The semiconductor industry is one of the most chemical-intensive industries in the modern world. Because the least bit of dust may ruin the manufacture of an expensive semiconductor, the electronics industry developed the cleanroom where dust levels are kept very low. To save energy costs, and to lower dust levels to near zero, for years filtered air was recirculated in the cleanrooms. As a result, workers were repeatedly and continually exposed to recirculated mixtures of chemicals because the the fumes and vapors of many hundreds of chemicals were not removed by air and particulate filters.
Although hundreds of chemicals have been used in this industry, OSHA only requires that a few common solvents be monitored in the cleanrooms. While these measurements occasionally may exceed threshold level values and permissible exposure levels, there is no required monitoring for “total organics”and no industrial hygiene monitoring of carcinogens and chemicals that cause birth defects.
The first public report of the health of semiconductor workers was by the California Department of Industrial Relations in 1981. Wade et al. 1981. Participating companies carefully limited their cooperation, but the study exposed a list of carcinogens used in semiconductor chip manufacture. The list included arsenic compounds, asbestos, beryllium and its compounds, carbon tetrachloride, chromium, and nickel. The industry made sure there was no discussion of the nature or extent of worker exposure to carcinogens, and this very limited survey provided no industrial hygiene data or health data.
In 1986, a survey [Garabrant et al. 1986] extended the list of known carcinogens to include benzene, chloroform, dichloromethane (methylene chloride), 1,4-dioxane, tetrachloroethylene, trichloroethylene, epichlorohydrin, formaldehyde, cadmium and cadmium compounds. Epoxy resin systems widely used in the encapsulation of chips added more chemicals to the list of carcinogens to which semiconductor workers were exposed: glycidyl ethers, ethylene oxide, propylene oxide, 3,4-epoxy-1-butene, 2,3-epoxyhexadecane, m-phenylenediamine (MPDA), 4,4’methylenedianiline (MDA), and diaminodiphenylsulfone (DADDPS, DDS, dapsone), o-toluidine, 2,4-toluene diamine, methylenebis-o-chloroaniline (MOCA), benzidine, urea-formaldehyde, and quinones.
In 1999, the seriousness of exposures suffered by cleanroom workers was confirmed by U.S. Bureau of Labor Statistics data which showed that 2.4 percent of the workloss cases for workers in all manufacturing industries were the result of “exposure to caustic, noxious, or allergenic substances,” but the corresponding rate for the electronics industry was 6.0 percent and for semiconductor workers it was an astounding 9.7 percent.
Actually the real numbers are much higher, since a study of the reporting of occupational illnesses in California found that semiconductor companies only reported less than half of all cases that should have been reported by OSHA criteria. McCurdy et al. 1991.
These findings are important because research has proven that cancer is caused by chemical exposures, not heredity. An extraordinarily reliable “matched pair” study of 89,576 twins showed that environmental exposures tremendously outweigh the role of heredity in causing cancer and that inherited genetic factors play a very minor contribution to a person’s susceptibility to cancer. Lichtenstein et al. 2000. Chronic exposures particularly in the workplace, on a day to day basis over a period of years, are the most significant cause of cancer, not the workers’ genetic code. See also Selevan 1991.
REPRODUCTIVE OUTCOMES
Reproductive toxicity is of special interest because chemicals that cause miscarriages and birth defects also are carcinogens. Eisses 1999.
After IBM and Fairchild Semiconductor contaminated the public water supplies of the Great Oak Water Company in San Jose, the state Department of Health Services found widespread leakage of solvents and other chemicals from underground chemical storage tanks into the ground water. An initial epidemiological study of reproductive outcome measured in the population who drank contaminated water showed increases in the rates of spontaneous abortion and congenital malformations of infants exposed during pregnancy. Rudolph et al 1986.
Since the large majority of semiconductor workers are women of childbearing age, the risk of adverse reproductive outcomes was examined among workers at a Massachusetts semiconductor company because semiconductor workers have much greater exposure to chemicals than those exposed through drinking water. Personal interviews were conducted with manufacturing workers, spouses of male workers, and an internal control group of non-manufacturing workers. Pastides et al. 1988. Elevated rates of spontaneous abortion were observed for women working in clean rooms (31.3 abortions per 100 pregnancies for photolithography workers, 38.9 for diffusion workers, and only 17.8 for unexposed women).
IBM, then the largest manufacturer of integrated circuits, engaged the School of Hygiene and Public Health at Johns Hopkins University to study reproductive problems among its employees. The retrospective portion of the study, conducted at facilities in New York and Vermont, was reported in 1992. The Johns Hopkins study showed an increased rate of spontaneous abortion among women who worked in two specific cleanroom areas. Gray 1993; Corn et al 1993; Correra et al. 1996.
The Semiconductor Industry Association (SIA) sponsored research at the University of California, Davis to conduct a retrospective cohort study of 6,088 women. In this group there were 904 eligible pregnancies ascertained by completion of a detailed telephone interviews and 113 of these resulted in a spontaneous abortion eligible for inclusion in the analysis. The crude risk ratio for women working in fabrication areas vs. non-fabrication areas was a significant 1.45 (95% CI 1.02-2.05). This reduced slightly to 1.43 (95% CI 0.95-2.09) after adjustment for various potential confounders.
This study shows that a reproductive risk is associated with semiconductor manufacturing work. Schenker et al. 1992 and 1995.
Worse, this excess of miscarriages occurred in settings where industrial hygiene air measurements were in compliance with current occupational standards.
This research raises significant alarms: present standards are inadequately to protect workers, the routes of exposure not included in the standards are significant and must be addressed, relevant agents must not measured with specificity, and agents obviously are acting in dangerous synergy.
Synergism is the simultaneous action of separate agencies which, together, have greater total effect than the sum of their individual effects. The phenomenon is routinely seen in the interaction of drugs, such as alcohol and tranquilizers, but never mentioned on Material Safety Data Sheets required for the sale of toxic chemicals by the Environmental Protection Agency.
The U.S. Environmental Protection Agency in its 20 years of existence has operated on a grossly erroneous premise which is totally divorced from reality. The EPA assumes that dangerous chemicals only are used individually and never in mixtures. As a result, the synergistic effect of cleanroom chemicals that are used in mixtures is not a subject of warning, preventative efforts or investigation by the ostriches of the EPA.
CANCER
Semiconductor workers are subject to a risk of occupational cancers in parallel with the risks of occupational reproductive effects.
A 1983 report evaluated the general cancer incidence pattern in the electronics industry. Vagero et al 1983. It used the Swedish Cancer Environmental Registry, which was created by linkage of the 1960 census to the Swedish Cancer Registry of 1961 to 1973. All subjects who were classified in the census as working in the electronic or electrical manufacturing industry in 1960 were compared with the general population aged 15 to 64. The study population included 54,624 men and 18,478 women;1855 and 1009 cancers were reported, respectively. The control population contained more than 3 million individuals. The relative risk estimates for lung, bladder, and malignant melanoma for electronics industry employees were significantly increased to 1.52, 1.22, and 1.35, respectively.
In the Swedish Registry study a subpopulation of workers in the electronics industry was further analyzed for cancers of the mouth, pharynx, and respiratory system. Among males the incidence of lung tumors was significantly elevated (RR=1.36). There were 13 cases of pharyngeal cancers giving a risk estimate of 3.0. In a subgroup composed of workers who largely held assembly jobs, there were five nasal cancers, representing a risk increase of more than fourfold.
In 1985, a chemist working in the Material Analysis Department at the IBM research facility in San Jose, California wrote a memo to IBM Corporate Headquarters. The memo alerted IBM officials to a cluster of cancers in his colleagues. Among the group of 12 workers in a research and development laboratory, two died of brain cancer, two died of lymphatic cancer, and two died of gastric cancers. When two more developed bone cancer, the survivors sought to bring IBM’s attention to the issue. While such clusters are notoriously difficult to evaluate, this set of events was particularly striking, and IBM began an internal study of cancer. Unfortunately, they have not published their findings, though the study has been available to IBM researchers and consultants for many years.
IBM also commissioned a study, published in 1996 of brain cancer mortality among electronics workers. The IBM Corporate Mortality File cited in the study report presumably recorded deaths for all U.S. employees of IBM, from outside sales agents never exposed to chemicals to the bunny-suited workers in the clean rooms. The study had severe limitations prompting the authors to state that “(s)ome of the observed associations are difficult to interpret because exposure information pertaining to division and job groups is lacking.” Beall et al. 1996.
This study found that mortality from brain cancer among male electronics workers sloped upward as duration of employment lengthened. This is consistent with trends previously observed in the scientific literature: namely, that the risk of dying from brain cancer is highest among electrical and electronics workers with long-term work histories – specifically, of 10 years or more — and with probable exposure to solders and organic solvents. Thomas et al. 1987.
The 1987 study found that the risk of astrocytic tumors among electronics manufacture and repair workers was increased tenfold among those employed for 20 or more years. The authors pointed out that, “Numerous solvents used throughout the electrical and electronics industry are known neurotoxins, causing peripheral neuropathy, central nervous system depression, and neurobehavioral dysfunction.” This study called for epidemiologic studies and for screening of new and existing materials for mutagenic activity prior to worker exposure.
The US Environmental Protection Agency (EPA) sponsored a program known as the Common Sense Initiative (CSI) that sought to involve various industries in the planning and implementation of health research projects. The EPA, working with California’s Department of Health Services, developed a broad-based community consensus proposal to utilize California’s Health Registries to study the rates at which disease occurs among electronics workers and their families.
The purpose of the study was to develop a record-keeping system with which the computer and electronics industry could monitor incidence rates for cancers, birth defects, and other health parameters.
Led by representatives of IBM, an industry group defeated the proposal in 1998 and gave the EPA little hope of further cooperation. LaDou et al 1998.
In December, 2001 the Health and Safety Executive (HSE) in the United Kingdom announced the results of its study of cancer rates in a small sample of workers (71 deaths) at the National Semiconductor (UK) plant at Greenock, Scotland. Although the sample size was small, the results substantially reinforce the concerns that prompted the investigation and suggest a work-related cause for several kinds of cancer. HSE identified a higher than expected incidence of three particular types of cancer among women in the work force and one type in men. Bailar et al. 2000.
There were 11 cases of lung cancer in women – two to three times as many as expected. Of the 11 female lung cancers, five had a latency of less than 10 years, with ages at diagnosis ranging from 46-55. Five more had a latency of 10 to 20 years, with a range of ages at diagnosis of 44 to 61. One woman had a latency of over 20 years and was aged 60 at diagnosis. Only one of the female lung cancers was non-fatal. There were two cases of lung cancer in men, about half as many as expected. One of the two male lung cancers was nonfatal, the other patient died from a myocardial infarction. Information on smoking was not available to HSE’s investigators.
There were three cases of stomach cancer in women – four or five times as many as expected. All the women with stomach cancer were diagnosed between five and 10-years after first starting work at the factory and all were aged less than 45 at diagnosis. Two of the cancers were fatal. No cases of stomach cancer in men were reported. Because of the small numbers of cases, the implications of these finding remain somewhat less certain
There were 20 cases of breast cancer among women – five more cases (about 30 per cent more) than would have been expected: Four of the 20 breast cancers cases had died by the end of 2000. Ten of the cases had a latency of less than 10-years, with ages at diagnosis ranging from 31 to 60.
There were three deaths from brain cancer among men – about four times as many as would be expected. Three of the four male brain cancers had died by end of 2000. Three cases had a latency of less than 10 years with ages at diagnosis of 23, 35 and 56 respectively. The nonfatal case was aged 32 at diagnosis.
The HSE study is significant because even though a very small group was studied, a substantial number of them had little or no exposure to the chemicals of concern. It is therefore extremely remarkable that given this dilution, four apparent excesses in cancer were found, in a study with a weak design and a total of only 71 deaths.
Although many American semiconductor companies are large enough to conduct cancer epidemiology studies of their own workers, the industry has not initiated any studies and there has been no adequate study of cancer in the semiconductor industry anywhere in the world even though one-third of all semiconductor company employees work as chemical handlers, maintenance workers, or as production workers in the cleanrooms, where chemical exposures occur.
The reluctance of the industry to conduct such studies in order to protect its workers is deplorable, particularly in light of the failure to study the synergistic effects and health impact of multiple chemicals used in manufacturing and the high probability of substantial harm being inflicted on workers as studies have shown to date.
ETHYLENE GLYCOLS: ONE CHEMICAL STUDY
Many of the chemicals used in the microelectronic manufacturing process have been the subject of numerous animal studies, which the electronics industry has ignored. As proof of that claim, the history and scientific literature concerning ethylene glycol is particularly instructive.
First, the hazards of ethylene glycol have been known for decades. Yet, from the days it was first suspected as a mutagen, carcinogen and teratogen, Material Safety Data Sheets published by chemical manufacturers have failed to comment on its dangerousness.
Second, despite known risks of harm and recurring cancers in workers and birth defects in their children, the pursuit of profits has prevailed in the electronics industry. Preventative exposure efforts have been nil and not one chemical manufacturer, industry trade organization, or chemical society has called for human epidemiological studies for workers exposed to ethylene glycol.
Third, the tragedy of chemical deaths and birth defects has been expanded from its home in the United States to impoverished economies where the needs of governments for jobs far outweighs the capacity of health and safety resources and controls. That these chemicals caused the same injuries no matter where they are used is evident in current litigation involving of IBM families from San Jose, California, Rochester, Minnesota, Burlington, Vermont, East Fishkill, New York and Corbeil-Essonnes, France, and in National Semiconductor families in Santa Clara, California and Greenock, Scotland
Exposing workers to ethylene glycol is illustrative of how the industry injures its workers and ignores the aftermath of its wrongdoing.
Ethylene glycols [including ethylene glycol monomethyl ether (EGME), also known as 2-methoxyethanol and methyl cellosolve; ethylene glycol monoethyl ether (EGEE), ethylene glycol monoethyl ether acetate (EGEEA), also known as cellosolve acetate] and the products that result from their metabolization, damage the male and female reproductive systems and have a profoundly toxic effect on human and animal development. Hardin 1983.
For decades, evidence of the consequences of exposure to EGME, a colorless, odorless gas used in a variety of industrial and chemical processes, has been reported in numerous scientific publications. See e.g. Stenger et al. 1971; Nagano et al. 1979; Hardin et al. 1981; Miller et al. 1981; Rao et al. 1983; Brown 1984; World Health Organization 1990.
Atrophy of the testicles, infertility, degenerative changes in spermatocytes, and decreased survival rates in male rats exposed to EGME in utero have all been documented in the world scientific literature for two decades. Nagano reported in 1979 that male rats exposed to EGME suffered dose-related decreases in testicular weight and damaging sperm. In 1983 Foster reported significantly decreased testicular weights in rats exposed to EGME, with degeneration of spermatocytes as soon as 24 hours after a single dose at the highest dose tested. Degenerative changes were noted in spermatocytes and examination of testes 24 hours after a single dose of EGME showed spermatocytes with swelling and disruption.
Further studies (Chapin et al. 1984) showed an even broader range of spermatocytes targeted by EGME and in 1985 reported that unexposed female rats mated with male rats exposed to EGME had 1) a dose-related reduction in pregnancies and 2) that the number of live fetuses per pregnant female was significantly affected in a dose-related fashion. Chapin also reported in 1985 that rats exposed to EGME had significantly reduced sperm counts, and a substantial increased in the percentage of abnormal sperm. The same authors further reported in 1985 that at doses designed to be at “no effect” levels, EGME altered sperm concentration and morphology. In 1987 Anderson reported significant increases in abnormal sperm morphology at all doses of EGME given to mice in an oral study. See also Oudiz et al. 1986.
When doses of EGME have been administered through inhalation, the gas has been shown to cause testicular damage in mice (Miller et al. 1981), rats (Miller et al. 1983), and rabbits (Miller et al. 1983).
In several studies of animal reproduction, EGME vapors caused significant decreases in male rat fertility (Rao et al. 1983) and pronounced testicle atrophy (Doe et al. 1984). In 1984, Samuels et al. reported that a single inhalation of EGME produced significantly reduced testes weights in test rats compared to controls, and in a second study documented that certain rat testes cells were damaged four days after EGME exposure. See also Kumagai et al. 1999; Shih et al. 2000.
EGME also produces damaging effects when absorbed through the skin. Hardin et al. 1984. A 1986 study by Hobson et al. notes severe testicle atrophy in male guinea pigs that had EGME applied to their backs for six hours. When the animals were examined 13 weeks after application, all had pronounced wasting of the testicles and a complete loss of spermatogenic cells.
The significance of the results of these studies is only increased by the results of studies of the effects of EGME on humans.
In a September 2000 report on 32,442 men who had their semen analyzed at the Sperm Analysis Laboratory in Copenhagen between 1963 and 1995, the effects of EGME exposure were shown to be low semen concentration, poor sperm motility, and a high proportion of abnormal spermatozoa, all of which are associated with an increased risk of testicular cancer (Jacobsen et al. 2000).
In addition to reproductive system damage, EGME has also been shown to cause other severe malformations as well as birth defects in both man and animal. These range from ocular defects to skeletal anomalies, craniofacial anomalies, cardiac malformations, kidney defects, missing and extra fingers and toes, and missing and foreshortened limbs. Holmberg et al. 1980,1982 and 1986; Kucera 1968.
Studies published from the 1970s forward show that EGME and methoxyacetaldehyde MAA, a product that results from the metabolization of EGME, both produce frequent, severe malformations in a dose-dependent way.
In 1981, Nagano et al. reported on the toxic effects in fetuses exposed to EGME during their mother’s pregnancy. Out of 130 live fetuses, 57 were born with abnormalities: 24 had exencephaly (brain residing outside the skull), 3 had an umbilical hernia, 29 had one or more abnormal fingers, and 1 had both exencephaly and an abnormal finger. There was also a significant increase in anomalies of the vertebrae, including spina bifida, a curvature of the spine, and fused ribs.
External bodily sites that EGME targets are wide ranging: the limbs and digits, the abdominal wall, eyes, and tail. Digit abnormalities that result include the absence of nails on fingers or toes, additional finger or toe nails, missing digits, and fused digits. Limb anomalies include, the inward rotation of the hind limbs and joints remaining in a flexed position.
Damage to soft tissue areas resulting from EGME exposure is also widespread. Target sites include the vascular system, the kidneys, and the ovaries. Tikkanen1988. Specific cardiovascular anomalies include ventricular membrane defects and a narrowing of the aorta, while renal defects include misshapen kidneys and a dilated renal pelvis. Ovarian abnormalities include hyperplastic ovaries, which are enlarged due to an abnormal increase in cells.
Animal studies also indicate numerous target sites for EGME-caused skeletal malformations. In one study, fetuses from the same litter had shortened nasal, maxillary, and mandible bones; in another, 11 fetuses from 5 litters had missing bones; and in a third, 4 fetuses from 3 litters had extra cervical ribs and shortened, fused, forked, and calloused ribs. In a study of rats and mice, EGME produced a significantly increased incidence of delayed ossification as well as extra lumbar ribs. In still another report, male mouse offspring exposed to EGME showed a significant increase in the incidence of a hypoplastic, or cell-deficient, testicle (8 fetuses in 6 litters). Also in 1984 Brown reported on the teratogenicity of the metabolite of EGME–methoxyacetic acid–in rats. The compound was fetotoxic and teratogenic, with skeletal malformations, hydrocephalus and urogenital abnormalities being the most common malformations.
In 1987, Hardin et al. reported that EGME, ethylene glycol dimethyl ether, and diethylene glycol dimethyl ether all produced a significant incidence of paw defects in mice, but EGME produced the highest rate: 87.5% of the litters and 68.5% of the fetuses treated with EGME had paw defects. HInd paw defects predominated, with fused digits the most common.
Other studies also document severe EGME damage to animal embryos. In 1987 Greene et al. found that mouse embryos whose mothers were given EGME orally had malformed forelimb buds and extensive digit malformations. In a 1989 study of macaque monkeys, Scott et al. concluded that EGME is toxic to the developing primate embryo even at a relatively low dose, supporting the evidence that all animal studies point to EGME as a human teratogen. The Scott study is noteworthy because it found that a dead embryo from the high-dose group was missing one digit on each forelimb, an anomaly not seen in macaques that hadn’t been exposed to EGME and previously reported in mice (Horton et al. 1985), rats (Ritter 1985), and rabbits (Hanley et al. 1984).
A metabolite of EGME of particular toxicologic interest is methoxyacetic acid (3MAA2). In fact, it is believed that the similarity in malformations caused by EGME, ethylene glycol dimethyl ether, and diethylene glycol dimethyl is linked to these gases’ metabolic conversion to 3MAA2.
MAA induces chromosomal aberrations in human lymphocytes and, along with EGME, has the ability to disrupt the endocrine system. Both MAA and EGME also target ovarian luteal cells in humans at the same concentrations as they do rat luteal cells, which results in the disruption of menstrual cycles and the inhibition of ovulation. Davis et al. 1997. The targets of EGME and its metabolites in developing organisms are both dividing somatic (non-sexual reproductive cells) and germ (reproductive) cells.
EGEE causes severe malformations, embryotoxicity, fetotoxicity, and birth defects in every animal species tested.
EGEE also impairs sperm production, producing abnormal sperm forms and significant decreases in sperm numbers. Exposed male rabbits demonstrated reduced testes weight and degeneration of seminiferous tubules (Barbee et al. 1984). In a study designed to determine the testicular cell types most susceptible to EGME, the spermatocyte is the primary taraget cell for the histologic effects of EGME in the testis of rats. The animal data show that the frequency and severity of malformations increase as the dosage increases no matter how animals are exposed.
Inhalation studies also show severe fetal damage caused by EGEE. Pregnant rabbits exposed to EGEE had dose-dependent adverse outcomes including significant reductions in the number of live fetuses at lower doses. Hardin et al. 1981. Fetal examinations revealed a significant increase in the incidence of renal, cardiovascular, and abdominal wall defects as well as significant increases in skeletal abnormalities. In studies of female rats exposed to EGEE before and after breeding, all litters produced by females given the highest dose were resorbed (total fetal loss); fetuses produced by females given a lower dose showed significant reductions in fetal body weight and crown-rump length. The lower-dose group also demonstrated a significantly increased incidence of cardiovascular anomalies, such as transposed arteries and skeletal defects.
The above studies, and others, confirm the developmental and reproductive toxicity of ethylene glycols, yet people in industrial situations are still being exposed to these and other chemicals simultaneously. Being exposed to multiple developmental toxins involves a higher risk than exposure to one toxin alone, and is evidenced in more frequent harmful outcomes and a significant increase in the severity of effects. This synergistic phenomenon is readily seen in the effects caused by exposure to a wide range of industrial chemicals, and is well recognized in medical circles.
For instance, the synergistic effect of exposure to both the chemical teratogen EGME and radiofrequency (RF) radiation has been well demonstrated. In one study (Nelson et al. 1991), mean fetal malformations/litter were seen in 14% of fetuses exposed only to EGME, in 30% of fetuses exposed only to RF radiation, but in 76% of fetuses exposed to both. All 18 litters exposed to both EGME and RF radiation were affected, with a mean of 12 fetuses affected per litter. Paw and tail malformations prevailed, with the most severe paw malformations occurring in the combined exposure group.
Had there been no synergy between EGME and RF radiation, the percent of malformed fetuses in the RF-only group plus the percent of malformed fetuses from the EGME-only group would not have differed significantly from the percent in the group with combined exposures.
However, the study clearly demonstrated significant synergy. Not only was there a great increase in the frequency of malformations, but the malformations in the combined exposure group were the most severe.
Yet another study, the pesticide research of Arnold et al. 1996, proves that hormone-disrupting chemicals that are known to cause mild effects individually produce dramatic hormonal effects when used in combination. Chemicals that separately do not disrupt hormones tremendously magnify the ability of other chemicals to do so, and in combination can be up to 1,600 times more powerful than any chemical alone.
Because industrial solvents have identical or structurally similar metabolites that compete for the same metabolic pathways, simultaneous exposure to a mix of these compounds can threaten to overwhelm the body’s metabolic processing ability, increasing the risk of adverse effects from such exposures. For example, the increased danger of alcohol consumption to a person in the process of metabolizing inhaled organic solvents is a well-recognized phenomenon. Additionally, when humans are exposed to chemicals through both inhalation and the skin, the potential for harm is greater than when there is only one route of exposure.
The animal data cited above are relevant to human outcomes for several reasons. For one, many reproductive processes are similar across mammalian species; animal developmental toxicology studies are thus widely recognized in the scientific community for their predictive value. A second reason is that animal studies demonstrate that developmental toxins produce specific malformations in a dose-dependent fashion. They provide an excellent model for the effects of these chemicals on humans, and their results help the medical and scientific communities to evaluate the link between chemical exposure and birth defects.
The process of applying animal data to humans involves a number of considerations, one of which is to compare the relative sensitivity of humans to the sensitivity of the animal species being studied. One way to approach this issue is to consider acute maternal toxicity, that is, the point at which a mother is harmed directly by the chemical exposure. Another consideration is the concentrations at which the compounds that induce birth defects in animals harm the fetus without causing overt maternal toxicity. If the acute effect level for a compound is lower for humans, then the latent effect level for that compound will also be lower.
Convincing evidence exists that acute maternal toxicity in animals occurs at much higher levels than it does in humans; humans are considerably more sensitive to many organic solvents than animals. For example, methanol concentrations that produced malformations in animal offspring without producing maternal toxicity were in the range of 20,000 parts per million. But sustained exposure to the same level would most certainly produce significant evidence of maternal toxicity in pregnant women, showing that the human female is probably more vulnerable to the toxic, fetotoxic, and teratogenic effects of methanol than the rodents used in the study. Similar findings exist regarding methyl cellosolve; human data on glycol ether exposure demonstrate that cellosolves have the potential to cause developmental toxicity in humans. One example of this involves a child born with cataracts and soon diagnosed with Hallermann-Streiff syndrome. At the time of her conception, her father loaded and delivered glycol ethers made by chemical manufacturers and suppliers.
It is known that men exposed to EGME and EGEE have an increased incidence of reduced sperm count. This indicates that glycol ethers, and EGME and EGEE in particular, can damage the human reproductive system (Welch et al. 1988, NIOSH 1986). Children whose mother worked with glycol ethers were born with hypospadia [malformed penises] in a case report by Bok in 1990.
In 1992, Cordier et al reported the results of a multicenter case-control study of 984 cases of major congenital malformations. The study demonstrated a significant association between maternal exposure to glycol ether solvents and congenital malformations. In 1997, Cordier also found significant associations between glycol ether exposure and multiple malformations. Of the 984 cases in the study, 100 had cleft lip/cleft palate.
In 1992, Johns Hopkins University first reported the results of a multiyear, multimillion-dollar epidemiologic study conducted among workers at IBM plants in East Fishkill, New York, and Essex Junction, Vermont. Correa et al. 1996. The study documented the presence of potent reproductive and developmental toxicants at both sites. IBM itself reported the study’s findings to the Environmental Protection Agency: a significant elevation in the rate of miscarriages and a significant delay in conception among women working with ethylene glycol ethers and related solvents. A similar trend was found among the wives of exposed male workers (Correa et al. 1996), and a study of the semiconductor industry showed a correlation between working around ethylene-based glycol ethers and adverse pregnancy outcomes (Schenker et al. 1995). A previous study (Pastides et al. 1988) also reported on an elevated spontaneous abortion rate among semiconductor fabrication workers at Digital Equipment Corporation. See also Khattak et al. 1999; Windham 1991.
The Johns Hopkins study looked at conceptions that occurred during two periods: 1980 to 1989 and 1990 to 1992 and found a pattern of exposure to ethylene glycol ethers affecting pregnancy outcomes during both time frames. Exposure to ethylene-based glycol ethers and related compounds (xylene and n-butyl acetate) had significant biological consequences for those exposed, including an increased risk of miscarriage and subfertility.
As a matter of reproductive epidemiology, when one or more adverse reproductive effects is significantly higher than expected in an exposed population, it is biologically plausible that other adverse reproductive effects may be attributable to the same cause. Thus, the subfertility and increased spontaneous abortion rates of the IBM workers are attributable to the same exposures as those causing elevated rates of malformations among the IBM employee offspring.
It is also well recognized that, depending on the level and timing of exposure to a given compound, exposure may lead to different outcomes. In animal studies, higher doses of developmental toxicants are required to induce fetal deaths (spontaneous abortions) than doses necessary to cause malformations in offspring. This well-known, biologically plausible phenomenon is well known in the field of animal toxicology.
The Johns Hopkins study demonstrated that the environment in which the IBM plant workers were employed was capable of inducing spontaneous abortions, and for that reason these poisons were present in sufficient amounts to cause birth defects.
The conclusions in the Johns Hopkins study are strong and reliable because of the consistency of the findings. This consistency, even more than the statistical findings, confirms the association between exposure to ethylene glycol ethers and related solvents and subfertility or spontaneous abortions.
Finding this result in two reproductive parameters — subfertility and miscarriage — further strengthens the conclusion that major malformations among certain survivors are attributable to the same risk factors.
During the 1980-89 period of the Johns Hopkins study, those male and female Fishkill and Essex Junction cleanroom workers who had the greatest opportunity for exposure to ethylene glycol ethers produced 69 living children. Based on background rates for major malformations in the population at large during that time, it would be expected that, at the most, two children would be born with major birth defects.
Unfortunately, the full count of major malformations or multiple malformations in the children born to the IBM cleanroom workers during the study period is unknown. So too is the count for the entire period these plants have been in operation, and for any segment of their operation. IBM has refused to produce any information it has regarding birth defects in the offspring of cleanroom workers, and has appealed judicial orders to produce this information. As a result, the only known cases are those of children whose parents have voluntarily come forward.
The information gathered shows that IBM cleanroom workers from East Fishkill, New York and Burlington, Vermont have reported 45 individual children with eye, gastrointestinal, genitourinary, reproductive, central nervous system, musculoskeletal, respiratory, cardiovascular, and chromosomal birth defects.
Of these self-identified birth defects among IBM cleanroom workers’ offspring, there are a number of exceedingly rare events. These events do not appear without biological basis and without sharing a common cause. The extraordinarily high level of birth defects is not coincidental; it is highly improbable that all of the events are unrelated. For these events to be truly random, and completely unrelated to the parents’ work environment at IBM, the universe of total live births would have to have been many times the size of this cohort.
Between 1980 and 1989, female cleanroom workers in the Johns Hopkins study reported 377 live births; wives of male cleanroom workers reported 576 pregnancies and 82 miscarriages. Adjusting for stillbirths, induced abortions, and pregnancies not yet at term produces a figure of approximately 825 live births for cleanroom workers during the period.
Though the Johns Hopkins team tried to be comprehensive, some live births were excluded from the study even though the parents met the study definition. In addition, the study did not systematically incorporate live birth outcome information for contract workers and for workers who left IBM during the decade.
By conservative estimate there were 1,000 live births to cleanroom workers including contract workers, temporary workers, and those leaving IBM prior to the start of the study. Though there are reasons to believe that the number of live births in the decades immediately before and immediately after the study was lower because of fluctuations in the IBM work force, it is assumed for purposes of this analysis that there were a total of 3,000 live births to cleanroom workers at the Fishkill and Essex Junction facilities during the 30-year plant operation period.
When compared to the general population, the known incidence of birth defects in the estimated 3,000 children born to IBM workers involved in the Johns Hopkins study is many times the norm, and is irrefutable confirmation that these defects were caused by exposure to ethylene glycol solvents.
Notes
* The history of the electronics industry reported here is the substantive work of John C. Bailar III, Ph.D., Department of Health Studies, The University Chicago, Chicago, Illinois; Joseph LaDou, M.D., director of the International Center for Occupational Medicine at the University of California School of Medicine, San Francisco, California; Robert Harrison, MD, Chief of Occupational Health Surveillance and Evaluation Program, California Department of Health Services, San Francisco and others, except for errors are which are the editor’s. The portion of this report on ethylene glycol represents the substantive contribution of Martyn Smith, Ph.D. Professor of Toxicology, Department of Biomedical & Environmental Health Services, School of Public Health, University of California., Berkeley, California, except as to errors which are the editor’s.
Anderson et al. 1987. Effect of ethylene glycol monomethyl ether on spermatogenesis, dominant lethality and F1 abnormalities in the rat and the mouse after treatment of Fo males. Teratogenesis, Carcinogenesis, Mutagenesis 1987: 7:141-58.
Arnold et al. 1996. Synergistic activation of estrogen receptor with combinations of environmental chemicals. Science 1996: 272:1489-92.
Bailar et al. 2000. Open Letter to the Health and Safety Executive. Int J. Occup Environ Health 2000: 6 (1):71-72.
Barbee et al. 1984. Subchronic inhalation toxicology of ethylene glycol monoethyl ether in the rat and rabbit. Environmental Health Perspectives 1984: 57:157-64.
Beall et al. 1996. Brain tumors among electronics industry workers. Epidemiology. 1996: 2:129.
Bok et al. 1990. Maternal exposure to ethylene glycol monomethyl ether acetate and hypospadia in offspring, a case report.” Br. J. Ind. Med. 1990: 47: 352-353.
Brown 1984. The teratogenicity of methoxyacetic acid in the rat. Toxicology Letters 1984: 22:93-100.
Chapin et al. 1984. The effects of ethylene glycol monomethyl ether on testicular histology in F344 rats. J. Androl 1984: 5 (5):369-80.
Chapin et al 1985. Effects of ethylene glycol monomethyl ether (EGME) on mating performance and epididymal sperm parameters in F344 rats. Fundam. Appl. Toxicol. 1985: 5 (1):182-9
Cordier et al. 1992. Maternal occupational exposure and congenital formations. Scand J. Work Environ Health 1992: 18:11-17.
Cordier et al. 1997. Congenital malformations and maternal occupational exposure to glycol ethers. Epidemiology 1997: 8 (4):355-63.
Corn et al 1993. Prospective Exposure Assessment Supplement to the Final Report, Johns Hopkins University School of Hygiene and Public Health, Baltimore, 1993.
Correa et al. 1996. Ethylene Glycol Ethers and Risks of Spontaneous Abortion and Subfertility. Am. J. Epidemiology 1996:143 (7):707-717.
Davis et al. 1997. Ovarian luteal cell toxicity of ethylene glycol monomethyl ether and methoxy acetic acid in vivo and in vitro. Toxicology and Applied Pharmacology 1997:142:328-37.
Doe et al. 1984. Comparative aspects of the reproductive toxicology by inhalation in rats of ethylene glycol monomethyl ether and propylene glycol monomethyl ether. Toxicology and Applied Pharmacology 1984: 69:63-67.
Eisses 1999. Concurrent teratogenic and mutagenic action of 2-methoxyethanol in drosophila melanogaster larvae resulted in similar phenotypes: close resemblance to directed mutations. Teratogenesis, Carcinogenesis, and Mutagenesis 1999:19:183-204.
Foster et al. 1983. Testicular toxicity of ethylene glycol monomethyl and monoethyl ethers in the rat. Toxicol. Appl. Pharmacol. 1983: 69 (3): 385-99.
Garabrant et al 1986. Carcinogens and cancer risks in the microelectronics industry. State of the Art Reviews: Occupational Med. 1986: 1:119-134.
Gray 1993. Final Report Retrospective and Prospective Studies of Reproductive Health among IBM Employees in Semiconductor Manufacturing, Johns Hopkins University School of Hygiene and Public Health, Baltimore, 1993.
Greene et al. 1987. Cytotoxic effects of ethylene glycol monomethyl ether in the forelimb bud of the mouse embryo. Teratology 1987:36:23-34.
Hanley et al. 1984. Comparison of the teratogenic potential of inhaled ethylene glycol monomethyl ether in rats, mice and rabbits. Toxicology and Applied Pharmacology 1984: 75 (3):409-22.
Hardin et al. 1981. Testing of selected workplace chemicals for teratogenic potential. Scand J. Work Environ Health 1981: 7 (4):66-75 and literature cited.
Hardin 1983. Reproductive toxicity of the glycol ethers. Toxicology 1983: 27:91-102.
Hardin et al. 1984. Developmental toxicity of four glycol ethers applied cutaneously to rats. Environmental Health Perspectives 1984: 57:69-74.
Hardin et al. 1987. Relative potency of four ethylene glycol ethers for induction of paw malformations in the CD-1 mouse. Teratology 1987: 35:321-28.
Health and Safety Executive (HSE) Investigation Into Cancer Rates in the Work Force at the National Semiconductor Plant at Greenock, Scotland, December 11, 2001. www.hse.gov.uk/statistics/nsukrept.pdf.
Hobson et al. 1986. A subchronic dermal exposure study of diethylene glycol monomethyl ether and ethylene glycol monomethyl ether in the male guinea pig. Fund Appl Toxicol. 1986: 6:339-48.
Holmberg et al. 1980. Congenital defects of the central nervous system and occupational factors during pregnancy: a case-reference study. American Journal of Independent Med. 1980:1:167-76.
Holmberg et al. 1982. Oral clefts and organic solvent exposure during pregnancy. Int Arch Occup Environ Health 1982: 50:371-76.
Holmberg et al. 1986. Solvent exposure and birth defects: an epidemiologic survey. Safety and Health Aspects of Organic Solvents, Alan R. Liss, Inc., 1986: pp. 179-85.
Horton et al. 1985. Developmental phase-specific and dose-related teratogenic effects of ethylene glycol monomethyl ether in CD-1 mice. Toxicology and Applied Pharmacology 1985: 80:108-18.
Jacobsen et al. 2000. Risk of testicular cancer in men with abnormal semen characteristics: cohort study. British Med. Journal 2000: 321:789-92.
Khattak et al. 1999. Pregnancy outcome following gestational exposure to organic solvents. JAMA 1999: 281 (12).
Kucera 1968. Exposure to fat solvents: a possible cause of sacral agenesis in man. J. Pediat. 1968: 72:857-59.
Kumagai et al. 1999. Uptake of 10 polar organic solvents during short-term respiration. Toxicol. Sci. 1999: 48 (2):255-63.
LaDou et al. 1998. The International Electronics Industry. Int. J. Occup. Environ. Health,1998: 4 (1):1-18.
Lichtenstein et al. 2000. Environmental and Heritable Factors in the Causation of Cancer –Analyses of Cohorts of Twins from Sweden, Denmark, and Finland. New England Journal of Med., July 13, 2000.
McCurdy et al. 1991. Reporting of occupational injury and illness in the semiconductor manufacturing industry. Amer. J. Public Health 1991: 81 (1):85.
Miller et al. 1981. Comparative short-term inhalation toxicity of ethylene glycol monomethyl ether and propylene glycol monomethyl ether in rats and mice. Toxicology and Applied Pharmacology 1981: 61:368-77.
Miller et al. 1983. Ethylene glycol monomethyl ether I. Subchronic vapor inhalation study with rats and rabbits. Fundamental and Applied Toxicology 1983: 3 (1):49-54.
Nagano et al. 1979. Japan J. Ind. Health 29 (in Japanese).
Nagano et al. 1981. Embryotoxic effects of ethylene glycol monomethyl ether in mice. Toxicology 1981: 20:335-43.
Nelson et al. 1991. Teratological assessment of methanol and ethanol at high inhalation levels in rats. Fundamental and Applied Toxicology 1991: 5:727-36.
NIOSH 1986. Health hazard evaluation report: Precision Castparts Corporation. HETA 84-415-1688.
Oudiz et al. 1986. In vivo and in vitro evaluations of spermatoxicity induced by 2-ethoxyethanol treatment. Toxicology and Applied Pharmacology 1986: 84:576-83.
Pastides et al. 1988. Spontaneous abortion and general illness symptoms among semiconductor manufacturers. Journal of Occupational Med. 1988: 30: 543-51.
Rao et al. 1983. Ethylene glycol monomethyl ether II. Reproductive and dominant lethal studies in rats. Fundamental Applied Toxicology 1983: 3:80-85.
Ritter 1985. Teratogenicity of dimethyloxyethyl phthalate and its metabolites methoxyethanol and methoxyacetic acid in the rat. Teratology 1985: 32:25-31.
Rudolph et al 1986. Reproductive hazards in the microelectronics industry. State-of-the-art reviews. Occup. Med. 1986:1:135–143.
Samuels et al. 1984. The effects on the rat testis of single inhalation exposures to ethylene glycol monoalkyl ethers, in particular ethylene glycol monomethyl ether. Arch. Toxicol. 1984: 7:167-70.
Schenker et al. 1992. Epidemiologic Study of Reproductive and Other Health Effects among Workers Employed in the Manufacture of Semiconductors, Final Report to the Semiconductor Industry Association, December, 1992.
Schenker et al. 1995. Association of spontaneous abortion and other reproductive effects with work in the semiconductor industry. Amer. J. Ind. Med. 1995: 28:639-59.
Scott et al. 1989. Teratologic potential of 2-methoxyethanol and transplacental distribution of its metabolite, 2-methoxyacetic acid, in non-human primates. Teratology 1989: 39:363-73.
Selevan 1991. Environmental exposures and reproduction. Reproduction and Perinatal Epidemiology, ed. M. Kiely, 115-30. Boca Raton, Florida, CRC Press, 1991.
Shih et al. 2000. Measurement of percutaneous uptake of 2-methoxy ethanol vapor in humans. Journal of Occupational Environmental Med. 2000: 42 (5):475-82.
Stenger et al. 1971. The toxicology of ethylene glycol monoethyl ether. Arsneim Forsch 1971: 21:881-85.
Tikkanen 1988. Cardiovascular malformations, work attendance and occupational exposures during pregnancy in Finland. Amer. J. Ind. Med. 1988: 14:197-204.
Thomas et al. 1987. Brain tumor mortality risk among men with electrical and electronics jobs. JNCI. 1987: 2:233-8.
Vagero et al 1983. Incidence of cancer in the electronics industry, using a new Swedish Cancer Environment Registry as a screening instrument. Br. J. Ind. Med. 1983: 40-:188-192.
Wade et al 1981. Semiconductor Industry Study. State of California, Division of Industrial Relations, Division of Occupational Safety and Health, Task Force on the Electronics Industry. 1981.
Welch et al. 1988. Effects of exposure to ethylene glycol ethers on shipyard painters: II. Male reproduction. Amer. J. Ind. Med. 1988: 14:509-26.
Windham 1991. Exposure to organic solvents and adverse pregnancy outcome. Am. J. Ind. Med. 1991: 20:241-59.
World Health Organization 1990. International Programme on Chemical Safety, Environmental Health Criteria, No. 115 (1990): 2-methoxyethanol, 2-ethoxyethanol and their acetates.
If you or a family member have been wrongfully injured call us at 1.888.777.1776 or use this form, delays can hurt your case, so please don’t hesitate to contact us.
