whether reported contaminant levels were determined from whole milk or milk fat, which can give substantially higher values.
Dr. Rogan examined a number of previous studies which found that PCBs and DDT and its derivatives were present in virtually all human milk samples from all over the world, even though DDT was banned in the USA in 1972 and PCBs in 1976. They are still in widespread use in other areas of the world.
PCBs, used mainly in the electrical industry as transformer-insulating fluids, leaked in huge concentrations into rice oil in Japan and Taiwan in the late 1960s and 1970s, respectively. Adult health consequences included an acne-like rash, headache, nausea, and diarrhea. Newborns were small for gestational age, had dark skin pigmentation, conjunctivitis, and jaundice. Children who were breastfed had higher blood levels of PCBs and levels increased with the duration of breastfeeding. Follow-up of some of these children at about nine years showed slight but clinically significant neurologic and developmental impairment. This was an unusual exposure, not an everyday occurrence.
There were no reports of DDT toxicity in either occupationally exposed workers or experimentally exposed volunteers receiving doses considerably greater than in the general population.
Animal studies were a different story, however, but Dr. Rogan cautioned about equating effects in animals with humans. Animal infants, particularly monkeys, have less body fat than human infants, so contaminants are more likely to attack target organs. Rogan acknowledged that further studies needed to be done, and he has proceeded to do one of the best.
In 1986, Rogan and others published the first in a continuing series of reports on milk samples from about 900 women in North Carolina and the development of their children, born between 1978 and 1982. This 1986 study published in the American Journal of Public Health specifically measured levels of PCBs and DDE in human milk, blood, and other body tissues at birth, six weeks, three months and six months, and from some of the mothers, at one year, and eighteen months.
Almost all samples showed detectable levels of both PCBs and DDE. The highest levels were in older women, Black women, first-time mothers, cigarette smokers, and alcohol drinkers. PCBs, in particular, were higher in women consuming sport fish during pregnancy. Another important finding was the decline in contaminant levels in the milk during the course of lactation and with the number of children breastfed.
A second study in 1986 published in the Journal of Pediatrics, was the first to report on about 900 of the North Carolina children and the "Neonatal effects of transplacental exposure to PCBs and DDE." There were no associations between PCB and DDE levels and birth weight, head circumference, and hyperbilirubinemia. The Brazelton Neonatal Behavioral Assessment Scales were administered after birth. Both chemicals were associated with hyporeflexia (lower reflex response). Higher PCB levels were also associated with hypotonicity (lower muscle tone). At that time, Rogan expressed concern about continued exposure to PCBs and DDE through breastfeeding.
In 1987, the next report followed the children to one year of age; it was published in the American Journal of Public Health. The purpose of the study was to ascertain if the presence of PCBs and DDE in their mothers' milk had affected these children's growth or health. Neither chemical showed an adverse effect on weight or frequency of doctors' visits for illness, although differences were seen between breastfed and bottle-fed children. Bottle-fed children tended to weigh more and have more frequent gastroenteritis and otitis media.
In 1988 in the Journal of Pediatrics, it was reported that the children were examined for any mental and psychomotor developmental effects at one year of age from transplacental exposure to PCBs and DDE. Higher transplacental exposure to PCBs was associated with lower psychomotor scores at both six and twelve months of age. The key finding here was that exposure to either chemical through breastfeeding was unrelated to the Bayley Scales of Infant Development test scores. No adverse effects on growth or occurrence of illnesses in the first year of life have been attributed to the presence of these chemicals in human milk.
In 1991 three more studies in this series were published:
A study published in the Annals of Epidemiology looked at "Child Development at Eighteen to Twenty-four Months" versus PCBs and DDE exposure. The trend toward lower psychomotor scores on the Bayley tests was still present and determined to still be due to transplacental exposure to the chemicals. There was no evidence of any effect from the larger but later exposure through breast milk, but at that time, Rogan expressed concern that such effects might show up as the children reached an older age.
Another study in the Journal of Pediatrics followed up the children, now age five-and-a-half to ten-and-a-half years, with the McCarthy Scales of Children's Abilities and by checking report cards, when available, to see if the psychomotor deficit through two years of age, due to transplacental but not breastfeeding exposure to PCBs, was still present. The deficits seen at two years on the Bayley Scales of Infant Development were no longer apparent.
The third 1991 study in Regulatory Toxicology and Pharmacology was particularly important. Rogan and colleagues developed various types of mathematical analyses to answer concerns about possible lifetime cancer risks from human milk exposure to chemical contaminants. The pesticides considered were DDT, chlordane, heptachlor, dieldrin and their metabolites, and industrial compounds or byproducts, such as PCBs and dioxins.
An initial statement was made in that article that the contaminant levels in human milk would prevent its sale as a commercial food for infants, based on current standards. This type of statement is made quite casually from time to time, with no regard for the concern it causes breastfeeding mothers.
Regulatory standards have been set by the World Health Organization and US Food and Drug Administration to attempt to control the levels of contaminants in foods. Current "allowable" levels have been set low to prevent toxicity in adults from exposure over a lifetime. In reality, there is a range in concentration of the chemicals, sometimes 100 to 1000 times higher, over which no adverse reactions have been reported. It is also important to remember that human milk constitutes only one very limited route and time of exposure to contaminants over an individual's lifetime. In addition, the rapidly growing infant is gaining body fat, and thus a place to effectively store contaminants, during the time it is receiving human milk. What this means is that even though human milk may contain higher levels of contaminants than some commercial food stuffs, there are substantial, built-in safety margins.
Rogan's analytical methods viewed risk from:
- EXPOSURE to these chemicals in human milk over the duration of breastfeeding,
- CARCINOGENICITY of the various chemicals,
- LIFE EXPECTANCY based on death rates from cancer attributable to the chemicals,
- POSTNEONATAL MORTALITY differences between breastfed and bottle-fed children, including the well-documented two-and-a-half times increased risk of death in the first two years of life from not being breastfed,
- CHANGE IN LIFE EXPECTANCY between breastfed and bottle-fed children, assuming that breastfed children have excess cancer risk from the contaminants.
The results were quite striking. Estimated loss of life expectancy from cancer from human milk exposure to the contaminants studied was less than three days; in contrast, the decrease in life expectancy from excess postneonatal mortality in infants not breastfed compared to the increase in life expectancy in breastfed infants is about seventy days. There was not only no advantage to avoiding breastfeeding, there was a disadvantage. Breastfed children, even those at the extreme doses of contaminants considered to be in human milk, had a net lower risk of death and longer life expectancy.
Rogan concluded that at current levels, the estimated cancer risks associated with contaminants in breast milk do not outweigh the benefits of breastfeeding. Because of these results, he also questioned the usefulness of testing human milk for contaminants in the general population, although he felt that individual women facing occupational or accidental exposure from PCBs in particular might consider testing.
In 1992 in NeuroToxicology, Rogan once again reviewed what had been learned in the preceding years about the neurotoxicology of PCBs and related compounds. Information about breastfed children concluded that despite the presence of PCBs in breast milk, no association between breast milk exposure and any measured outcome has been seen other than lower activity levels at four years among long-term breastfed children at the highest PCB levels.
And finally, in 1993 in Early Human Development Rogan examined breastfeeding and cognitive development in the children he had followed for so long. What he discovered was that the breastfed children, even though they had received higher contaminant loads through human milk than the children who had been bottle-fed from birth, scored significantly higher on the Bayley and McCarthy mental and psychomotor development tests (at all time points from two years through five years) and had higher English grades on report cards from grade three or higher. There were also trends toward higher scores with increasing length of breastfeeding. Rogan concludes that breastfeeding, even with known contaminants in human milk, offers the potential for enhancing the children's development within the population at no risk and little cost.
So, in answer to one of my original questions--can breastfeeding still be recommended for infants in this contaminated world.? Yes, and the research is there to confirm that answer.
But what about the contaminants found in the infant feeding alternative, commercial infant formula?
The Rogan series of studies cited earlier also analyzed cow's milk and infant formula that was being fed to infants and children not receiving human milk. No overall significant levels of pesticides and organochlorine contaminants were found, although individual samples may be higher.
Even though processing standards and contaminant limits have been set in the USA, remember that cows are part of the contaminated environment, eating grasses, grains, and other feed potentially treated with pesticides and drinking water potentially contaminated with a variety of chemicals. Dairy cattle are also treated with a variety of drugs, including antibiotics, and these can find their way into the milk supply as well as potentially into infant formulas which are predominantly cow's milk-based.
One set of compounds not yet discussed are the heavy metals, such as lead, cadmium, and mercury. The heavy metals bind more readily to human milk proteins as opposed to the previously described pesticides and organochlorines which bind to lipids.
In general population exposure, lead levels in human milk are normally lower than the lead levels in milk-based infant formulas. This is especially true when concentrated infant formulas are mixed with water containing high lead levels or packaged in lead-soldered cans. There are generally higher blood lead levels in formula-fed infants than breastfed infants, too. The February 1994 issue of Pediatrics takes an extensive look at the lead crisis in infants in general, including the continuing controversy over the new lower blood lead concentration limit recently set.
Cadmium levels in breast milk are about the same as in cow's milk. Please note, however, that cadmium and DDT levels are higher in the breast milk of smokers.
Mercury levels typically are lower in human milk than those of lead and cadmium. The highest levels have been found in the milk of fish eaters, particularly sport fish. Concern is also raised from time to time about the contribution of mercury to human milk from silver amalgam dental fillings. This bears research investigation.
Much of the preceding information comes from a highly respected reference text entitled Chemical Compounds in Human Milk, edited by two world-renowned researchers in the field, Allan A. Jensen of the Danish Institute of Technology, Department of Environmental Technology, and Stuart A. Slorach of the Food Research Department at the Swedish National Food Administration. This text reviews the global contaminant picture and its impact on human milk to 1991 in terms of general population versus occupational exposure.
Other facts of interest from this book include the following:
Dietary intake of contaminants during lactation is not a significant source of contamination in breast milk; the contaminants come from the mobilization of adipose tissue into the lipids in human milk. During pregnancy, however, dietary intake may significantly affect maternal contaminant levels because of an increase in adipose tissue in conjunction with weight gain, as well as development of key organ systems in the fetus itself.
Contaminant levels are also dependent on personal exposure from a woman's location in the environment. There are higher levels of persistent organochlorines in women living in coastal regions, presumably from eating more readily available fish. However, women who consume great quantities of beef and dairy products should also be aware of concerns.
There are lower levels of DDT and DDE in the milk of vegetarian mothers, but stable PCB levels indicate that contaminant sources other than food are involved, for example air pollution outdoors and indoors.
The following quotation sums up the risk-benefit considerations as summarized in this authoritative text:
Reports in the mass media about the presence of toxic chemicals in breast milk often alarm mothers, who wonder whether they should get their milk analyzed and, if the levels are relatively high, stop breastfeeding. Virtually all national and international expert committees have hitherto concluded--on the basis of available information--that the benefits of breastfeeding outweigh the possible risks from chemical contaminants present in human milk at normal levels. There is little point in investigating milk from individual concerned mothers, unless there are indications of abnormally heavy exposure, for example, at the workplace or due to poisoning incidents.
Gladen, B. C. et al. Development after exposure to dichloroethene polychlorinated biphenyls and dichlorodiphenyl transplacentally and through human milk. J Pediatr 1988; 113(6):991-95.
Gladen, B. C. and W. J. Rogan. Effects of perinatal polychlorinated biphenyls and dichlorodiphenyl dichloroethene on later development. JPediatr 1991; 119(1pt1):58-63.
Jensen, A. A. and S. A. Slorach, eds. Chemical Contaminants in Human Milk 1991; Boca Raton, Florida: CRC Press, Inc.
Rogan, W. J. et al. Pollutants in breast milk. N Engl J Med 1980; 302(26):1450-53.
Rogan, W. J. et al. Polychlorinated biphenyls (PCBs) and dichlorodiphenhyl dichloroethene (DDE) in human milk: effects of maternal factors and previous lactation. AJPH 1986; 76(2):172-77.
Rogan, W. J. et al. Neonatal effects of transplacental exposure to PCBs and DDE. J Pediatr 1986; 109(2):335-41.
Rogan W. J. et al. Polychlorinated biphenyls (PCBs) and dichlorodiphenyl dichloroethene (DDE) in human milk: effects on growth, morbidity, and duration of lactation. AJPH 1987: 77(10) :1294-97.
Rogan, W. J. and B. C. Gladen. PCBs, DDE, and child development at 18 and 24 months. Ann Epidemiol 1991; 1(5):407-13.
Rogan, W. J. et al. Should the presence of carcinogens in breast milk discourage breast feeding? Reg Toxicol Pharmacol 1991; 13:228-40.
Rogan, W. J. and B. C. Gladen, Neurotoxicology of PCBs and related compounds. NeuroToxicology 1992; 13:27-36.
Rogan, W. J. and B. C. Gladen. Breast-feeding and cognitive development. Early Human Dev 1993; 31:181-93.
Wolff, M. S. et al. Blood levels of organochlorine residues and risk of breast cancer. JNCI 1993: 85(5):648-52.
Minimizing Contaminant Risks During Pregnancy and Lactation