Document 65q9E0jnEj0pKGa7oZ3j7yMxd

CHAPTER 2 AN OVERVIEW ON THE HEALTH EFFECTS OF HALOGENATED DIOXINS AND RELATED COMPOUNDS-- THE YUSHO AND TAIWAN EPISODES G. Reggiani F. Hoffmann-La Roche & Co. Ltd. Basel, Switzerland I. INTRODUCTION There are numerous health effects and symptoms and signs reported in humans following the exposure to halogenated diox ins and related compounds. They are given in the tabulation below and have been adapted after Kimbrough (1) . System or finding Signs of lesion Chemical product General Weakness, headache, weight loss. impotence, insomnia loss of appetite, nausea, abdominal pain ACCIDENTAL EXPOSURE TO DIOXINS 39 Technical PCP, poly chlorinated naphtha lenes , -dibenzodioxins, -dibenzofurans, -biphenyls, polybrominated biphenyls, diben zofurans ISSN0.(2-191160-9 Hepatic Chlor- or bromacne, hyperpigmentation, hair loss, nail growth, porphyria Liver enlargement, impaired liver function, fatty degeneration, fibrosis, hepato cellular necrosis Reduced vital capacity, chronic bronchitis Muscle -and joint pain, numbness and spasm in the limbs, reduced sensory and motor conduction velocity, sensory disturbances (taste, hearing, smell, sight impairment) Chemical product Penta- and hexachloronaphthalenes, tetrachlorodibenzodioxins, tri- and tetrachlorodibenzofuran, hexaand heptachlorobiphenyls , tetxabromodibenzofuran, polybrominated naphthalene Chlorinated naphthalenes, tetrachlorodibenzodioxin, technical pentachloropheno1 Technical pentachlorophenol, polychlorinat ed and polybrominated biphenyls Polychlorinated bi phenyls , -dibenzofurans, -dibenzodioxin Polychlorinated biphenyls and -dibenzofurans Technical pentachlorophenol, polychlorinated biphenyls, -dibenzo furans and -dibenzo dioxin 2 HeaJlh EiTe System of findi Stomato logical alterati Clinical chemist i labrate finding* II C. Reggiani duct achlorotetra- dioxins, achlorohexarobi'abromopoly.phthalene phthalenes, Ubenzotnical Phenol :achloro:hlorinatominated :d bilenzosnzodioxin ad biphenyls furans zachlorochlorinated dibenzodibenzo- 2 Health Effects of Halogenatcd Dioxins and Related Compounds 41 System of finding Stomato logical alterations Clinical chemistry laboratory findings Signs of lesion Pigmentation of oral mucosa, anomalies in number and growth of teeth Increased serum triglycerides, decreased serum bilirubin Chemical product Polychlorinated bi phenyls and -dibenzofurans Polychlorinated bi phenyls , -dibenzofurans and -dibenzodioxins The clinical features have been observed mainly in cases of occupational exposure where dosage is unknown, route of administration is assumed, latency period to onset of toxic effects is poorly defined, length and intensity of exposure is seldom recorded or available, number of cases entering the study is imposed by circumstances beyond the investigators, baseline data for control are often not adequate, and a com parison with a matching sample is lacking. In addition, the sample consisted mainly of adult males. The majority of the cases of exposure reported have been cases of serious toxicity, i.e., of unquestionable toxic mani festations which assume exposure and, consequently, absorption of many times the threshold dose level to produce toxic effects, Furthermore, these have been cases of mainly acute exposure following a change in a manufacturing procedure or a lack of safety measures or accidental, exposure. These health effects might somewhat provide indications for long-term exposure to low doses of the same chemicals or for their effects when stored in the body in trace quantities. Minute amounts of many chemicals can be detected in the human body using very sensitive analytical chemical methods as these methods improve (2) the detection of very low quan tities of a chemical in body fluid should provide additional data on adsorption (Table I) and consequently exposure. II. DETERMINATION OF EXPOSURE Contact with one or several chemicals assumes an exposure to and consequent absorption of these chemicals. However, the assessment of the level of exposure and of the quantity of the actual absorption are not as easy to ascertain. Terms such as high, medium, and low exposure (dosage exceeding background r*J*3 *mmw tiravi TABLE I. Trace Analysis and Its Implications (Tetrachlorodibenzodioxins)a Amount Weight (gm) Trace contamination level per gm Number of molecules Level of analytical expertise required to detect residues Milligram (mg) Microgram (ug) Nanogram (NG) Picogram (pg) Femtogram (fg) Attogram (ag) Mologram ? 10"3 10-6 10"9 10-12 10-15 10-10 10-21 1 ppm 1 ppb 1 PPt 1 ppq lOl8 1015 1012 109 106 103 (1) Titration Spectrophotometry Gas chromatography Mass spectrometry Mass spectrometry Mass spectrometry Mass spectrometry After Cairns et a l 1980 (2). expc (0 O Oi H) k H pi M iH iO ^ M O U * H H- H) X* HI p < g It p n o b rtrtltltH i wiopip'tigH-t .n'*S H M Bl a 15 H m i; g pi pi h in >d >i u o m 3n?(r5*^|1! H3- 5Ot j 5a ttti' c 5 n i o exposure) are all relative and vary quite widely as noted in the literature. Of the clinical signs and symptoms of toxicity, only chloracne is sufficiently specific for and, therefore, indicative of exposure (3). However, not all chemicals are chloracnegenic and those which are have different levels of acnegenicity. Furthermore, the skin lesions tend to be spontaneous and com pletely heal. Thus, they may have occurred and disappeared and are not observable at the time of clinical examination. In addition, in its mildest form (a few blackheads on the face around the malar region) , it can escape detection by the ex amining clinician. Finally, the organ susceptibility in man may differ from individual to individual, thus determining whether one develops other toxic effects, irrespective of chloracne. The detection of the chemical(s) in body fluids and tissues will provide a nfew dimension in the assessment of exposure i.e., the process, of uptake of the chemical from the environ ment into the systemic circulation. Given the parameters ex posure, the detection of the chemical in the blood and urine will provide the evidence that the chemical was not only avail able but also absorbed. The improved sensitivity of technical analysis will also open the way to the study of the kinetics of the chemical in the human body. Thus, time and level of exposure can be calculated and correlated with the clinical signs and symptoms of toxicity. III. RISK ASSESSMENT OF LOW DOSES Experimental data (4-9) have shown that for some chemicals effects are linked with the structure of each isomer. One iso mer can induce enzyme production and acnegenic properties at very low doses, while a change in the position of one chlorine atom in the molecule in another may not produce these effects at all. With the exception of 2,3,7,8-TCDD very little is known about the toxicity of these other chemicals. In fact, tor the majority of these chemicals, long-term toxicity stud ies and studies on their effects on reproduction are still lacking. There is considerable concern in the medical community about the proper methodology to-be used for the study of the long-term effect of exposure to low doses of these chemicals. Although the present diagnostic means are adequate for identi fying subtle clinical., histologic, and biochemical changes, doubts are reaised about the ability to understand the meaning of these findings and their role in predicting the long-term effects on human health. There is uncertainty concerning 'Wat 44 G . Reggiani whether borderline changes observed in morphologic (liver his tology) and blood chemistry tests, immunological tests, cyto genetic tests, nerve conduction velocity tests, and respira tory function tests are the expression of sirple physiological adjustment or an unfavorable reaction of an organ which will lead eventually to irreversible damage. In the great majority of cases the chemically induced changes in some functions or structures are probably only an indicator of an exposure to `quantities of chemicals which are below the toxic level. The correlation of the level of the chemical(s) in body fluids, may provide a sensitive system that can be used to monitor ex posure of humans to the chemicals and the importance of that exposure in the long-term susceptibility to disease. They are not, however, a symptom of disease. Failure to agree on the meaning of these findings masks the definition of "observable adverse health effects." IV. HUMAN RESPONSE TO EXPOSURE In the final analysis it is heavy exposure to chemicals that provides an opportunity of measuring human response. Ex trapolation from one dose level to another might provide data for estimating the- human risk associated with low exposure levels. There have been a number of episodes of heavy exposure to chemicals. They have involved both sexes of all ages, and a sizable portion of the population. In addition, the dose, route of absorption, and length of exposure could be somewhat-' defined. A number of these cases are presented below. In 1973, in Michigan, over 10,000 residents were exposed to polybrominated biphenyls (PBBs). The chemical was absorbed through the consumption of meat, milk, and other dairy products. Of this population,, the level of exposure for 4000 residents was determined, although the accuracy of these results were not definitive. The PBB serum levels ranged from 0 to 1900 yg/liter but the daily and total consumption was not established (10,11). Thus far, it has not been shown conclusively that PBBs caused illness in most of the exposed population. In 1976, in Seveso, Italy, over 5000 residents were ex posed to 2,3,7,8-TCDD. For several years, these residents have lived and are still living in a territory contaminated with a known amount of the chemical. The daily and total amount of exposure and the route of absorption of this chemical are not known. However, chloracne was observed in a small subgroup of the population, although this health effect does not provide conclusive information about the route of absorption, the daily intake, the length of exposure, and the total dosage (12). we: l2a2i are (abt exec man te fee poi ran. 196 ted 200 an.* pre and the cal 9^ Ja; sori< in. pe ri P of st tt tl w! a I t G. Reggiani (liver hisits, cyto respira/siological lich will it majority rtions or isure to avel. The fluids, monitor exe of that They are ae on the observable hemicals pans. Exavide data xposure xposure to es, and a e dose, e somewhat ow. e exposed as absorbed 'iry products. residents Its were not 1900 pg/liablished ely that .on. were exsidents have .ted with a amount of :al are not subgroup of t provide n, the daily -e (12). 2 HcaJih Effects of HaJogenated Dioxins and Related Compounds 45 At the Love Canal in Niagara Falls, New York, 239 families were exposed to more than 200 organic chemical compounds in a landfill. These 239 families had to be relocated. Of the 22,000 tons of chemical wastes buried in the landfill, there are 200 tons of TCP with approximately 300 ppb of 2,3,7,8-TCDD (about 60 g). The conditions of exposure of this episode are exceedingly complex. The population at risk was exposed to many chemicals and one cannot say if one, several, or an in teraction between them, are likely to be responsible for ef fects observed (13). Approximately 1600 residents w ere exposed to a mixture of. polycFTLonnateQ blpnenyls (PCB's) , polychlorinated dibenzofurans (PCDF1s) and polychlorinated auaterohenvls (PCO's) in "1968 in South "Japan. These residents consumed Yusho contaminated with an edible^ oil (Yusho poisoning) (14-17) . A similar event occurred in 1978 in central Taiwan, where 2000 residents were exposed to the same mixture of PC3's, PCDF's and PCQ's (18,19). In the data available, only the last two episodes provide precise information about the daily intake, length of exposure and total dosage, and latency to onset of effects and time of the appearance of the single symptoms and signs of the clini cal picture of* the poisoning. These episodes are discussed, in greater detail in the next section. V. THE 1968 YUSHO AND THE 1978 TAIWAN EPISODES A sporadic outbreak of a peculiar skin disease in South Japan between March and October 1968 was the beginning of the .so-called "Yusho disease." It was caused by contamination of rice bran oil by Kanechlor, a commercial brand of a polychlor inated biphenyl which leaked from a heating pipe. About 1600 people were identified as having ingested the contaminated rice oil. On June 7, 1968 a 3-year-old girl was admitted to the De partment of Dermatology of the Kyushu University with symptoms of chloracne. Her parents and her elder sister showed the same symptom, leading health authorities to become suspect of the cause of the disease. Similar cases were uncovered in the weeks and months that followed (14-17).. In Taiwan, during the last months of 1978, for reasons which still require explanation, rice cooking oil produced by a local company was contaminated with material containing PCB. In March, 1979, the first cases of unusual ocular and derma tologic manifestations were observed. They occurred mainly in the central region of Taiwan and their number increased rapidly to about 2000. In October, 1979, the Central Health */ 46 C . ReggiAni Authorities in Taiwan issued a statement announcing that the cause of the disease had been the ingestion of edible oil con taminated with PCB (18,19). VI. COMPOSITION OF THE TOXIC MATERIAL-- QUANTITY AND DURATON OF INGESTION 'The toxic chemicals involved in the two episodes discussed in the previous sections (20-25) were polychlorinated biphenyls (PCB's), polychlorinated dibenzofurans (PCDF's) and polychlor.inated quaterphenyls (PCQ's). heir chemical structures are presented below. f~\ Cl, 'O' `Cl* Polychlorinated dibenrofurao (PCDF) Polychlorinated biphenyl (PCB) Polychlorinated quaterphenyt (PC) There are 209 possible isomers for the PCB's and 135 for the PCDF's. As can be seen in Table II, the concentration of these chemicals in the rice bran oil differed for Japan and for Taiwan. The toxicity of the oil from Japan was 25 times higher than that from Taiwan. Another difference was the dis tribution of the isomers. In Taiwan there was a higher con centration of the tetra- and pentachlorobiphenyls whereas in Japah, it was the hexa- and heptachlorobiphenyls. As for the PCDF's, the toxic oil contained isomers with 5 chlorine atoms . Health E TABLE I ted O i l ' -----Japan Taiwan TABLE : P C B 's PCDF'S with - was t; T) epidej 325 c showe consu pared cals 1:4, firsin T shor This quar day: G. Reggiani iat the oil con- -ND DURAdiscussed 1 biphenyls xDlychlor\res are 135 for ation of an and 5 times the diser con reas in for the rine atoms Health Effects ofHalogcnated Dioxinsand Related Compounds 47 TABLE II. Concentration of PCBs, PCDFs and PCQs in Contamina ted Oil in Japan and in Taiwan PCBs (ppm) PCDFs (ppm) PCQs (ppm) Japan 920 5 866 Taiwan 40.5 0.26 36.5 TABLE III. Isomers of Toxic Oil in Japan and Taiwan Isomer Japan Taiwan PCB's PCDF'S Tetrachlor Pentachlor Hexachlor Heptachlor Trichlor TetrachlorPentachlorHexachlor- + ++++ ++++ + ++ +++ -H- +++ ++++ + + ++ ++++ ++ with smaller amounts of the other congeners (Table III). This was true of the oil from both Japan and Taiwan. The quantity of the ingested chemicals was calculated by epidemiological surveys (17,19/26), carried out in Japan in 325 cases and in Taiwan (924 cases)(Table IV). The surveys showed that there was a marked difference between the average consumption of rice oil per month in Japan (0.25 kg) as com pared to Taiwan (1.42 kg). The ratio of contaminating chemi cals ingested per month in Japan as compared to Taiwan was 1:4, respectively. The latency period to the onset of the first clinical symptoms was shorter in Japan (2 months) than in Taiwan (2.7 months) . The total length of intake was also shorter in Japan (3.5 months) than in Taiwan (8.7 months). This is probably due to the fact that in Japan only a limited quantity of the rice oil was contaminated and only for a few days at the beginning of February, 1968. *As soon as this TABLE IV. Quantity of Ingested Toxic Oil and Duration of Ingestion3 Quantity in rice oil fppm=vng//cg; Quantity rice oil used per month (mg) Ingested during latency period (mg) Ingested for total length of intake (mg) Dose per day (mg) Dose per kg body weight Japan Taiwan Japan Taiwan (0.25 (1.42 Chemicals kg) kg) Japan Taiwan Japan Taiwan Japan Taiwan Japan Taiwan (2 (2.7 (3.5 (8.7 months) montAs) months) months) PCBs 920 40.5 230 57.5 460 155 805 500 8.4 1.9 0.14 0.03 mg mg PCDFs 5 0.26 1.25 0.37 2.5 0.99 4.4 3.2 0.04 0.01 0.06 0.2 U9 vg PCQS 866 36.5 216.5 57.8 433 140 756 450 7.2 1.7 0.12 0.02 mg mg Values are caiculatecf for average body weight of 60 kg. 325 cases in Japan, 924 cases in Taiwan. Health Efl (32) . rC ohui ph rpo* wro *o h jbt* H ro u p * g H - (0 it iO rhti E) r o w ro *d <; p. 3 01 rprot* frof 8. (pft* ro ro w n01 CrrHaooi roMr3root. H0 i wpons-- h gD_i ro tu Ui H~ji Brno.* sh|U rt ro h --3 rt PO 3 Hi O tro Tpf* P^3* rt ' p<-pP;**rVo) _a rt H p W P* <O ro>to. p- h W rt on< O Hi Hi O H ro o l-< H Health Effects of HaJogenatcd Dioxins and Related Compounds quantity had been consumed there was no additional opportunity for poisoning to occur. In Taiwan, it is not known how much of the rice oil had been contaminated and how long it had been available to consumers. The association of the clinical mani festations with the oil contamination was officially announced in October, 1979, i.e., 8 months after the observation of the first cases. The ratio in the total amount of PC3's, PCDF's, and PCQ's ingested in Japan (875 mg, 4.4 mg, and 756 mg, res pectively) and Taiwan (500 mg, 3.2 m g , and 450 mg, respective ly) is 1:1.6. These last values are probably more instructive with respect to the clinical manifestations observed in both places than the daily dose and the dose per kilogram body weight, where the ratio is roughly 1:4. The route of absorp tion for all cases was the gastrointestinal tract. In Japan it was possible to estimate (17) the incidence (rate) of the clinical manifestations in the population in re lation to dosage (Table V ) . It was shown that the toxic dose 100 (eliciting toxic effects in 100% of the cases) could be reached at a dose of 720-1400 mg PCS's , i.e., 3.6-7.2 mg PCDrs and 570-1100 mg PCQ's, and that below that level the incidence of disease decreased. The clinical severity of the symptoms (27) graded by the dermatologist from 0 to 4 (Grade 0: subjective complaints; Grade 1: pigmentation of skin and mucosae, eye discharge; Grade 2: comedone formation; Grade 3: moderate acne: Grade 4: exten sive acne eruptions) correlated closely in Japan with the to tal amount of oil consumed but not with the amount of oil con sumed per kilogram of body weight per day. The acne lesions in the children were generally milder than those of the adults. The daily amount per kilogram body weight calculated for the children was larger than for the adults but the total amount was less. In Taiwan about ^ of the cases were severe and clinical severity seemed to be higher in the younger age brack ets. There seems to be a direct correlation between total in take of toxic oil and severity of symptoms, but this cannot be confirmed statistically (28-30). VII. CLINICAL SYMPTOMATOLOGY Probably the best (31) clinical picture was provided by the examination of a large Taiwanese family of 27 members, com prising three generations. The living conditions were the sane and the toxic oil consumption had been identical for all members with respect to daily and total intake (Table VI). Table VI shows the clinical symptoms and signs in order of frequency' and time of appearance. Symptoms affecting the eyes were the first to appear after a latency period of 2-3 months (32) . TABLE V. Frequency and Severity of Symptoms in Relation to PCB Concentration in Yushb; Oil Japan 1968-1979* s(Q-O -U -U u r in p- s in rn-C mCD n0 s0r 4J hj O CM u o 1 \D ^ 4J 03) o 01 0 0U O' n O' o oe 03 H o f-4 E <oN E CM vID> c3 10 n K W 01 3 V ot JJ oa. m Jj jOz'H CO m n c\m> tu Q Ua. a "w --iTnT uaai 3 4-> <8 00 r-C E m u IQ w -TOr-vt 2E* rr CrotM~kCmrCI1-M10MDt -r-*t Bl 0 ot u Q 0U 0 O' 6 O' E CM CM r-t t-" rCIM> ID nItD 2 --*H CA3 2:5 nv uu soi sr r-t 2 Health Effects TABLE V I. tions Famil 1979a Symptoms Increased e Swelling of Acneiform e Pigmentatic ( j Pigmentatic | Pigmentatic ! Black colo; Hypoestesi Deformity Pigmentati Numbness c Blurred vi Keratotic and soles Follicular ^Aftei The s< examined . Veteran Gth Depar Biochemis tologic s shows tha ably rela but the c V II). Compr performed were admi General I 2 Health Effects of HaJogcnaicd Dioxins and Related Compounds 51 TABLE VI. Symptoms and Signs of 27 Members of a Three Generations Family 6-7 Months after Beginning of Poisoning, Taiwan, I979a Symptoms and signs Male (IS) Female (12) Increased eye discharge Swelling of eyelids Acneiform eruptions Pigmentation of nails Pigmentation of conjunctivae Pigmentation of lips Black color of nose Hypoestesia Deformity of nails Pigmentation of gingivae Numbness of limbs Blurred vision Keratotic plaques of palms and soles Follicular hyperkeratosis 93.5% 86.6 86.6 86.6 80 80 66.6 66.6 60 53.3 53.3 53.3 46.6 40 91.6% 91.6 83.3 83.3 83.3 66.6 75 75 58.3 66.6 50 50 66.6 41.6 3After Wei-Min Li et al., 1981 (31). The same clinical symptoms were observed in groups of cases examined and treated by the Department of Dermatology of the Veteran General Hospital, Taipei, Taiwan (33) , as well as by the Departments of Ophthalmology, Neurology, Pediatrics, and Biochemistry of the same Hospital. A comparison of the derma tologic symptoms of two similar groups of cases from Japan shows that there are quantitative differences, which are prob ably related to differences in the constitutions of the samples, but the quality of the clinical features was the same (Table VII) . Comprehensive neurological examinations have been (34) performed on 39 cases of a group of 122 severe cases which were admitted to the Department of Dermatology of the Veteran General Hospital of Taipei between January and November, 1980. TABLE VII. Comparison of Dermatologic Signs between 122 Cases in Taiwan (A) and 138 cases in Japan (B) Wale Female Sign A (42) B (72) A (80) B (66) Acneiform skin eruptions (Comedones, follicular hyper keratosis, yellow cysts, abscesses) Dry Skin Deformities of nails Nail pigmentation Black color of nose Hyperkeratosis of palms and soles 71.4% 50.0 68.5 56.7 40.7 19.5 86.1% 77.5% 30.6 22.2 75.0 4.2 1.4 56.0 67.5 61.5 57.3 22.5 77.3% 36.4 24.2 71.2 4.5 4.5 H HI Q H i #1 o Q o ro *< O 0 3 OJ H 3 a H'O H W- 03 Dj 3 H H- 03 n Qi M P B H O 9 g H h - ' Z i Q H> > OH,n>oDOi-jaifiOH 0 330*0p>(i(IlMH til t* pi H* H - *Q i+n h F-ftrtrtH qfcrRit> rt if c r+ H O W 0* 3 " n i0n H Pi ns V , H <3rDt 8 OHtMrrt.- d PH* 3 O a r 3 oo a rPo *d cHon.r+ pap P 3HowNN*' Xto P (1j OP *(ca t(Dn H (nHV & M rt K> Xrt 5Sr* 2 HeaJih Effects of HaJogenaied Dioxins and Related Compounds 53 TABLE VIII. Neurological Signs and Symptoms of 39 Cases Selected in a Group of 122 with Skin Lesions3 Symptoms and signs Number Headache Dizziness Paraestesia, numbness in the distal part of the limbs Hypoaestesia, hypoalgesia in the distal part of the limbs Pain over back, limbs, and orbits Intermittent blurred vision 15 (38.4%) 12 (30.8%) 25 (64.1%) 13 (33.3%) 16 (41.0%) 11 (33.3%) All reflexes were normal-- no pathological reflexes; no muscle weakness or wasting. All had been treated unsuccessfully by local district doctors for 6-10 months. All cases had skin lesions of varying de grees of severity. The 39 cases referred to the Department of Neurology had neurologic symptoms upon clinical examination (Table VIII). The cases presented with a mild peripheral neu ropathy with involvement of primary sensory neurons without inclusion of the motor neurons. There was also a mild involve ment of the central nervous system. The electrographic measure ment of nerve conduction velocity provided evidence that in 35 of the 39 cases not only sensory but also motor conduction ve locity was slightly reduced. Nerves conduct at velocities ranging from 1 to 100 m/sec. The speed of nerve conduction indicates the functional status cf a group of axons. The ve locity in the PCB-contaminated cases was slightly longer than in a control group, averaging 63.3 m/sec. for the motor neu rons and 52.9 m/sec for the sensory neurons, instead of 58.3 and 46.8 m/sec, respectively. The neurological signs and symptoms also indicate involve ment of the central nervous system, although electroencephalographic analysis did not show abnormalities. In Japan, many Yusho patients complained of neurological symptoms, such as, numbness of limbs', hypo- and hyperesthesia, feeling of weakness, and muscle spasm. Reduced sensory nerve conduction velocity was observed in 9 out of 23 cases examined soon after poisoning, while reduced motor nerve conduction ve locity was seen in only two cases. Ultrastructural examination of the sciatic nerve of rats treated with PCB's did not show any change. No abnormalities ____ nical Findings of Newborn Babies of PCB-Contaminated Mothers, Taiwan, 1979 S4 G. Reggiani of the central and peripheral nervous system have been produced in monkeys treated with 0.5 mg PCB's and 2.5 vg PCDF's for 6 weeks (35). Blood chemistry tests were performed (36, 37) on several of the cases admitted to the Department of Dermatology of Tai pei. The level of triglycerides was elevated in all age brack ets. The SGOT and SGPT values, as well as those of the alka line phosphatase, were within or at the upper borderline of the standard values, with the exception of the lowest age brackets (3 months of age). The blood picture and hemoglobin were nor mal. Conventional function tests of the liver (bilirubin, SGOT, SGPT, BSP retention, and others) did not show any abnormali ties of this organ in the Yusho cases. Although few morpho logic changes were detected by light microscopy, the serum tri glyceride was abnormal (38). The increase persisted for a few years (Table IX). In about 20% of the Yusho cases chronic bronchitis was re ported 10 years after the poisoning. The PCB concentration in the sputum and blood of these cases seemed to correlate with the severity of the symptoms. This has not thus far been reported in Taiwan. VIII. PREGNANCY AND CHILD DEVELOPMENT The course of pregnancy and subsequent child development was studied in 5 pregnant women (39,40) from Taiwan, who had conceived at the time they were using PCB-contaminated oil and who developed the most common dermatological manifestations (mucocutaneous pigmentation, acneiform eruptions, deformity of the nails, ocular manifestations, etc.}. Birth was full term in 4 women and 1 month premature in one who gave birth to twins (Table IX). Of the skin lesions, different degrees of generalized pig mentation, and deformity of the nails was observed in all of the babies. Although bone development was normal during the first year of life, the babies suffered from gastrointestinal and respiratory illnesses. They were not breast fed. Skin lesions and pigmentation disappeared 2-5 months after birth in all but one of the babies (case No. 4), who at the end of the first year of life still had no teeth and was unable to sit and to crawl. Retarded teeth and bone development was also observed in Japanese babies bora from mothers who had been poisoned with the Yusho oil (41). A comparison with findings in Japan has shown a similar general picture. Eleven women with overt syn>toms of poison ing and two women married to men who had been exposed to the -- -- G. Reggi; the tn 0Hi *0 oPt0an|.* p tHn' 5P- PH tr pPOtr3pe' rt *: tr hre p tn r0t PI-1 t0n tPn* rt re r3Oei 0Hi fhretr Hi rrMet OpM-' ri rt p3- Hraei tXp3n* HPO3rret trp3PPn+** D<PCH3n* r3retX" H. P3pi HP** O (0 M Pp>* PN(Ot)* 0 v t Up* luto Di rfet ip. j p3 .r--e H (6 W f t - < fhtr rO+ c I 1o11 h->Pgm* Ohfpf tt-. POn*' 03C* 0po(30 * ft it D H yn> tn P3 p Ou r t oa ft P i re a* re nPi O K ipl fl> H i P * I" rt 3 **j re - re &tCDr (D a O P* o HI rt. s cr H op ?r 3* pni re irQo P_ ^ 0 hi w (r<Po H 3 p tn a Hoi *Ho 11 Si ft* ft in p p* P1- (Ti CCt CP O- TABLE IX, Clinical Findings of Newborn Babies of PCB-Contaminated Mothers, Taiwan, 1979 Case number Sex Birth weight (gm) Pigmentation of skin, nails, and mucosae Acneiform eruptions Dry skin desquamation Swelling of Meibomean glands Deformity of nails Black color of nose Neurologic disorders 1 234 Male 3600 Male 2500 Female 2400 Male 3600 + + + ++ + Both 5th fingers + - -- + - - +++ ++ + ++ ++ +++ +Both 5th fingers + - +++ +++ + - 56 Female twins 1800 1000 ++ ++ -++ ++ ++ --- Triglycerides (mg/dl) Iqi(H (PCB+PCDF) Fig. 1. Ten year follow-up of serum triglyceride levels in 24 Yusho patients Standard values: 47-162 mg/dl. n rf h h tr m dr ir ro rt> d K H IV W < f t 3 't I T S h g , R ' 8 - B 3 e: 3H *<a0 (nT> 3Q5 cr<o ip h (D o. ti tr M to & ^ g s e s s s-*s n H r r- Ml l< 2 Health Effects of Halogenatecl Dioxins and Related Compounds 57 poison but who did not present overt symptoms of th disease delivered 11 live borns and two stillborn babies. Of these, 10 had dark brown pigmentation of skin mucosae and'nails, as well as increased eye discharge. Five were considered to be undersized. A battery of laboratory tests, however, showed no abnormalities. The skin,coloration faded within a few months. Another woman suffering from the disease delivered 3 babies within 4 years after the incident occurred. All showed dark brown skin pigmentation. Pigmentation of the skin also occur red in a baby breast fed by a woman affected by the disease. However, no cases of unusual malformations or increased inci dence of spontaneous abortions were reported. IX. METABOLISM The rate of absorption, metabolism, distribution, and eli mination of these chemicals in humans are practically unknown. However, some data concerning blood levels, storage in adipose tissue and liver, and dissipation from these stores are avails able (42-44). The concentration of these chemicals in the blood, liver, and adipose tissue was always higher in cases of poisoning than in the normal population (Table X). Blood PCB levels in normal subjects were below 4 ppb in Taiwan and were about 3 ppb in Japan. Levels in poison cases from Taiwan were ten times higher than normal; they are even higher than the levels found in Japanese cases who showed an average of 6.7 ppb. How ever the interpretation of these results are difficult because measurement of the blood PCB levels in the Japanese cases was performed 5 years (1973-1974) after the poisoning, whereas the same tests were carried out in Taiwan 1 year after the poison ing. X. BLOOD AND TISSUE LEVELS AND RELATIONSHIP OF CLINICAL EFFECTS The correlation between blood levels, severity of clinical synptoms, and blood chemistries has been examined many times, but no close relationship has been found to exist. A high PCB level in blood does not necessarily parallel severity of skin, lesions or rise in the serum triglyceride levels (45,46) . On the other hand, blood and tissue concentrations of the chemi cals seem 'to be higher and remain at that level in exposed as compared to the normal population (Table XI). 9 58 G. Reggiint TABLE X. PCB Levels in Blood of 66 Cases of Contamination in Taiwan, 1979 Number Sex Ape PCB levels (ppb) 26 Male 31.4 (8-60) 47.5 (19-120) 37 Female 25 (12-69) 33.7 (11-71) 2 Female 2 months 18 (15-21) 37 Normal persons 2.9 1.7 (Japan) Normal persons (Taiwan) Less than 4 TABLE XZ. Concentration of PCB in Tissues of Yusho Patients3 Case no. Concentration Time of death or operation Adipose tissue ...... Liver 1 October, 1969 0.1 0.07 2 July, 1969 3.7 0.14 3 July, 1969 15 0.2 4 November, 1969 8.4 0.7 5 December, 1970 0.9 0.07 6 May, 1972 6.5 0.8 7 September, 1972 2.9 8 April, 1977 0.4 0.07 9 March, 1977 1.4 0.06 10 September, 1977 1.1 0.02 Control cases (31) 1973 >1.0 0.05 aAftex Kikuchi et al., 1979 (46). h Eflects of Halogcnaied Dioxins and Related Compounds nemical dissipation time seems to be longer for some rs than for others (Table XII). However, the interpretaof these findings is hampered by the limited number of ases providing tdiis data. Pentachloro isomers apparently longer dissipation times.than tetrachloro isomers. The ibution of the PCBs isomers also differs for cases of poig than for the normals (Table XIII) . In the normal popun there is a greater concentration of pentachloro than hloro isomers, whereas in the Japanese cases the hexao and heptachloro isomers are greater. This finding cornds to the composition of the ingested oil. :i. CONCLUSIONS h e ingestion of PCBs, PCDFs, and PCQs during several is, at an average daily dose of 1.9-8.4 mg PCBs, 0.01-0.04 :DFs, 1.7-7.2 mg PCQ's, and a total dose of 500-805 mg 3.2-4.4 mg PCDF's, and 450-756 mg PCQs resulted in a .fic clinical picture, namely, skin manifestations domina3y acneiform lesions, pigmentation of skin, mucosae, and 5, and ophthalmic lesions, pigmentation of skin, mucosae, lails, and ophthalmic lesions. In 20-30% of these cases, neurologic disorders of sensory and motor nerves were also rved. Conventional liver function tests as well as convenal blood analyses were normal. Serum triglycerides were 2fold higher than standard values. Levels returned to al after 3 to 5 years in the Japanese cases, rhere are some general conclusions which can be drawn from information gained by the two episodes which are possibly d for all chemicals. The primary manifestation of the oning with these chemicals is skin lesions, which is also acteristic of those chlorinated chemicals which are chlorgenic. Impairment of other physiologic functions, such the liver and nervous system, and fat and porphyrin metasm are less frequent and less severe. It can be reasonargued, further, that if the skin lesion does not appear r functions and organs are not affected. Of course, this not hold if the contaminated chemicals are not chloracne(e.g., 2,4-DCP which is specifically neurotoxic and not racnegenic). Skin manifestations heal slowly (47) in about of the cases. In some cases, sequelae remain which can be cted clinically when other symptoms and signs have disared. In addition, symptoms such as ocular lesions, perial neuropathies, liver function impairment, and increase erum triglycerides return to normal from a few months to w years following exposure, depending on the severity of poisoning. As is frequently the case with all these chemi- there is never a large increase in SGOT (SGPT indicates a fa TABLE XIII. Concentration of PCB Isomers in Blood of Yusho Patients and Normal Persons ' in CM PCD isomers 2,4,5, 3' , 4'-Pentachloro 2,3,4, 3' ,4 *-Pentachloro 2'1,4',5'-Hexachloro ~ 2,3,4, 2'',4',5'-Hexachloro 2,3,4, 5'',3'4' -Hexachloro , 2,3,4, 5, 2',4', 5'-Heptachloro 2,3,4, 5, 2' ,3' , 4 1-Heptachloro Yusho patients (N-9)(ppb) 0.13 0.05 N.,D. 0.52 0.22 0.65 0.21 0.28 1 0.07 0.44 * 0.15 0.28 0.09 Normal person (N=17)(ppb) 0.22 + 0.09 0.05 + 0.02 0.15 + 0.07 0.17 + 0.07 0.04 0.02 0.12 + 0.07 0.11 + 0.04 aDetection limit - 0.01 ppb. N.D. = less than 0.01 ppb. ^After Masuda, 1980. 62 G . Reggiani important hepatocellular lesions). There seems to be a cor relation between severity of the clinical symptomatology and the amount of chemicals accumulated in the body. '-' Inferences about reproductivity are limited by the small number of cases reported. Conception seems to be possible even in cases of, overt poisoning. Pregnancy seems to follow a normal course. Early fetal losses (abortions) have not been reported, but some late fetal losses (stillborn) may be related The newborn child is extensively exposed to the chemicals through the mother which leads to specific skin manifestations, to decreased birth weight, and retarded development, in some cases. Congenital anomalies have not been reported, with the exception of deranged tooth eruption and growth as a late manifestation. The immune response of the newborns has not been examined. The rather high frequency of acute infections during the first year of life points to an impairment of immune capability. In the Japanese episode the toxic oil contained PCB's with mostly 6 and 7 chlorine atoms isomers and only small amounts 5 chlorine isomers. More than 40 isomers of PCDF have been found in the Yusho oil (48), including as their major components the highly toxic 2,3,7,8-tetra- and 2,3,7,8-tetrachlorodibenzofurans. The ratio of PCB's to PCDF's is 200 for Yusho oil (Table XIV), while the ratio is 50,000 -for kanechlor. The ratio in adipose tissue is practically the same as in the ab sorbed product, but it is five times higher in the liver. The clinical features of the episodes in Japan and Taiwan are probably the result of a concurrent toxicity of the speci fic isomers of the mixture ingested. The dibenzofurans are probably mainly responsible for the chloracne lesions. Poly chlorinated dibenzofurans have a strong chloracnegenic effect, whereas polychlorinated biphenyls are only weakly chloracnegenic. It is not possible to assess the risk of the chemicals in volved in these episodes. The LD^O for 2,3,7,8-TCDF varies from 5 to 10 yg/kg in the guinea pig to over 6000 yg/kg in the mouse, with rabbit and monkey in between (49). The total amount of PCDF^s absorbed by the Yusho and Taiwan cases is 3-4 mg; the daily dose varied between 0.01 and 0.04 mg, i.e., 0.20 .6 yg/kg- . Tetra and penta isomers of the PCDFs prevailed in the mix ture but the actual amount of the 2,3,7,8-TCDF isomer is not known. It is most probably less than the dose which is lethal for the guinea pig. The acute and chronic experimental toxicity of the mixture of chemicals involved in the Yusho and Taiwan episodes is not available. An extrapolation of some of the data produced by the study of some isomers has only limited value because a TAB in Kan' Yusl G. Rcggiani be a cor:ology and the small possible to follow ave not been ay be related, e chemicals nifestations, t, in some d, with the a late en examined, ng the first ability. :& PCB's with ;11 amounts have been ijor components :hlor od ibenzo:usho oil >r. The , in the abi liver. i and Taiwan of the specifurans are ions. Poly genic effect, chloracne- chemicals inCDF varies pg/kg in the he total cases is 3-4 'g, i.e., 0.2- *3 in the mixomer is not ;ich is lethal <f the mixture .sodes is not produced by because a 2 Health Effect* of Halogenatcd Dioxin* and Related Compounds 63 TABLE XIV. Concentration of PCB's and PCDF's and Their Ratios in Yusho Cases3 PCBs (ppm) PCDFs (ppm) Patio (ppm) .Kanechlor. (unused) 1,000,000 20 50,000 Yusho oil " 1000 5 200 Yusho patients consumption (mg) adipose tissue (ppm) total (mg) 875 1.3 13 4.4 0.009 0.09 198 144 Liver concentration (ppm) total organ (mg) 0.05 0.075 0.0013 0.0019 39.5 After Masuda et al.f 1980 (22) change in the number and position of the chlorine atoms in the molecule can modify the clinical picture. Therefore, not only the quantitative but also the qualitative prediction of the toxic effects in humans of PCDFs and PCBs is questionable, if the composition of the mixture varies. The impact of physical factors, i.e., heat on the technical products, will probably produce a constant change in ratio of the chemicals and of their isomers in the mixture and, therefore, their toxicity. The risks and health hazards of each case.must be assessed on its own merits. SUMMARY Quantitative toxicology in animals has provided accurate and comprehensive data for some of the halogenated dioxins and^ related compounds, For many others it has only partially been carried out or is still lacking, In humans the information is lagging behind for several reasons, The determination of exposure should be improved, taking advantage of the increased sensitivity of analytical methods for assessing evidence of absorption in man. Clinical symptomatology should be estimated in relation to the kinetics of the chemicals in the body. Signs of physiological adjustments of organs and functions involved in the detoxification process should not be considered an ad- # 64 C. Reggni verse or irreversible health effect. Precise information about daily intake, length of exposure and total dosing, and latency to onset of effects and time of appearance of the symp toms, and signs of the clinical.picture of the poisoning are available only rarely Human'response to these *chemicals is, at best, measured,.during episodes of heavy exposure. The PCB's, PCDF's and PCQ*s poisonings in Japan (Yusho) in 1968 and in Taiwan in 1979 offer opportunities to correlate clinical features to the dose.absorbed and stored in the body. The follow-up studies of the.Yusho_cases also provide informa tion about idiecours and evolution of the symptomatology. The health`effects observed, in the two episodes can be in dicative for the effects to be expected in cases of long-term exposure to the same chemicals at different dose levels. REFERENCES' . -a Kimbrough, R. D. (1980)./Occupational exposure. "Halo- genated Biphenyls, Terphenyls, Naphthalenes', Dibenzodioxins . and Related Compounds". Elsevier North Holland, Amsterdam. Cairns, T., et al. (1980). .-Review of the dioxin problem. Mass spectrometric- analysis-of-TCDD.in environmental samples. '^S i o m e d M a s s .Spectr-, 7 11-12, 4B4-492. - 3. Crow, K.`'{1981)-.- Thei-cutanec aid 'fystematic signs of poisoning with- chloracnegens i n .the* Kumah'subject. Symp. Chlorinated Dioxins -and-Related Compounds, October 25-29, 1981, Arlington;;Virginia 1=,^ Poland, A'.' and Glover; E. (1977). Chlorinated biphenyl induction of'aryl hydrocarbon-hydroxylase activity; a - study of structure-activity relationship. Mol. Pharmacol. 13, 924-938.' \c,\ *'"V' '7 Poland, A; a n d ' G l o v e r E . -(1980) .... TCDD: studies on the mechanism of action. '.Develop, Toxicol. Environ. Sci. 6, 6 . 223-239. . ^ > .3 /* Poland, ^A~. et^al. (1976) Stereospecific, high affinity binding" of'TCDD by hepatic; cytosol.^*7* Biol. Chem. 251, 16, 4936-4946. ' ,%*,"^ 7 *^r Poland, A.- et al.. (1979) Studies on the mechanism of action of chlorinated dibenzodioxins and'related compounds. Ann. N. Y.' Acad. Sci., 320, .214-230. .Baars, A. J- et al.-(1960). Induction of rat hepatic glutathione-transferase,-activities by 2,3,7,8-TCDD. Proc. Workshop Impact of Chlorinated Dioxins and Related Com pounds' on the Environmentr Oct* 22-24, 1980, Rome. 9. Leng,. M. (1979) . Comparative toxicology of various chlor inated dioxins as related to chemical structure. CIPAC Proc. Symp. Ser. 1. 2 He; 10. 11. 12. 13. 14. ' 15. 16. 17. " 18. ' 19. * '20"" .21. , 22. 23. 24. C. Reggiani :ion ig, and the sympig are zals is, iusho) in relate the body. informalogy, an be inong-term .Is. "Haloanzodioxins Amsterdam. problem, ntal gns of t. Symp. er 25-29, iphenyl ty; a Pharmacol. a on the . Sci. 6, affinity "jem. 251, - nism of d compounds. epatic CDD. Proc.' ted Comme. ious chlor- CIPAC 2 Health Effects of Haiogenated Dioxins and Related Compounds 65 10. Landrigan, P. J. et al. (1979). Cohort study of Michigan residents exposed to PBBs. Epidemiologic and immunologic findings. Ann. H. Y. Acad. Sci. 320, 284-294. 11. Billiant, L. B. et al. (1978). Breast milk monitoring to .measure Michigan's contamination with PBB. Lancet, 643- 646 (23 Sept.). 12. Caramaschi, F. et al. (1981). Chloracne following environ mental contamination by TCDD in Seveso, Italy. Intern. J. Epidemiol. 10, 2, 135-143. 13. Kimbrough, R. D. (1981). Studies of human populations exposed to environmental chemicals. Workshop Assessment of Multichemical Contamination, April 28-30, 1981, Milan, Italy. 14. Harakuni, U. et al. (1979). Present state of Yusho pa tients. Ann. H. Y. Acad. Sci. 320, 273-283. ' 15. Kuratsune, M. (1980). Yusho. "Haiogenated Biphenyls, Ter- phenyls, Naphthalenes, Dibenzodioxins and Related Compound. Elsevier North Holland, Amsterdam. 16. Kuratsune, M. et al. (1971). Yusho, a poisoning caused.by rice oil contaminated with PCBs. HSMHA Health Rep. 86, 12, 1083-1091. 17. Kuratsune, M. et al. (1972).* Epidemiologic study on Yusho, a poisoning caused by ingestion of rice oil contaminated with a commercial brand of polychlorinated biphenyls. - Environ. Health Perspect., 1,: 119-128. ` *T: 18. Hsi-Sung Chen, P. (1981). Polychlorinated' biphenyls. En vironmental occurrence, biological and toxicological ef fects. Clin. Med (Taipei) 7, 4--8 T 19. Chung-Fu Lan et al. (1981). An epidemiological study on PCB poisoning in Taichung area. Clin. Med. (Taipei) 7, -- - 96--100. ^^ r .j 20. Hsi-Sung Chen, P. -et~a T (198i)"V Toxic' compounds^in th cooking oil which caused PCB poisoning in Taiwan. I. Le vels of PCBs and PCDFs. Clin. Mid. (Taipei) 7, 71-76. 21. Hsi-Sung Chen, P. et al. (1981). Toxic compounds in the cooking oil which caused PCB poisoning in Taiwan. II. The presence of polychlorinated, quaterphenyls and poly chlorinated terphenyls. Clin. Med.'(Taipei) 7, 77-82. 22. Masuda, Y. et al. (1980). Polychlorinated dibenzofurans and related compounds in patients with Yusho. Proc. Work shop on Impact of Chlorinated Dioxins and Related Com pounds on the Environment, Oct. 22-24, 1980, Rome. 23. Kamps, L. R; et al. (197B). Polychlorinated quaterphenyls identified in rice oil. associated with Japanese "Yusho" poisoning. Bull. Environ. Cont. Toxicol. 20, 589-591. 24. Yamaryo, T. et al. (1979) . Formation of polychlorinated quaterphenyls by heating PCBs. Fukuoka Med. Acta 70(4), 88-92. r-w,-; s',-U V pi? V-. ` vi .A.. G . Reggiani 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. Masuda, Y. and Kuratsune, M. (1979)- Toxic compounds in the rice oil which caused Yusho. Fukuoka Med. Acta 70 (4), 229-237. Hayabuchi, H. et al. (1979). Consumption of toxic rice oil by "Yusho* patients and its relation to the clinical response and latent period. Food Cosmet. Toxicol. 17, 455-461. Po-Chak Cheng and King-Yin Liu (1981). Dermatopathological findings of PCB poisoning patients. Clin. Med. (Taipei) 7, 41-44. Hsi-Sung Chen, P. (1981). Dermatological survey of 122 PCP poisoning patients in comparison with blood PCB le vels. Clin. Med. (Taipei) 7, 15-22. Asahi, M. et al. (1979). Dermatological symptoms of Yusho alterations in this decade. Fukuoka Med. Acta 70(4), 172-180. Masakazu Asahi et al. (1981). Dermatological findings and their analysis in the general examination of Yusho in 1976-1980. Fukuoka Med. Acta 72(4), 223-229. Wei-Min Li et al. (1981). PCB poisoning of 27 cases in three generations of a large family. 'Clin. Med. (Taipei) 7, 23-27. Yao-n Fu (1981). Ocular manifestations of PCB poisoning and its relationships between blood levels and ocular findings. Clin. Med. (Taipei) 7, 28-34. Wen-Jen Wang et al. (1981). Investigations on severe PCB poisoning In-patients. Clin. Med. (Taipei) 7, 62-65. Lie-Gan Chia et al. (1981). Neurological manifestations in PCB poisoning. Clin. Med. (Taipei) 7, 45-61. Yoshihara, S. et al. (1979). Preliminary studies on the experimental PCB poisoning in rhesus monkeys. Fukuoka Med. Acta 70, 4, 135-171. Okumura, M. et al. (1979). Laboratory examination of the patients with PCB poisoning. Fukuoka Med. Acta 70, 4, 199-207. Hixayama, C. (1979). Hepatocellular dysfunction in pa tients with PCB poisoning. Fukuoka Med. Acta 70, 4, 238-245. Okumura, M. et al. (1979). Ten year follow-up study of serum triglycerides levels in 24 patients with PCB poison ing. Fukuoka Med. Acta 70, 4, 208-210y Kit-Ching Wong and May-Yae Hwang (1981). Children born to PCB poisoning mothers. Clin. Med. (Taipei) 7, 83-87. King-Lee Law et al. (1981) . PCB poisoning in newborn twins. Clin. Med. (Taipei) 7, 88-91. Miller, R. W. (1971). Cola-colored babies. Chlorobiphenyl poisoning in Japan. Teratology 4, 211-212. 42 43 44 4f 4. *vV. G. Rcggiani >ounds in \cta 70 1c rice clinical oJ. 17, athological (Taipei} of 122 PC3 le- s of Yusho 70(4) , ndings and sho in cases in (Taipei) poisoning ocular severe 7, 62-65. istations T is on the* Fukuoka -on of the i 70, 4, t in pa70, 4, study of PCS poison- ren born to 83-87. newborn orobiphenyl 2 Health EfieetJ of Halogenaied Dioxins and Related Compounds 67 42. Hsi-Sung Chen, P. et al. (1981). Levels and gas chromato graphic patterns of PCB in the blood of intoxicated pa tients after ingestion of toxic cooking oil. Clin. Med. (Taipei) 7, 35-40. 43. Chen, P. H. et al. (1980). Levels and gas chromatographic patterns of polychlorinated biphenyls in the blood of pa tients after.PCB poisoning in Taiwan. Bull. Environ. Cont. Toxicol. 25, 325-329. 44. Shigematsu, N. et al. (1979). Tissue distribution and biological effects of PCBs components, especially to the respiratory tract. Fukuoka Med. Acta 70, 4, 246-251. 45. Chiou-Jye Chen and Rey-Long Shen (1981). Blood PCB level and serum triglyceride in PCB poisoning. Clin. Med. (Taipei) 7, 66-70. 46. Kikuchi, M. et al. (1979). Autopsy report of two Yusho patients who died nine years after onset. Fukuoka Med. Acta 70, 4, 215-222. 47. Masakazu, A. et al. (1981). Dermatological findings and their analysis in the general examination of Yusho in 1976-1980. Fukuoka Med. Acta 72, 4, 223-229. 48. Rappe, C. et al. (1979). Identification of polychlorinated 1__ dibenzofurans retained in Yusho patients. Chemosphere 4, 259-266.' ' ` ' ' -- '* *- '49- Moore, J. A. et al. (1979) . Comparative toxicity of three halogenated dibenzofurans in guinea pig, monkey and mice. ; Ann. N. Y. Acad. Sci. 320, 151-163.