Document QXdz8e50xdYoVza6yRoqz1D2R

Ann *.1 t'hnimoi nl. T.n,, ,,! 1987 2787-III HUMAN HEALTH EFFECTS OF POLYCHLORINATED BIPHENYLS (PCBs) AND POLYBROMINATED BIPHENYLS (PBBs)1 Renate D. Kimbrough Center Tor Environmental Health, Centers Tor Disease Control, Public Health Service, US Department of Health and Human Services, Atlanta. Georgia 30333 INTRODUCTION Polychlorinated biphenyls (PCBs) are chemical compounds with the empirical formula C^io-nCI*. with n = 1-10. They are a mixture of chlorinated biphenyl congeners. Theoretically, 209 such congeners are possible, but at ' least 20 congeners have never been identified in commercial products. In addition. PCBs may contain polychlorinated dibenzofurans and chlorinated quaterphenyls as impurities. PCBs were discovered before the turn of the century, and the useful industrial properties of mixtures obtained by chlorina tion of biphenyl were recognized early. In 1966 the discovery of PCBs in environmental samples (I) spurred renewed interest in the analysis and toxic ity of these compounds. In recent years many industrial nations have taken steps to control the flow of PCBs into the environment. PCBs and PCB-containing formulations are restricted (an exception is sometimes made for mono- and dichloro-PCB) for most uses, except for categories such as closed-system electrical equipment and hydraulic fluids in mining equipment. Commercial production of PCBs began in the United States in the late 1920s. In 1971. Monsanto Chemical Company voluntarily stopped open- ended uses of PCBs, and subsequently only the lower chlorinated biphenyls The US Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper. 87 "Reproduced, with permission, from the Annual Review of Pharmacology and Toxicology. Volume 27, 9 1987 by Annual Reviews Inc." 6BRN0D200Q KK KIMHKOUGII were produced (Aroclor 1242 and 1016). In 1977 the company ceased produc tion entirely (2). Many PCBs manufactured in the past (3) are still in use in old transformers, but even this use is decreasing. The estimated cumulative production and consumption of PCBs in the United States in the period 1930-1975 (in millions of pounds) was as follows: total production, 1400; imports, 3; domestic sales, 1253; exports, 150. PCBs are inert chemicals that arc fairly resistant to degradation. Because of their stability and lipophilicity, they have accumulated in the environment and in organisms. They have been identified in indoor air (4) at concentrations of 0.1 pg/m5 (5), in flsh (6, 7) and other food products, and in sediments from lakes and rivers (8, 9). PCBs have also been identified at varying con centrations in soil (0.01 mg/kg-100 mg/kg) (10). They do not occur naturally. Thus, their presence in the environment is linked with human activities, and concentrations are higher in urban and heavily industrialized areas than in rural and remote areas. However, trace amounts are also found in remote areas, since the air may transport such chemicals over large distances. Although PCBs are no longer used commercially in the United Stales because of their persistence, they are still present in our environment. A number of transformers and capacitors that contain PCBs, however, are still in use. Results of laboratory experiments showed that pyrolysis of PCBs at temperatures of 200-600oC could result in the formation of significant amounts of the more toxic polychlorinated dibenzofurans (PCDFs) (II), In February 1981, an electrical fire occurred in a New York State office building in Binghamton, N.Y. The lire, which originated in a switch gear in the basement, caused the bushings to crack on a nearby transformer. About 180 gallons of PCB dielectric fluid Pyralon (65% Aroclor 1254,35% chlorin ated benzenes and trace additives) were lost. A fine layer of oily soot covered many of the internal surfaces of the 18 floors of the building. Analysis of a soot sample showed that it contained various isomers of chlorinated di benzofurans, 2,3,7,8-teirachlorodibenzodioxin, other chlorinated dibenzodioxins, and chlorinated biphenylenes. Some of these chemicals are much more toxic than the PCBs. Because of these findings the Binghamton stale office building was closed, and workers wearing respirators and pro tective clothing began an extensive cleanup of the building. The cleanup operations lasted four years, and the cost has been enormous (12). Since then several other transformer fires have occurred. These fires have not resulted in as much contamination as the one in Binghamton, partly because the transformers in the other fires were usually located in a separate vault (13). The problems with polybrominated biphenyls (PBBs), which are also a mixture of chemicals, have been quite different. In 1970 a chemical company in Michigan manufactured polybrominated biphenyls as flame retardants. The same company also produced magnesium oxide, a chemical commonly mixed HUMAN HI-ALTH HFIIiCTS OF IT'lls & PitUs K9 into Iced for livestock. The llamc retardant was culled Fircniaster and the magnesium oxide, Nutrimaster. In 1973 some bags of Fireniastcr were accidentally sold as Nutrimaster and mixed into animal feed. This resulted in widespread contamination in the slate of Michigan (14). Since 1974 PBBs have not been produced in the United Suites. At the time of the exposure little was known about the toxic effects of PBBs. Ten years after the Michigan residents were exposed, no clinical illness has been causally linked to PBB exposure in this group, although chloracne was apparently noted in some workers who manufactured PBB. ; HUMAN EXPOSURE Because PCBs arc ubiquitous and very persistent in the environment, humans have been and will continue to be exposed to them, particularly in in dustrialized countries. PCBs may be inhaled in small amounts through the air or ingested through food. In the United Stales today, people are primarily exposed to PCBs by consuming fish from contaminated waters (9). In the past, some farm families were exposed to PCBs from dairy products; these PCBs originated from coating material used in the inside of silos (IS). In addition, workers who repair transformers and woricers who handle toxic wastes may also be exposed (16). The PCB products that were manufactured by Monsanto in the United States had the trade name "Aroclor." The particular kind of Aroclor is identified by a four-digit number. The first two digits refer to the 12 carbon atoms, and the second two refer to the percent, by weight, of chlorine in the mixture. Thus, Aroclor 1254 contains about 54% chlorine, and Aroclor 1260, about 60% chlorine. The composition of this mixture of chemicals, with different properties, changes once it gets into the environment and into organisms. Some com ponents of the mixture are more easily degraded in the environment than others. As a result the PCBs identified in the environment resemble Aroclor 1254 but arc not identical to it. Similarly, the PCB mixtures found in humans usually resemble Aroclor 1254 if exposure occurred primarily through the environment. A different composition of the PCBs may be found in serum or adipose tissue samples from occupationally exposed workers. For instance, if the workers are primarily exposed to Aroclor 1016 or Aroclor 1242, which contain much less of the more highly chlorinated homologs, then their gas-chromatographic patterns resemble a combination of Aroclor 1016 or 1242 and Aroclor 1254. For this reason Smith et al (17), in evaluating occupational exposure, divided PCBs into high and low chlorinated biphenyls. The gas-chromatographic pattern of the PCB mixture present in humans, can be used to determine whether the exposure occurred primarily o O g CQ U l LEXOLDMONOOI872 90 KIMBROUGH through occupation or through the environment, or. in the case of occupation al situations, whether most of the exposure was recent or occurred many years ago. Although 93% of the US population eats fish, the average annual per capita consumption is small: 15 lbs. per year (6,7). If PCBs ate to be quantitated in fish, the edible portion rather than the whole fish must be examined. Because of their lipophilicity, PCBs are preferentially stored in the hepatopancreas of the fish, giving ctroneously high levels if the whole fish is analyzed. Sim ilarly, levels in cooked fish are lower (6|. Generally, PBBs are not found in the environment because they have had less commercial use than PCBs. PBB contamination is essentially restricted to Michigan's lower peninsula. Most persons who lived in Michigan during the 1973-1974 period have low-level PBB body burdens (18). The greatest degree of contamination occurred mainly in areas with contaminated farms; this segment of the population still has appreciable body burdens (19). Since PBBs and PCBs are lipophilic, they are preferentially stored in adipose tissue. They are also present, to a smaller extent, in scrum and other organs and in human milk. The concentration of these materials in different organs depends upon the lipid content of such organs, with the exception of the brain where the concentration is lower than the lipid content would indicate. PCBs and PBBs pass the placenta and are primarily excreted through bile and milk. In addition to lipid content, the ratios between adipose tissue, blood, and vital organs are influenced by exposure level, sex. age, length of exposure, and also by whether exposure is cunent. At very low con centrations an analytical imprecision influences the ratios much more than at higher concentrations (19). Since human milk is relatively easy to obtain, it has been used to monitor human exposure. Jensen (20) recently summarized results of such monitoring studies. Average levels of PCBs below 2 ppm (mg/kg) in milk fat have normally been found, although women living in heavily industrialized urban areas may have higher levels. The fat concentra tion in human milk averages 2.6-4.5% (21). At 2% fat, I liter (I) of milk would contain 0.04 mg or 40 fig if the PCBs were present in milk fat at a concentration of I ppm. If an infant weighed 5 kg and imbibed 750 ml of milk per day, it would take in about 6 /xg/kg, a dose that exceeds the 1.5 /xg/kg dose calculated as acceptable by Cordle et al (6). At 1% milk fat this dose would be reduced to 3 /xg/kg. As the infant gains weight, the dose on a kilogram body weight basis will be reduced to some extent, however; milk is the sole food source for only about six months. After the first week, the daily milk intake is estimated to be 150 ml/kg body weight per day. This consump tion gradually falls after two months and declines to 120 ml/kg body weight at four-to-six months. Finally, the amount of PCBs and other halogenated organic chemicals declines with time. However, at low concentrations this HUMAN III.AI.TH lllf-CTS OF PCBs & PBBs 91 may not be obvious because of continued exposure of the mother and the variability of the analytical results. In addition to PCBs, human milk contains trace amounts of many other persistent chemicals. Whctlicr the infant's consumption of such chemicals has any adverse health effects is not known. Most persons, particularly in industrialized countries, have had some expo- sure to polychlorinated biphenyls even if they do not eat fish. The con centrations at which such exposure presents a risk are not clear. Recently, Cordle et al (6,7) calculated the dose of PCBs to people consuming fish from Lake Michigan. They concluded that persons eating Lake Michigan fish ingested an average of 46.S mg of PCBs per year, this amount ranged from 14.17 to 114.31 mg/year/person. The calculated mean daily dose received by the exposed group was 1.7 /xg/kg/day and ranged from 0.09 to 3.94 /xg/kg/ day. Thus, the average sports fisherman consuming contaminated fish would receive a total PCB dose equal to 200 mg in about 4.3 years. No adverse health effects or groups of symptoms clearly related to PCB exposure could be identified in this exposed group. The presence of PCBs in the exposed persons has not caused any observable adverse health effects similar to those observed in the Yusho population (see below). However, this finding does not exclude the possibility that the effects are too subtle for detection or that they require long-term observation. Similarly, in Michigan an analysis of 1,075 human milk samples showed that all contained PCB residues and that the residues ranged from trace amounts to 5 ppm (mg/kg) based on fat level. The public health significance of PCB residues in human breast tnilk and their effects on breast-fed infants are unclear. Since there are no human data on which to base public health policy, risk predictions for PCBs have been based on results from animal studies, particularly the positive bioassay studies. Reviewing these data, Cordle cl al (6, 7) concluded that a 2-ppm (mg/kg) tolerance for PCB in fish be established, since a I-ppm (mg/kg) tolerance does not greatly reduce the estimated risk. . As previously mentioned, some of the isomers of the PCBs and PBBs are much more easily degraded or metabolized. Because they can be metabolized, they are more easily excreted. Others may be retained in the body for long periods; in general, the PBBs appear to be more persistent in human tissues than the PCBs (19, 22). ; n CM Z gj (0 ' POLYCHLORINATED BIPHENYLS Background When PCBs were first used industrially some workers developed chloracne. Results of early animal studies seemed to suggest that PCBs might have some toxic effects on the liver. Beyond that observation no information was avail- LEXOLDMONOOI873 92 KIMUKOUGII able. Because PCBs were so inert chemically, (hey were not considered to cause a great deal of toxicity. In 1968 a poisoning outbreak occurred in Japan (23) that affected over 1,000 persons. These individuals had purchased rice oil, in large drams, front a single source and had used this rice oil for cooking, j Chloracne was one of the leading signs in those who became ill. It was soon discovered that PCBs had been used as a heat-exchange fluid in the factory where the rice oil originated. PCBs had leaked out of the columns in which they were contained into the rice oil when the rice oil was heated. Since the disease was caused by ingesting contaminated rice oil, it was called Yusho (rice-oil disease). When the outbreak first occutTcd, its association with exposure to PCBs was not clear. At that lime the capabilities for measuring these types of chemicals in tissues and body fluids were limited, particularly in Japan. Therefore, early in the investigation total chlorine, rather than PCBs, was measured. Retrospectively determining the precise dose these patients received is difficult. Whether the consumed oil was uniformly con taminated is also not clear. However, a relationship between the amount of rice oil ingested and some symptoms could be established (24). Because of this poisoning outbreak and other environmental problems, animal studies were started in Japan, in the United States, and in other countries to elucidate the toxic effects of PCBs. These data are summarized in several detailed reviews (16, 25, 26). Animal Studies This article addresses primarily the human health effects of PCBs and PBBs. Therefore we highlight only recent results from animal studies that might give a better understanding of implemented public health policies and potential human health effects. One of the difficulties in using animal data to predict human health effects for PCBs and related compounds is that animal species vary greatly in their responses. Further, many of the animal studies use relatively high doses. Therefore, determining how such animal studies relate to the human situation is difficult. Some animal species, such as the subhu man primates, the guinea pig, and the mink, are much more sensitive to the toxic effects of PfBs than the rat or the mouse; also the types of toxic effects and morphological changes in the organs of different species vary. Most animal studies conducted during the 1970s used mixtures of PCBs. In general, PCBs were found to affect reproduction and the immune response, and to cause liver tumors in rodents (16). When different mixtures of PCBs were studied, however, the results were inconsistent. For instance, the mix ture Aroclor 1254 affects reproduction in rats at much lower doses than docs Aroclor 1260 (27). More recently some of the isomers of the PCB mixture were found to be much more toxic than others (28-32). The more toxic isomers constitute only HUMAN llliALTH EFFECTS OF PCBs & PBBs 93 a very small portion of the mixture, particularly those with less chlorine by weight, such as Aroclor 1242 or Aroclor 1016. Recently, Schaeffer et al (33) found that a German PCB mixture--Clophen A-30, with an average composi tion of 1% monochlorobiphenyl, 20.7% dichlorobiphcnyl, 57.4% trichlo- robiphenyl, 17.3% tetrachlorobiphenyl, 1.8% pentachlorobiphenyl, 1.0% hexachlorobiphenyl, 0.6% heptachlorobiphcnyl, and 0.1% octachlor- obipltenyl--produced a 3% incidence of hepatocellular carcinoma, whereas Clophen A-60 produced a 61% incidence of hepatocellular carcinoma in Wistar rats. The incidence of the disease in the controls was 2%. The Clophen A-60 had an average composition of 0.2% monochlorobiphenyl, 1.1% di chlorobiphcnyl, 2.2% trichlorobiphenyl, 3.1% tetrachlorobiphenyl, 19.8% pentachlorobiphenyl, 43.2% hexachlorobiphenyl, 25.3% heplachlor- obiphenyl, 4.7% octachlorobiphenyl, and 0.3% nonachlorobiphenyl. Sim ilarly, Norback & Wellman (34) and Kimbrough et al (35) were able to produce hepatocellular carcinomas in rats with Aroclor 1260, the more highly chlorinated Monsanto product. When Aroclor 1254 was fed to rats, fewer liver tumors developed in g exposed rats (36); however, the incidence of gastric intestinal metaplasia and o adenocarcinoma of the stomach increased (37, 38). Whether PCB fractions 0 without hexachlorobiphenyls, heptachlorobipbenyls, and octachlorobiphenyls produce hepatocellular carcinomas in rodents should be explored. g Particularly in the United States, mixtures such as Aroclor 1242,1254, and ^ 1016 were used more than Aroclor 1260. Because of these differences in potency, the PCBs in heavily contaminated areas of our environment should be characterized according to their isomeric composition. For instance, whether the PCBs in Lake Michigan are of the same composition as those found in New Bedford Harbor, Massachusetts, is not clear. Aside from tumor formation, PCBs cause a variety of other biological effects, such as the induction of enzymes (39). In some species they may cause atrophy of the thymus, intrahepatic bile duct hyperplasia, hyperplasia of the epithelial lining of the urinary bladder, atrophy of tire sebaceous glands, and hyperkeratosis of the ducts (40). Some isomers are fetotoxic, and some produce metaplasia of the sebaceous glands, nailbeds, ameloblasts, thymus corpuscles, and gastric mucosa (28, 41). Subhuman primates, mink, and guinea pigs are particularly sensitive to the toxic effects of PCBs; other species, such as the rat, the mouse, and the dog, can tolerate much higher doses. From empirical observations, humans also appear to be less sensitive to the toxic effects of PCBs. The ability to store these chemicals in adipose tissue may be protective. Generally, the subhuman primates and mink have less adipose tissue than humans. Animals with greater ability to store vitamin A on a quantitative basis, such as the hamster and the ret, are somewhat less susceptible to the toxic effects of these types of compounds (42). The LEXOLDMONOOI874 94 KIMHKOl'OH mechanism by which these types of chemicals affect hepatic retinoids is not clear. Apparently, the duration of the reduction of hepatic retinoids does not correlate with the induced aryl hydrocarbon hydroxylase (AHH) activity (43). Body Burdens Many investigators have repotted PCBs in human tissues (44, 43). In the United States, according to data from the Centers for Disease Control (CDC), mean PCB serum levels are about 3-7 ng/ml (pbb), although some patients may have higher serum levels without any documented unusual exposure. These data were also summarized by Kreiss (46). Levels in adipose tissue and in human milk fat are 100-200 times as high, since PCBs are highly lipid soluble (20). Mes et al (47) repotted that PCB levels in adipose tissue of accident victims ranged from 0.9-9.4 mg/kg. Sahl et al (48) surveyed PCB blood levels in 738 pre-employed and 1,038 currently employed workers of a utility company. The median blood level before employment was 4 mg/I and the range, 1--37 mg/1. These levels were quite similar to those in the currently employed group. Patients who died of cancer in Denmark had somewhat higher levels of PCBs in their adipose tissue (49). Since terminal cancer patients have usually lost a great deal of weight, bioconcentration may have occurred. Similarly, in patients with highly impaired liver function, tissue concentrations of xenobiotics may be slightly higher than those in healthy persons. However, levels remained higher if parameters such as weight, height, occupation, and resi dence were considered (30), whereas levels of PCBs in breast fat tissue from patients with breast cancer were similar to those of controls (SI). Furthermore, Lawton et al (32) demonstrated that random errors and interlaboraiorv variations in procedure and methods of data reporting can influence serum and adipose PCB levels. Unless an interlaboratory qualitycontrol system is set up, measured levels between laboratories are not neces sarily comparable. For instance, Lawton et al (52) found that the results of repeated analyses on serum samples of known composition showed the 93% prediction interval for an individual measurement to be about 42%. This interval depends on the method of extraction, the procedure used, and the means of quantitation. Summary of Human Epidemiology Studies Recently, investigators studied the predominantly black population of Triana, a small rural town in the southern United Slates (53). This population was excessively exposed to DDT residues by consuming contaminated fish. The residents also had PCB body burdens. Fish consumption correlated positively with PCB blood levels; no other source of PCB exposure could be established. These researchers noted that PCB serum levels increased with age and that HUMAN HEALTH EFFECTS OF PCBs A PBBs 93 levels were lower in females of each age group. Similar findings were made for DDT residues. The serum cholesterol level was positively associated with the log PCB level, independent of age, sex, fish consumption, body-mass index, and alcohol consumption. Rates of borderline and definite hyperten sion for study participants were 30% higher than those expected on the basis of national rates (54). Log PCB serum values contributed significantly to explaining the variability of log systolic and diastolic blood pressure in multiple regression analysis (53). Median total cholesterol levels of in dividuals in the United States increase with age from about ISO to 160 mg/dl al age 20 to over 200 mg/dl at age 50. PCBs in blood are influenced by serum lipid content, and populations with inherently lower total serum cholesterol levels appear to have a different PCB serum to adipose tissue ratio. The age-associated increase in blood PCB levels could be related to the long half-life of some PCB isomers that are preferentially retained in mammals (29); as long as exposure continues, a true steady stale between intake and excretion is never reached. Other variables affecting body burdens may be differences in metabolism with age. In the Triana studies, the blood levels of total DDT residues also increased with age, and others have made similar observations (56, 57). Lawton et al (38) studied workers who had been exposed to electrical-grade Aroclor 1016, 1242, and/or 1234; the study covered the period from before the workers were exposed to two years after PCB exposure ceased. Serum levels for the lower chlorinated PCBs in 1977 ranged from 57-2270 ppb and in 1979, from 12-392 ppb; for the higher chlorinated PCBs serum levels ranged from 6-142 ppb in 1977 and from 4-108 ppb in 1979. These findings again illustrate the preferential excretion of lower chlorinated PCB. Lawton et al (38) also found that cholesterol levels correlated with log serum PCBs. Similar associations with log serum PCBs were found for log gamma glutamyl transpeptidase (GGTP) and, in some cases, for log alanine aminotransferase. When the PCB concentrations were expressed as levels in serum lipids, all the associations between serum lipids or enzymes and log serum PCBs disappeared except for those between log GGTP and log PCBs. Similarly, Chase et al (59) found no significant correlation between either serum triglycerides or aminotransferases and the PCB levels in adipose tissue. How age and length of exposure affect these parameters is not adequately explained in the article. Finally, Akagi St Okumura (60) were not able to confirm a positive association between PCB blood levels and elevated blood pressure in Yusho patients. Thus, as Brown (61) has suggested, the positive association between PCB serum levels and elevated triglycerides and serum cholesterol can be ex* plained by the increased solubility of PCB in serum with higher lipid content. In several cross-sectional studies of exposed workers, only minor abnormalities not necessarily related to PCB exposure have been detected % KIMBROUGH (62-66). In cross-sectional studies, however, the ability to evaluate chronic health effects is limited. In several studies, a positive association between results of one liver function test--the test for y-glutamyllranspcplidase--and PCB blood levels has been found. Kimbrough (67) has summarized earlier studies on the health effects of PCBs observed in workers. In 1930 and 1940, chloracne, a disfiguring skin disease, was reported among workers exposed to PCBs. One of the clinical features of chloracne is the chloracne cyst, which is skin colored and measures from 1-10 mm in diameter, with a central opening. The other dominant lesion is the comedo. The skin lesions may only involve the face, but many also extend to other parts of the body. Microscopic examination of human skin biopsies from chloracne cases shows markedly dilated hair follicles filled with keratin. The sebaceous glands involute partially or completely. The epithelial cells lining the hair follicles and the adjacent surface epithelium proliferate, and acantho sis is present. In old lesions, the epithelial lining of the greatly dilated hair follicles becomes atrophic. Jones & Alden (68) examined 17 of 23 workers engaged in the production of PCBs. The workers had chloracne involving the face, genitalia, trunk, and extremities. Before the outbreak of chloracne in the plant, the electrical property of the PCBs had fallen below specifications, and the color had deepened. In the report, symptoms of illness were extensively described for the first worker who was diagnosed as having chloracne. This worker com plained of lassitude, loss of appetite, and loss of libido. Over the years, other cases of chloracne following exposure to PCBs have been reported. Most of these involved exposure to vapors that developed when PCBs were heated (69). At times, the skin rashes that developed in workers were accompanied by pruritus. Some workers also complained of burning of the eyes, nose, and throat; dry throat; nausea; and dizziness. Meigs et al (70) reported chloracne in workers who had been exposed to PCB vapors for 5-14 months. The concentration of PCBs in the workers' breathing zone was 0.1 mg/m1. Evidence of slight liver injury was also present. Ouw el al (71) found air levels in a capacitor plant that ranged from 0.32-1.44 mg/m1 Aroclor 1242 (PCB). Here workers complained of burning eyes, face, and skin in general, and persistent body odor. One worker suffered from chloracne, five com plained of eczematous rashes, and a few had abnormal liver function tests. These workers had a mean PCB blood level of about 400 ppb (/tg/kg). In most studies of workers with chloracne, evidence of liver injury was also found; in one study, workers who did not have chloracne were found to have abnormal liver function (69). PCBs also affect the liver by inducing mixed-function oxidases. Alvares et HUMAN HEALTH EFFECTS OF PCBs & PBBs 97 al (72) determined that in five workers occupationally exposed to Aroclor 1016--a PCB mixture primarily composed of dichlorobiphenyls, trichlor- obiphenyls, tetrachlorobiphenyls, and pentachlorobiphenyls--plasma anli- pyrine half-life was significantly lower than that in matched controls, suggest ing the induction of mixed-function oxidases in the liver. These workers had been exposed to Aroclor 1016 for at least two years and had no obvious I symptoms of PCB poisoning. Other health effects are eye and upper respiratory irritation. Warshaw et al (65) studied a group of 326 workers in a capacitor plant with a mean employment of more than IS years and mean employee ages of 41.1 years for males and 47.3 years for females. Work-related eye or upper respiratory irritation was repotted by 48% of the workers, and 10% had experienced tightness in the chest. Spirometric studies were conducted on 309 workers; 66 of them were dropped from the study because they had been exposed to talc, textile dust, or asbestos. Thus, 243 men were available for analysis. In males, there were about twice as many smokers and exsmokers as nonsmokers. In females, the proportion of nonsmokers was higher. Thirty-four of the workers in (14%) had a reduced vital capacity, and 27 of these demonstrated a restrictive pattern of impairment. Because of additional variables such as smoking and tst asbestos exposure, these findings are difficult to interpret. Taylor et al (73), in an attempt to determine whether the fetus would be Z affected in capacitor workers, examined pregnancy outcome and birth weight, gj and found that the gestation period was reduced by one week. The infants weighed slightly less than the controls; this finding could be explained by the reduced gestation period. Smoking and alcohol consumption were not con tD trolled, however, furthermore, whether the socioeconomic status of this group of women was similar to that of the control group is not clear. Thus, until other studies confirm these findings, they should be viewed with caution. In several papers Jacobson and his associates reported behavioral changes and a reduced gestation period in association with higher fish intake or higher intake of PCBs (9, 74-76). Furthermore, Jacobson et al (74) reported that intrauterine PCB exposure may have a delayed effect on central nervous system functioning. Since genetic makeup, the mother's lifestyle, and acute illness also affect these parameters, these findings are difficult to interpret. Furthermore, many other chemicals are also excreted in human milk (20). Rogan & Gladen (77), for instance, found that mothers with high levels of DDE |l,l'-(2,2-dichloroethenylidene)-bis-4-chlorobenzene| in their milk tended to wean their infants earlier, as they did not thrive. Apparently, PCB levels in milk were higher in older women, women who drank alcohol regularly, and primiparas (78). Whether high levels of DDE affect lactation is not clear. In animals, DDT LEXOLDMONOOI876 98 KIMBROUGH homologs, but not specifically DDE, have been shown to have estrogenic effects (79). Before these findings can be clarified, additional studies must be done. No conclusive evidence thus far reported shows that occupational exposure to PCBs causes an increased incidence of cancer. Bahn el al (80) repotted results of a preliminary study of a group of SI research and development employees and 41 refinery plant employees at a New Jersey petrochemical facility. Between 1949 and I9S7 these workers had been exposed to Aroclor I2S4. Three melanomas and two carcinomas of the pancreas were found. This incidence was significantly higher than expected. Exposure to other chemicals also occurred, however, and the cohort was small. Brown & Jones (81) conducted a retrospective mortality study of 2.S67 workers in two capacitor plants. The relatively few deaths (163) severely limited the statistical power of the study, and the average follow-up was only 13 years, whereas latency periods of 20-30 years are not uncommon for cancer. Over 50% of the sample had exposure to PCBs for two years or less. Deaths from liver cancer, cirrhosis of the liver, and rectal cancer were slightly higher than expected, but not significantly for both sites combined. The observed increase for cancer of the rectum was statistically significant among females at one of the plants. In a follow up study (82) no additional cancers of the rectum were noted, and the standardized mortality ratio (SMK) dropped from 336 to 211. However, two additional cancers of the liver and biliary tract were observed, bringing the total of these tumors to five as reported on the death certificates. However, a review of the medical records raises questions about at least one of these tumors. Bertazzi et al (83) reviewed the mortality of 290 males and 1,020 females who had worked for six months or more in capacitor production. Males had a statistically significant increased number of deaths from all neoplasms. When deaths were analyzed by organ system, deaths front neoplasms of the di gestive system, the peritoneum, and the lymphatic and hematopoietic tissues were higher. Among females, all causes of deaths were significantly elevated. The actual numbers in this study, however, were small. Yusho and Yucheng Two outbreaks of poisoning have been reported that followed the ingestion of rice oil contaminated with polychlorinated dibenzofurans, biphenyls, and quaterphenyls (PCQs). The first outbreak occurred in Japan in the summer of 1968 and the second outbreak, in Taiwan in 1979. Ironically, the outbreak in Taiwan repealed what had occurred 10 years earlier in Japan. Many studies of these two outbreaks have been published in Japanese or Chinese. In 1984 some of the information in these reports was published in English in the HUMAN HEALTH EFFECTS OF PCBs A PBBs 99 American Journal of Industrial Medicine S:I-IS3; the information is also summarized in volumes 59 and 60 of Environmental Health Perspectives. In Japan and Taiwan the disease was first recognized because chloracne developed in the affected patients (84). In Japan, members of all of the affected households had purchased rice oil from a specific company, and the . toxic rice oil produced or shipped on February 3 and 6 of 1968 contained large amounts of Kanechlor 400, a brand of PCB with a chlorine content of 48%. At the time of the outbreak, no analytical methods specific far PCBs were available in Japan; the concentration of Kanechlor 400 in the oil was therefore estimated from the organic chlorine content to be 2,000-3,000 ppm. Kanech lor 400 had been used for heating the rice oil in a metal container at over 200C, at a reduced pressure of 3-44 mm Hg, to remove odorous material from it. Kanechlor (K) must have leaked from the heating pipe into the processed oil, but the actual mechanism of the contamination has apparently ^ not been determined. The reanalysis of the K-rice oil, once the methods were g developed, showed that some of the oil samples contained 1,000 ppm (mg/kg) ^ PCB. This concentration was much lower than had originally been estimated. g Therefore, other chlorine-containing compounds were assumed to be in the z oil, and additional samples were analyzed. The oil was found to contain an & average of 5-ppm polychlorinated dibenzofurans (83). According to Buser el 0 al (86). the Yusho oil contained more than 40 polychlorinated dibenzofuran isomers, including the highly toxic 2,3,7,8-tetrachlorodibenzofuran (TCDF) and 2,3,4,7,8-pentachlorodibenzofuran (PCDF). In addition, the oil con tained PCQs at a concentration of 866 ppm (87, 88). According to estimates made by Kuratsune (89), the total amount of PCB, PCDF, and PCQs consumed by the patients was, on the average, 633 mg of PCB, 3.4 mg of PCDF, and 596 mg of PCQ. This calculates to roughly 157 /ig/kg body weight/d. PCB, 0.9 ftg/kg body weight/d. PCDF, and 148 /ig/kg body weight/d. PCQ. At this dose the length of the latent period between exposure and onset of clinical illness was roughly 71 days, with a range from 20 to 190 days. Some of the oil the patients consumed may have contained higher or lower levels because in such situations contamination is usually not uniform. Furthermore, the patients consumed different amounts of con taminated rice oil. The severity of symptoms was positively associated with the amount of contaminated rice oil consumed (24). Early in the outbreak the patients had chloracne, dark-brown pigmentation of the nails, itching, pigmentation of the skin, swelling of the limbs, pig mented mucous membranes, eye discharge, hyperemic conjunctivae, jaun dice, swelling of the upper eyelids, a feeling of weakness, numbness of the limbs, and fever. Over 1,000 people were affected. Thirty-six babies showed fetal PCB syndrome, which consists primarily of a dark-brown pigmentation LEXOLDMONOOI877 100 KIMBROUGH of the skin (Cola babies). The cutaneous pigmentation was caused by an increase in melanin pigment in the epidermis (90). The mucous membranes were also pigmented. In all cases, the pigmentation disappeared by the time the babies were between two and five months old. In affected infants, the face was edematous, and spotty calcifications were noticed in the parietal and occipital areas of the skull. In a few of the infants, ihe teeth had erupted at birth. Subsequently, the adult patients with clinical disease complained of having to expectorate a great deal and, on auscultation, wheezing was noted; however, on examination, there was no evidence of bronchial asthma or pulmonary emphysema. In many of these patients, the respiratory symptoms have persisted, and ihe patients have chronically infected airways. In the early 1970s, some changes were noted in the patients* serum immunoglobulin levels, but the levels returned to normal. Over time the severity and Ihe extent of the skin lesions improved considerably in the exposed population. Fifteen years after ihe accident, only a very few patients had extensive chloracne (91). About five years after the outbreak of Yusho, tissue and body fluids of Yusho patients were analyzed for various congeners of PCB and PCDF. At this time, the PCB levels in adipose tissue were 1.9 1.4 ppm (mg/kg). In the liver they were 0.08 0.06 ppm and in blood, 6.7 5.3 ppb (/rg/kg); thus, they were not very different from levels in the general population in Japan. On the other hand, the isomeric distribution for Ihe PCBs in the Yusho patients varied from that in the control population in Ihe same area (92). About 40 PCDF congeners were identified in the rice oil that the Yusho patients ingested. Only some PCDF congeners were retained in the body for a long time; they included 2.3,6.8-TCDF. 2,3,7,8-TCDF. 1,2,4.7,8-PCDF, 2,3,4,7,8-PCDF, and 1,2,3,4,7,8-hexachlorinated dibenzofurans. Since these congeners do not have free adjacent carbon atoms, they are not as easily metabolized and excreted. More of the 2,3,4,7,8-PCDF than the other iso mers was retained in the patients' tissues. In the five patients studied, the concentration of this isomer ranged from 6.9 ppb (figlkg) in a specimen obtained in 1969 to 0.1 ppb (/xg/kg) in a specimen collected in 1977. Measurable concentrations of TCDFs were only detected in the earlier years. Although not the most toxic isomer, the 2,3,4,7,8-PCDF caused mixedfunction oxidase induction at a dose of I fig/kg in rats, and atrophy of the thymus, suggesting toxicity at a very low dosage level. Thus, the clinical manifestations observed in these patients were primarily caused by the PCDFs, specifically by the more toxic isomers. In the Yucheng episode, it was never determined with certainty how the rice oil was contaminated (93). In 1979 a school for blind persons informed a local health bureau in Taichung County that a strange disease characterized by an acnelike skin eruption had been occurring frequently among students and HUMAN HEALTH EFFECTS OF PCBs & PBBs 101 staff since the end of March. At Ihe same time, 85 of 150 workers in a nearby plastic shoe factory had the same symptoms. Later that year, this outbreak was also reported to a local health bureau. Victims in both outbreaks had consumed the same brand of cooking rice oil, which had been manufactured by the same company and which had been purchased in the same store. For this reason the rice oil was the prime suspect in the outbreak. In additional reports of outbreaks in other companies and in the general population, all victims had consumed the same type of C-rice oil. Finally, because the disease resembled the Yusho disease in Japan, samples of C-rice oil and patients* blood were analyzed in Japan and were found to contain either a Kancchlor-400 or a Kanechk>r-500 mixture at concentrations as high as 65 and I0B ppm (mg/kg), respectively. Over 2,000 patients were finally identified as having been poisoned by contaminated rice oil. Oil samples collected from other outbreaks contained PCBs at concentrations of 31-300 ppm (mg/kg). Retrospective studies determined that the period of PCB intake ranged from 3 to 9 months. The average total intake for each n person varied from 0.77 to 1.8 mg of PCB. Within the first year of the outbreak, the blood levels of PCB in 13 patients ranged from 3 ppb to 1,156 N ppb. Most of the patients had blood levels between 11 and 150 ppb (pg/kg). The symptoms observed in these patients were quite similar to those already Z described for the patients in the 1968 Yusho outbreak in Japan. (g The rice oil was not only contaminated with PCBs but also with PCDFs and U polychlorinated quaterphenyls. It contained the same major components of PCDFs observed in the rice oil in Japan--namely, 2,3,4,6,7-PCDF and 2.3.4.7.8-PCDF. Relatively high concentrations of 2,3,4,5,3' ,4'-hexachlo- robiphenyl were found in the blood and adipose tissue of the Yuchcng patients. This particular PCB isomer is biologically quite active, and the concentration of 2,3,4,3'4,-pentachlorobiphenyl was also elevated in these patients. Furthermore, as in the Yusho patients, the concentration of 2,3,7,8TCDF was comparatively low (92). Chen et al (94) analyzed additional samples of the oil, blood, and adipose tissue of the Yucheng patients. These investigators identified several TCDF and PCDF isomers. Apparently, 2.3.7.8- TCDF was only a minor component in the oil; the major component was 2,3,4,8-TCDF. One of the major furans in the toxic oil was 2,3,4,7,8- PCDF. The concentrations of the PCDFs in different oil samples ranged from 0.21 to 1.68 ppm. Polychlorinated quaterphenyls were present in concentrations ranging from 25 to 53 ppm (mg/kg). Overall, the concentrations of PCDFs and PCQs were lower in these oil samples than in the oil samples that had caused the Yusho outbreak. Whether these oil samples were representative is not really known. The 3,4,3'41 -tetrachlorobiphenyl was also identified in the oil that caused Yucheng disease in Taiwan. This isomer is considered to be LEXOLDMONOOI878 102 KIMBROUGH the most toxic PCB isomer present in commercial PCB preparations (95); it was present at a concentration of about 1%. Most other commercial PCB preparations, such as the Aroclors, in the United States have not been shown to contain this particular isomer. In addition to the epidemiological studies, some disease-specific in vestigations were also conducted. The blood pressure of the Yucheng and Yusho patients was not affected (60). Although some of the patients in the Yusho cohort have died of cancer (90), the number has been small; because the latency period may be long, the population should be followed fora longer period to determine whether the cancer incidence will increase. Although PCBs and related compounds are known to affect reproduction in animals, and although they affected some fetuses and neonates in the Yusho and Yucheng episodes, the information on reproduction and fetal toxicity in general is very limited. In one such study, Han (96) examined women working in a capacitor plant who also nursed their infants and who themselves had mild chloracne and erythema of the skin. The human milk of some of these women contained, on a whole milk basis, PCB levels that ranged from below SO ppb (Mg/kg) to about 400 ppb (/rg/kg). Forty children of these mothers were followed for a five-year period. Some children were found to have "decayed" nails, gingival pigmentation, mottled enamel, and denial caries. No relationships between these changes or symptoms to PCB blood levels, however, were observed. The general population in the United States and other countries also has body burdens of PCBs, PCDFs, and polychlorin ated dibenzodioxins (97, 98). However, these background concentrations-- particularly for the biologically active isomers--arc far lower than they were in the Yusho and Yucheng patients, even several years after exposure. Chang el al (99) examined the delayed immune response in 30 Yucheng patients and compared their responses with those of 50 controls. The mean age of patients in both groups was about 14 years. The authors injected a solution of streptokinase and strcptodomase subcutaneously into the flexor side of the forearm. The response was read at 24 hours (hr) and again at 48 hr after injection. Eighty percent of the controls had an induration of S mm or more in diameter 24 or 48 hr after they were injected; only 43% of the exposed group responded similarly. All of the poisoned patients had dermal lesions, and the percentage of patients with a positive response decreased with increasing severity of the skin lesions (chloracne). Furthermore, the degree of the dermal lesions appeared to be associated with the whole blood PCB concentrations. Patients with minor skin lesions that were classified as grade I appeared to have a normal skin response. The same authors found that PCBs caused a decreased concentration of IgA and IgM, but not of IgG, in serum. Furthermore, the percentages of total T cells, active T cells, and T mu cells decreased, whereas the percentage of 3 cells and T gamma cells were not HUMAN HEALTH EFFECTS OF PCBs ft PBBs 103 affected (100). These two reports are the first in which the effect on the immune response was actually correlated with body burdens of PCBs and in which only severely poisoned patients showed this effect. This finding is consistent with the findings from animal studies in which relatively high doses of PCBs affected the immune response and also caused some other adverse effects. In the Japanese and the Taiwanese Yusho and Yucheng poisoning out breaks, sensory neuropathy was reported in a number of patients for whom nerve conduction velocities were measured (101, 102). The blood levels of the various chemicals (PCBs, PCDFs, PCQs) were negatively correlated with the lowered nerve conduction velocity, suggesting that these types of chemi cals affect nerve conduction velocity. (It is not quite clear why most in vestigators measure nerve conduction velocity to detect sensory neuropathy. Other tests that would measure the detection of vibration, touch, and tempera ture would be more useful from a clinical perspective.) < Seppalaincn et al (103) examined 16 men working in a cardboard plant who were exposed to fumes that resulted from the explosion of 15 capacitors containing Clophen A-30. The first PCB air concentrations, measured S.S hr after the explosion, were 8,000 to 16,000 /ig/mJ air. PCDFs were also formed. The soot samples contained letrachlorodibenzofuran up to 90 |ig/g, of which 6.5 /ig/g was 2.3,7,8-tetrachlorodibenzofuran. In addition, mono- chloropyrcnes and dichloropyrenes were found. Most of the men had a transient sensory neuropathy in their lower extremities. Chang et al (104) reported increased urinary 6-aminolevulinic acid uropor phyrin excretion in 69 Yucheng patients over that of 20 controls. No informa tion on the patients' clinical conditions or on how these findings related to degree of exposure was given. No such observations have been reported from o q O g Japan. POLYBROMINATED BIPHENYLS Since the toxicity of PBBs both in laboratory animals and livestock was recently reviewed (105), we do not review it here in detail. In laboratory animals, PBBs generally cause effects simitar to those that the PCBs cause. They produce morphological changes in the liver, affect reproduction, and promote biochemical changes, such as hepatic porphyria and induction of mixed-function oxidases. Teratogenic effects have also been noted. In addi tion, atrophy of the thymus has been reported, and hepatocellular carcinomas have been produced in both rats and mice. The overall findings reported in animal studies are similar to those that have been reported for PCBs. Although the PCB contamination of the environment is a more general problem, the PBB contamination primarily affects certain areas within the stale of Michigan. Most persons living within the lower peninsula of Michi- LEXOLDMONOOI879 104 KIMBROUGH gait have had slight exposure, since the contamination resulted from dairy products and since normal marketing channels lor these products involved the mixing of milk from many producers in relatively few processing facilities. In addition, most cull dairy cattle are used for hamburgers and processed meat products that would also receive wide distribution. Thus, the marketing system diluted the degree of exposure for the individual; however, it increased the number of those exposed. In 1978, the distribution of PBBs was com prehensively studied in a probability sample of 1,738 persons. PBB levels in serum were determined, and 844 adipose tissue samples were also analyzed for PBBs. PBBs were detected in 97.3% of the adipose tissue samples, in 68% of the adult serum samples, and in 72.7% of the scrum samples from children. The mean PBB concentration in adipose tissue was 400 ppb (Mg/kg); in serum it was 1.3 ppb (Mg/kg) for adults and 1.8 ppb (/*/gkg) for children. The highest adipose tissue concentration was 37 ppm (mg/kg) (18). In addi tional studies, when cohorts of PBB-exposed residents of Michigan were compared with residents of the stale of Wisconsin, a higher prevalence of a variety of symptoms and complaints was noted in the Michigan residents (106). Similarly, in comparative ncurobchavioral studies, the Michigan pop ulation was found to be affected more than that in Wisconsin (107).' Since the findings were not correlated with body burdens of PBB in any of these studies, determining whether other factors may be responsible for these differences is difficult. In 1976, the Michigan Department of Public Health established a cohort of farmers who had been exposed to varying con centrations of PBBs in their products and their environment. A total of 3,877 persons were enrolled. They included farm residents, direct recipients of farm products, chemical workers and their families, and a lew persons who had been originally studied in a smaller previous study. The serum PBB levels in this entire group ranged from no detectable levels to 1,900 ppb (/ig/l), with a mean of 21.2 ppb (/tg/l) and a median of 3 ppb (/tg/l). Because of the wide range of exposure and because results could be analyzed by regression analyses with exposure as a variable, a comparison group for acute health effects was not included. This cohort was found to have various symptoms and conditions: however, these symptoms did not correlate with PBB body burdens. Symptom prevalence rates were slightly higher in persons with no detectable PBBs in serum than in those with measurable quantities. In all groups, including chemical workers and quarantined farm residents, the highest prevalence rates were in persons with the lowest serum PBB levels (22). Similarly, in this study and in a previous immunologic study (108) no dose-related depression of lymphocyte function in persons exposed to PBBs could be demonstrated. All these findings suggest that there may be no causal HUMAN HEALTH EFFECTS OF PCBs & PBBs 105 relationship between the abnormal lymphocyte functions observed in some persons or the prevalence of other symptoms and exposure to PBBs. This cohort of Michigan residents is still being followed by the Michigan Depart ment of Health in collaboration with the Centers for Disease Control. Several studies of subgroups of this population and surveys for chronic health effects have been conducted since the cohort was first assembled (109, 19). When scrum and adipose tissue concentrations were compared, a significant correla tion was found. The serum:adipose tissue concentration ratios ranged from l to 140 to I to 260 for pregnant women and male chemical workers, respec tively. Males from farms had a significantly different ratio of I to 325 to 329. Potential transplacental passage of PBBs was demonstrated, since they could also be found in the fetus and newborn. Cord blood contained one-tenlh of the concentration found in the maternal serum, which indicated partial placental passage. Human milk contained PBBs at 107--119 times the quantity round in maternal serum. PBBs were also detected in bile and feces, which indicates that these materials can be transferred into the intestinal tract. All of these concentrations were measured long after the population bad first been exposed to PBBs (19). Concentrations of PBBs observed in bile and feces were about one half to seven-tenths of the serum levels and ate probably about 0.5% of the adipose tissue levels. These findings indicate that PBBs are very slowly ; ^ o excreted, which is consistent with the findings of Tuey & Matthews in rats (110). The estimated half-life for PBB is 6.5 years. b More recently, two groups of Michigan residents--those with high PBB serum levels and those with PBB serum levels around I ppb--were matched K for age, sex. and smoking. For both groups, various clinical laboratory tests were conducted, blood pressure was measured, and height and weight were determined. In this study, 83 participants had PBB serum levels of 50 ppb (/xg/l) or more. In the middle group. 83 bad PBB levels of 5-49 ppb and 96 ' had PBB levels of 0-44 ppb (/ig/l) in serum. Urinary porphyrins were also measured in all of the participants. Thus far, the final results of this study have not been reported. For most of the parameters studied--which included serum glucose, triglycerides, high-density lipoproteins, various liver func tions. creatinine, uric acid, thyroid function, proteins, calcium and phospho rus in scrum, and also measurement of various porphyrins--no differences of clinical significance were found among different groups (M. Barone, personal communication). (Note: Even though significantly more women in the high PBB group used birth-control pills than women in the low PBB group, too few were using them to affect urine porphyrin levels.) In none of a variety of other studies conducted on this population as well as other groups in Michigan did any findings indicate that exposure to PBB had impaired the health of the exposed group. All of these studies have been reviewed by Fries (105), LEXOLDMONOOI880 106 KIMBROUGH ; Although this population was exposed during a 9-month period in 1973 and 1974, whether it will have chronic health effects is unknown. This particular cohort needs to be followed for 30 to 40 years before the question of chronic health effects can be intelligently addressed. Two problems with assessing chronic health effects are that the cohort, in spile of its size, is still relatively small and that the amount ofexposure it has received varied widely. Although some members of the group exposed to PBBs have relatively high body burdens, these burdens are still appreciably lower than those of rats in which liver cancer developed. In the study by Kimbrough et al (111), liver cancer developed in the rats that received a dose of 1.000 nig/kg body weight. This dose for humans would roughly translate into a dose of 70 grams per person. These amounts are much greater than the estimated mean total exposure per person. The highest exposure was about 11.7 grams, and the mean was 170 mg per person. In rats given 200 mg/kg, a dose that for humans would be between 12 and 14 grams, only neoplastic nodules developed in their livers; there was no evidence of hepatocellular carcinomas. Of course, whether humans would be more or less susceptible to the toxic effects of PBBs and whether their response would be similar to that of rats is not known. In conclusion, various toxic effects of PBBs and PCBs have been described in laboratory animals. In humans, acute poisoning outbreaks have only occurred following exposure to a combination of PCBs and PCDFs. When humans were exposed only to PCBs or PBBs, the only observed acute effects have generally been minor. So far. no significant chronic health effects have been causally associated with exposure to PCBs or PBBs. Use of trade names is for identification only and docs not constitute endorsement by the Public Health Service or the US Department of Health . and Human Services. Lileralure Cited 1. Jensen. S. 1966. Repun of a new chemi cal lu/anl. New Sri. 32:612 2. I9S2. Polychlorinated biphenyls (PCBs): manufacturing, processing, distribution in commerce, and use prohibitions: use in electrical equipment, fed. triia. 47:37342-60 3. I9S2. Polychlorinated biphenyls fPCBsl: manufacturing, processing, distribution in commerce, and use prohibitions: use in closed and conirotleil waste man ufacturing processes, fed. Regiu. 47:469X0-96 4. McLeod. K. E. 19X1. Polychlorinated biphenyls in indoor air. Environ. Set. Tfchnot. 13:926-2* 3. Kuu. F. W.. Yang. H. S. C. 1976. A note on pulychlurinaied biphenyls in air. In Prac. Nall. Conf. Potirhlorinalcd Biphenyls. Chicago. I97S. EPA-56W673-004. p. 1X2. Washington. DC: En viron. Prut. Agency 6. Curdle. F., Locke. R.. Springer.1 J. 19X2. Determination of human risk in regulating pulychlurinaied biphenyls (PCBs)---a case study. CSC Toxicol. Risk Assess. 2:211-23 7. Curdle, F.. Luckc. R., Springer. 1. 19X2. Risk assessment in a federal reg ulatory agency: an assessment of tide associated with the human consumption uf some species of fish contaminated with polychlorinated biphenyls (PCBs). Environ. Health Perspea. 43:|7|-X2 HUMAN HEALTH EFFECTS OF PCBs & PBBs 107 X. Sullivan. J. K., Dclfino. I. J., Buclow, C. R.. Sherry, T. B. 19X3. Polychlori nated biphenyls in the fish and sediment of the Lower fin River, Wisconsin. Ball. Environ. Contain. Tomcat. 30-58 64 9. Schwatu, P. M.. Jacobson, S. W,, Fein, G. G., Jacobson, J. L., Price. H. A. 19X3. Lake Michigan fish consump tion as a source of polychlorinated biphenyls in human conf actum, mater nal serum, and milk. Aar. J. Public Health 73:293-96 10. Tatsukawa, R. 1976. PCB pollution of the Japanese environment, la PCB Poisoning and Pollution, cd. K. Higuchi. pp. 147-79. Tokyo: Kodansha It. Marita. M., Nakagawa. J.. Rappe. C. I97X. Pulychlorinated dibeiuufuraa (PCDF) formation from PCB mixture by heal and oaygen. Bull. Environ. Conlam. Toxicol. 19:663-70 12. O'Keefe, P. W,, Silkworth. J. B.. Gierthy, 1. F.. Smith, R. M., DeCaprio, A. P., et al. 1983. Chemical and biological investigations of a transformer accident at Binghamton, New York, Environ. Health Perspeci. 60:201-9 13. Hutzinger. O.. Ghnlam. G. C., Brock, G. C.. Johnston, L. E. 1983. Formation of polychlorinated dibenzofuiant and di oxins during combusrian, electrical equipment fires and PCB incineration. Environ. Health Perspeci. 60:3-9 14. Caner, L. 1976. Michigan PBB in cident. Chemical mix-up leads to disas ter. Science 192:240-43 13. Willett. L. B.. Liu. T. T. Y.. Durst. H. I.. Cardwell. B. D.. Rcnkie. E. D. 1983. Quantification and distribution uf polychlorinated biphenyls in farm silos. Bull. Environ. Contain. Toxicol. 33:31 60 16. Kimbrough. R. D. 19X3. Laboratory and human studies on polychlorinated biphenyls (PCBs) and related com pounds. Environ. Health Perspeci. 39:99-106 17. Smith. A. B.. Schloemer, J., Lowry. L. K., Smallwood, A. W.. Ligo, R. N.. el al. 1982. Metabolic and health con- Xnces of occupational exposure to hloriruled biphenyls (PCBS). Br. J. hut. Med. 39:361-69 IX. Wolff. M. S.. Anderson. H. A.. Selikoff. J. J. 1982. Human tissue burdens of hulogenaicd aromatic chemicals in Michigan. J. Am. tied. Assoc. 247: 2112-16 (9. Eyster. J. T-. Humphrey, H. E- B., Kimbrough, R. D. 1983. Partitioning of polybrominaied biphenyls (PBB) in serum, adipuse tissue, breast milk, placenta, cord blood, biliary fluid ami feces. Arch. Environ. Health 31:47-33 20. Jensen. A. A. 19X3. Chemical contami nants in human milk. Residue Rev. *9:1-12* . 21. Jensen. R. G., Clark. R. M., Ferris, A. 1 M. 19*0. Composition of die lipids in human milk, ateview. Lipids 13:343-33 22. Landtigan. P. J.. Wilcox. K. R. Jr.. Silva. J. Jr., Humphrey, H. E. B.. Kauffman. C.. Heath. C. W. Jr. 1979. Cohort study of Michigan residents ex posed to poiybrominaKd biphenyls: epidemiologic and immunologic find ings. Am. N. r. Acad. Sri. 320:284 94 21. Kuntsuae, M,, Yoshimurs, T.. Matsuttka. J.. Yamaguchi. A. 1972. Epide miologic Hudy on Yusho, a poisoning caused by ingestion of rice oil con- laminated with a commercial brand of polychlorinated biphenyls. Environ. Health Perspeci. Exp. No. 1:119-28 24. Yoshimun. T.. Hayabuchi, H. 19X3. Relationship between amount of rice oil ingested by patients with Yusho and (heir subjective symptoms. Environ. Health Perspeci. 59:47-51 23. Kimbrough. R. D.. Buckley, J., Fish- bein, L.. Flamm. G.. Kasza. L.. et al. 1978. Animal toxicology. Environ. Health Perspeci. 24:173-113 26. Vos. J. G.. Faith. R. E . Luster. M. I. 1980. Immune alterations. In Halogenated Biphenyls. Terphenyls. Naph thalenes. Dibensodioxini and Related Compounds. Topics in Environmental Health, ed. R. D. Kimbrough. 4:241-58 Amsterdam: Elsevier Btorncdical But. 27. Linder, R. E-. Gaines, T. B.. Kim brough. R. D. 1974. The effect of pulychlorinated biphenyls on rat re production. food Cosmel. Tasicol. 12: 63-77 28. McNulty. W. P.. Becker. G. M.. Cory. H. T. 1980. Chronic toxicity of 3.4.3'4'- and 2.5.2'.S'-letrachlorobiphenyls in thesns macaques. Toxicol. Appl. Pharmacol. 56(2)182-90 29. Matthews. H. B., Anderson, M. W. 1975. Effect of chlorination on the dis tributism and excretion of polychlori nated biphenyls. Drug. Meiab. Dispos. 3:371-80 10. Matthews. H. B-. Kato. S. 1979. The metabolism and disposition of halogen- ated aromatics. Ami. N. Y. Acad. Sci. 320:131-18 31. Pstkinson. A.. Robertson. L., Uhlii, L.. Campbell. M. A.. Safe. S. 1981. 2.3,4.4' ,5-Pentachlorobi|ihenyl: differ- 6BRN002010 oooo Oo O Q _i O X LU 108 KIMBROUGH cniial effects on C57BL/6J and DUA/2J inbred mice- Biochem. Phormacoi. 31:2830-33 32. Poland. A.. Glover. E. 1977. Chlori nated biphenyl induction of arylhydrocaibun hythoaylase activity: a study of structure activity relationship. Mol. Pharmacol. 13:924-38 33. Schaeffer. E., Gtcim. H.. Goessncr. W. 1984. Pathology of chronic polychlori nated biphenyl (PCBl feeding in rats. Toxicol. Appt. Pharmacol. 73:278-288 34. Nutback, 0. H.. Wellman. R. H. 1983. Polychlorinated biphenyl induction of hepatocellularcarcinoma in the SpragucDawlcy rat. Enviroo. Health Pertpect. 60:97-103 33. Kimbrough, R. D., Squire. R. A.. Lin der, R. E.. Slrandberg. i. D., Montali, R. J.. Burse, V. W. 1973. Induction of liver tumors in Sherman strain rats by polychlorinated biphenyl Aroclor 1260. J. Nail. Cancer last. 33:1433-39 36. National Cancer Institute. 1977. B'uias- say of Arodor (trademark) 1234 for passible carcinogenicity. Springfield. VA: Nall. Tech. Inf. Sent, as PB-279 624/IGA. NCI-CG TO-38 Chem. Abstr. Scrv. No. 27323-18-8. DHEW Publ. NIH-78-838 37. Ward. I. M. 1983. Proliferative lesions of the glandular stomach and liver in F 344 rats fed diets containing Aroclor 1234. Environ. Health Persped. 60:89 96 38. Morgan. R. W.. Ward. I. M.. Han- mann. P. E. 1981. Arodor 1234 induced inteslional metaplasia and adenocarcino ma in the glandular stomach of F344 rats. Cancer Ret. 41:3032-39 39. Safe. S.. Bandiera. S.. Sawyer. T.. Rohensun. L.. Safe. L . et al. 1983. PCB: structure-function relationships and mechanism of action. Environ. Health Penned. 60:47-36 40. McConnell. E. E. 1980. See Ref. 26. pp. 109-90 41. McNulty. W. P. 1983. Tosiciiy and fetotoxicity of TCDD. TCDF. and PCB isomers in rhesus macaques (Macaco national. Environ. Health Penned. 6077-88 42. Thunberg. T. 1984. Effect of TCDD on vitamin A and its relation to TCDD tosiciiy. Banbury Rep. 18: Biol. Mach. Dioxin Action. New York: Cold Spring Harbor Lab 43. Brouwer. A.. Van den Berg. K.J..Kufc- Icr. A. 1983. Time and dose responses of the reduction in retinoid concen trations in C57BI/Rij and DBA/2 mice induced by J.d.S'd'-lctrachJorubi- phenyl. Toxicol. Appl. Pharmacol. 78: 180-89 44. Kutz, F. W.. Slrassmaa. S. C. 1976. Residues of polychlorinated biphenyls in the general population of the United Slates. See Ref. 3. pp. 139-48 43. Juskicwicz, T., Nicwiadowska, A., Radomanslti. T. 1977. Polychlorinated biphenyl residues in human adipose tis sue. Pol. Tvg. Lek. 32:173-175 46. Krciss, K. 13383. Studies on populations exposed to polychlorinated biphenyls. Environ. Health Pertpect. 60:193-199 47. Mcs. J.. Davies. D. J..Tunon, D. 1982. Polychlorinated biphenyl and other chlorinated hydrocarbon residues in adi pose tissue of Canadians. Bull. Environ. Contain. Toxicol. 28:97-104 48. Sahl, J. D.. Cracker. T,, Gordon. R. !.. Feeder. E. I. 1983. Polychlorinated biphenyls in the blood of personnel from an electric utility. J. Occup. Med. 27:639-43 49. Unger, M..Olsen. J. 1980. Organochlo- rine compounds in the adipose tissue of deceased people with and withuul can cer. Envjrurt. Ret. 23:257-63 30. Unger. M.. Olsen. I.. Clausen, J. 1982. Organochlorinc compounds in the adi pose tissue of deceased persons with and without cancer: a statistical survey of some potential confoundets. Environ. Ret. 29:371-76 51. Unger. M.. Kiaer. H., Blichert-Toft, M.. Olsen. I.. Clausen. I. 1984. Or- ganochlorine compounds in human breast fat from deceased with and with out breast cancer and in a biopsy malerial from newly diagnosed patients un dergoing breast surgery. Environ. Ret. 34:24-28 32. Lawton. K. W., Brown. I. F.. Ross. M. R.. Feingold. J. 1985. Comparability and precision of serum PCB measure ments. Arch. Environ. Health 40:29 37 i 33. Krciss. K., Zack, M., Kimbrough, R. D.. Needham. L. L.. Struck. A. L.. Jones, B. T. 1981. Cross-sectional study of a community with exceptional expo sure to DDT. J. Am. Med. Attoc. 245:1926-30 34. National Center for Health Statistics. 1978. Blood pressure levels of persons 6-74 years. US. 1971-1974. HRS 78 1618. scries II, No. 203. Hyallsville, Md. Natl. Cent. Health Slat.. DHEW 53- Krciss. K.. Zack, M., Kimbrough, R. D.. Needham, L. L.. Struck, A. L.. Jones, B.T. 1981. Association of blood Cessure and polychlorinated biphenyl vcls. J. Am. Med. Attoc. 243:2303-4 HUMAN HEALTH EFFECTS OF PCB* & PBBs 109 56. Kutz. I:. W., Yobs, A. K., Slrassman. S. C.. Viar. J. F. 1977. KITcclsof reduc ing DDT usage on total DDT storage in humans. Pestic. Mooit. J. 11:61-63 37. Davies, J. E., lidniundson, W. F\. Kaf- I'onclli. A., cl al. 1972. The role of so cial class in human pesticide puthnion. Am. J. Epidemiol. 96:334-41 38. l-awiun. R. W.. Ross: M. R.. Feingold. J.. Brown. J. I: 1983. EITects of PCB exposure on biochemical and heiualological findings in capacitor workers. Envirtni. Health Pertpect. 611:163-84 $9. Chase. K. H.. Wmig. U.. Thomas. D.. Slal. II. W.. Ilcrncy. II. W.. Simon. R. K. 1982. Clinical and nieiaholic abmvnialilies associalcd with occupa tional exposure ui polychkinaaled bi phenyls. J. Occup. Med. 24:109-14 6(1. Akagi. K.. Okumun. M. 1985. Association of blond pressure and PCB level in Yushu patients. Enrirtm. Health Pertpect. 59:37-39 61. Brown. J. F. Jr. 1984. Polychlorinated biphenyl (PCB) partitioning between adipose tissue and serum. Bull. Euviruu. Coutam. Toxical. 33:277-80 62. Emmett, E. A. 1983. Polychlorinated biphenyl exposure and effects in trans former repair workers. Envieou. Health Perspcd. 60:185-92 63. Humphrey. H. E. B.. Price. H. A.. Hudd. M. I. 1976. Evaluation of changes of the level of polychkirinaled biphenyls IPCIII in human tissue. Hoot Rep. EDA Contract No. 22J-7J-22W. Washington. DC: DHEW. FDA M. Baker. E. L. Jr., l-andrigan. P. ).. Clucck. C. J.. Zack. M. M. Jr.. Liddlc. J. A., cl al. 1980. Metabolic con sequences of exposure to polychlori nated biphenyls in sewage sludge. Am. J. Eitidfitiiol. 112:333-63 65. Warshaw, K.. Fishhein. A.. Thorahm. J.. Miller. A.. Sclikoll. I. J. 1979. De crease in vital capacity in PCB exposed workers in a capacitor manufacturing facility. Ann. N. Y. Acad. Sci. 320:277 83 66. Sak. M . Ahlers. I. 1977. Sennit lipid changes under conditions of occupation al exposure tu chlorinated biphenyls. Cetk. Peritutlot. 52:62-65 67. Kimbrough. R. D- 1980. Occupational exposure. See Ref. 26. pp. 373-97 68. Junes. J. W.. Aldcn. H. S. 1936. An ucnclimn dcrmalergosis. Arch. De rmatol. Syphilol. 33:1022-34 69. National Institute few Occupational Safety and Health (NIOSH). 1977. Criteria for a recommended standard. Occupational exposure hi polychlori nated biphenyls (PCBs). Washington. 1X2: Superintendent of Documents. US Govt. Printing Office. USDHEW tNIOSHl Publ. Nu. 77-225 70. Meigs. J. W.. Album. J. J.. Kanin. B. I. 1954. Chkmcnc from an unusual ex posure to Aruchhir. J. Am. Med. Attoc. 134:1417-18 71. Ouw. H. K., Simpson.G. R..Siyali, D. S. 1976. Use and health effects uf Arodot 1242. a polychlorinated biphenyl in an electrical industry. Arch. Eavirou. Health 31:189-94 72. Alvares. A. P.. Fischbein. A.. An derson. K. E., Kappas, A. 1977. Altera tions in drag metabolism in wsirfceis ex posed to polychlorinated biphenyls. CJin. Pharmacol. Ther. 22:140-46 73. Taylur. P R..Lawrence.C. L..Hwang, H. L.. Panetsoa. A. S. 1984. Polychlorinaicd biphenyls influence on binh- wcighl and gestation. Am. J. Public Health 74:1133-34 74. Jacobson, J. L.. Jacobson, S. W.. Schwanz. P. M.. Fein. G. G., Dowler, J. K. 1984. Prctulal exposure to an en vironmental toxin: a lest of the multiple effects model. Dev. Ptychid. 20:523-32 75. Fein. G. G.. Jacubson. J. L.. Jacobson. S. W.. Schwanz. P. M.. Dowler. J. K. 1984. Prenatal exposure to polychlori nated biphenyls: effects on birth size and gestational age. J. Pediatr. 102:313-20 76. Jacubson. S. W.. Jacobson. J. L.. Schwanz. P. M.. Fein. G. G. 1983. Intrauterine exposure of human new borns hi PCBs: measures of exposure. In PCBt: Human and Environmental Hazard't. cd. F. M. DTlri. M. Kamrin. Rp. 311-43. Boston: Buttcrwonh ogan. W.. Gladen. B. 1982. Duration of breast-feeding and environmental contaminants in milk. Am. J. Epidemud. 116:565A 78. Kogan. W. J.. Gladen. B. C.. McKin ney. J. D., Caireias, N.. Hanly, P., cl al. 1986. Polychlorinated biphenyls (PCBs) and dichlorudiphenyl dichloroeihcne (DDE) in human milk: effects uf maternal factors and previous lactation. Am. J. Public Health 76:172 77 79. Gcllcit. R. J.. Heinrichs. W. L.. Swerdloll. R. S. 1972. DDT homologs. Esirogen-like effects on the vagina, uterus, and pituilaiy of the rat. Endocrinology 91:1093-100 81). Bahn. A. K.. Rosenwaike. I.. Herr mann. N.. Grover. P.. Stellman. J.. O'Leary. K. 1976. Melanoma after ex posure to PCB. N. Engl. J. Med. 293:430 a N O Oz XT. CO u LEXOLDMONOOI882 110 KIMBROUGH 81. Brown. D. P.. Jones. M. 1981. Mortal ity and industrial hygiene study of work ers exposed to polychlorinated bi phenyls. Arch. Environ. Health 36: 120-29 82. Brown. 0. P. 1986. Mortality of work ers esposed to polychlorinated biphe nyls--an update. Arch. Environ. Health In press 8). Benazzi. P. A., Zocchelti, C., Gucr- cilena, S.. Foglis. M. D.. Pesaturi. A.. Ribahli, L. 1981. Mortality study of male and female workers exposed to PCBs. Presented at /nr. Symp. Prev. Occup. Cancer. Helsinki 84. Kutaisunc. M . Morikawa. Y., Hindu- la. T., Nishizumi. M.. Kobchi, S.. cl al. 1969. An epidemiologic study on Yusho or chlorobtphenyls poisoning. Fukuoka Acta tied. 60:513-32 (In Japanese) 85. Miyata. H.. Kashimoto. T.. Kunita N. 1977. Detection and determination of polychlorinated dibenzofurans in normal human tissues and Kancmi rice oil caused Kanemi Yusho. J. Food Hvg. Sac. 18.260-65 ' 86. Butter. H. R.. Rappe, C.. Gara. A. 1978. Polychlorinated dibenzofurans (PCDFs) found in Yusho oil and in used Japanese PCB. Chemosphere 5:459-49 87. Miyata. H.. Kashimoto. T. 1978. Stud ies on the compounds related to PCB (IV). Investigation on polychlomdi- benzofuran formation. J. Food Hvt. Sac. 19:78-84 (In Japanese) ' 88. Kamps. L. V. R.. Trotter. W. J.. Young.S. J..Carson.I. /..Roach. J. A. G., el al. 1978. Polychlorinated rpoter- phenyls idenlilicd in rice oil associated with Japanese 'Yusho* poisoning. Bull. Environ. Cantata. Toxicol. 20:589-91 89. Kuratsune. M. 1980. Yusho. See Ref. 26. pp. 287-302 90. Kikuchi, M. 1984. Autopsy of patients with Yusho. Am. J. Ind. Med. 5:19-30 91. Urabe. H.. Asahi. M. 1985. Past and current dermatological status of Yusho patients. Environ. Health Perspect. 59.11-15 92. Masuda, Y. 1985. Health status of Japanese and Taiwanese after exposure to contaminated rice oil. Environ. Health Perspect. 60:321-25 93. Hsu. S. T.. Mac. I.. Hsu S.. K-H.. Wu. S-S. Hsu. NH-M. Yeh. C-C. Wu. S-B. 1985. Discovery and epidemiology of PCB poisoning in Taiwan: a four-year follow-up. Environ. Health Perspect. 30:5-10 94. Chen. P. H.. Wong. C. K.. Rappe. C., Nygren. M. 1985. Polychlorinated biphenyls, dibenzofurans pod quaier- phenyls in toxic rice-bran oil and in the blood and tissues of patients with PCB poisoning (Yu-Cheng) in Taiwan. En viron. Health Perspect. 59:59-65 95. Abdel-Hamid, F. M.. Moore, J. A., Matthews, H. B. 1981. Comparative study of 3,4,3' ,4* telrachlorobiphenyl in male and female rats and female mon keys. J. Tasicol. Environ. Health 7:181-91 96. Ham. T. 1985. Health status and PCBs in blood of workers exposed to PCBs and of their children. Environ. Health Perspect. 59:85-90 97. Rappe. C.. Bergtgvist. P. A.. Hamson. M.. Lars-Owe, K.. Undstrbm, G., ct al. 1984. Chemistry and analysis of poly chlorinated dioxins and dibenzofurans in biological samples. Banbury Report IB: Biologies! mechanisms of dioxin action. Cold Spring Harbor Laboratory 17-25 98. Scheeler. A.. Schaffner, F., Tieman, T.. Taylor. M. 1984. Ulirastroctural al terations of liver mitochondria in re sponse to dioxins, furans. PCBs, and btphenylenes. See Ref. 42. pp. 177-90 99. Chang. K. J.. Hsieh. K. H.. Tang. S. Y., Tung. T. C. 1982. Immunologic evaluation of patients with polychlori nated biphenyl poisoning: evaluation of delaycd-tyne skin hypersensitive re sponse and its relation to clinical studies. J. Toxicol. Environ. Health 9:217-23 100. Chang. K. J.. Hsieh. K. H.. Lee. T. P.. Tang. S. Y.. Tung, T. C. 1981. Im munologic evaluation of patients with polychlorinated biphenyl poisoning:.de termination of lymphocyte subpoputalions. Toxicol. Appl. Pharmacol. 61(11:58-63 101. Chen. R. C.. Tang. S. Y.. Miyata. H.. Kashimoto, T.. Chang. Y. C.. ct al. 1985. Polychlorinated biphenyl poison ing: correlation of sensory and motor nerve conduction, neurologic symp toms. and blood levels of polychlorina ted biphenyls, quanerphenyls, and di benzofurans. Environ. Ret. 37, 340-48 102. Murai, Y.. Kuroiwa. Y. 1971. Periph eral neuropathy in cMocobiphenyls poisoning. Neurology 21:1173-76 103. Seppalainen. A. M., Vuojolahti, P.. Elo, O. 1985. Reversible nerve lesions after accidental polychlorinated biphenyl exposure. Scant! J. Work Environ. Health 11:91-95 104. Chang. K. J.. Lu, F. J.. Tung. T. C.. Lee. T. P. 1980. Studies on patients with polychlorinated biphenyl poison ing. 2. Determination of urinary copro porphyrin. uroporphyrin, delta-amino levulinic acid and porphobilinogen. HUMAN HEALTH EFFECTS OF PCBs & PBBs 111 Res. Commun. Chem. Pathol. Pharma col. 30(31:547-54 105. Fries. G. 1985. The PBB episode in Michigan: an overall appraisal. CRC Crit. Rev. Toxicol. 16:105-56 106. Anderson. H. A.. Lillis. R.. Selifkoff. J. J.. Rusenman. K. D.. Valciukas, J. A., ct al. 1978. Unanticipated preva lence of symptoms among dairy farmers in Michigan and Wisconsin. Environ. Health Perspect. 23:217-26 107. Valciukas. J. A.. Lillis. R.. Wolff. M. S.. Anderson. H. A. 1978. Comparative ncurobetuivioral study of a ptdybrominated biphenyl-exposed population in Michigan and a nonexposed group in Wisconsin. Environ. Health Perspect. 23:199-210 108. Bekesi. J. G.. Holland. J. F., Amlcrson. H. A.. Fischbcin. A. S., Rim. W., ct al. 1978. Lymphocyte function of Michigan dairy farmers exposed to polybrominated biphenyls. Science 199:1207-9 109. Kteiss. K.. Roberts. C.. Humphrey. H. E. B. 1982. Serial PBB levels. PCB levels, and clinical chemistries in Michi gan's PBB cohort. Arch. Environ. Health 37(3>:l4l-47 110. Tuey. D. B.. Matthews. H. B. 1980. Distribution and excretion of 2.2'.4.4,.5.5' bexabromobiphenyl in rats and man: pharmacokinetic model predtcliuns. Tasicol. Appl. Pharmacol. 33(31:420-31 111. Kimbrough. R. D.. Groce. D. F.. Kor- vcr. M. P.. Burse. V. W. 1981. Induc tion of liver tumors in female Sherman strain tats by polybtuminaied biphenyls. J. Natl. Cancer Inst. 66.535-42