Document GKZn63nMVboey8XMz52mRdm0n

LONG TERM HEALTH EFFECTS OF EXPOSURE TO 2 A 5 - T AND/OR ITS CONTAMINANTS i 237738 Summary'of Findings' }A clinical and epidemiological'-study of 436 subjects from a plant making. 2*4,S-T fr|n 1948-69 was carried out. The subjects in the data analysis consisted * ' o f 48 white maljas: 204 subjects were defined as exposed, 163 not exposed, and there were 51 subjects whose exposure was questionable. The remaining 18 who were not included in the epidemiological' comparisons were 5 black males, 10 white females, 1 male American Indian, 1 male of Hispanic origin, and 1 male not employed at Monsanto but exposed to material of the runaway reaction as a child. Since the number of this group was too small for comparisons they were not included in the statistical, analysis. The data on each subject included historical information about illness from childhood to 1979, results of a comprehensive physical examination, clinical laboratory tests, organ function tests and x-rays. The analysis of the data on 418 subjects indicates the following: 1. Chloracne is the dramatic clinical hallmark which clearly separates the expostfM'rom the not exposed or those of questionable exposure; 86% of the 2G4 -* . 'f-' exposed dev4ped chloracne and this condition persisted in 52% of the exposed. S* * v- There w e r e no cases ofN chloracne in the not exposed.pr those of Questionable *. * ,L. .r'-ei exposure. . *\ 2. Actinic elastosls, a degenerative condition affecting the elastic /* tissues in the dermis is known to be related to the degree of exposure to sunlight and is-ge anfclci'n coloration dependent. In this study the frequency of actinic ` '***' .. ' elastosis was 59% in the exposed as compared to 30% in the not exposed.'* When corrected for age there was still a significant difference between the.exposed and not exposed.; A biologically significant observation indicates that actinic elastosis in the exposed is found predominantly in subjects with persistent 237740 examination among both the exposed and the not exposed groups when the >50 year old were- ompar^d to the <50 year old. The pulmonary function examination revealed consistent differences in all parameters (FEV, FVC, FEVj/FVC, and MMEFR) between the exposed and not exposed groups only if they are present smokers. The confounding factors were that the exposed group was 10 ^ears older than the not-* exposed and that 13 of the exposed were coal miners before they started to work for Monsanto. Among current smokers with no coal mining experience however, there were significant differences in FEV^, FVC and fttEFR between the 2,4,5-T process exposed and the not exposed. 7. When clinical laboratory data other than lipids were reviewed no differences were found between the exposed vs. the not exposed, the chloracne vs. no chloracne group. 8. Lipid findings were considered separately and in relation to exposure, age, and the presence of chloracne. When exposure alone was considered there were no significant differences in the frequency of abnormal levels in all parameters tested. Frequency of abnormal HDL levels however were-somewhat greater in those f` % with persistent chloracne^ than those who never had or formerly had chloracne and there was a greater frequency of abnormal LDL levels among those who only had a history of chloracne. Since the total number of abnormal values is small, the biological significance of these statistical differences is uncertain. 9. In the rreurobehavioral aspects of the study no differences in neurological examination findings between the exposed and not exposed groups were found that were not age associated or with other known etiologies such as extensive burns, W.W.II injuries, cervical nerve damage, cervical spurs, etc. In full recognition of the inherent problems associated with carrying out and Interpreting nerve conduction velocity tests, a critical analysis of the qualified subjects was made. No significant differences in motor and sensory -3- 237743 unctions were observed between the exposed and not exposed groups, between those who had cfcloracne-and' those who did not, after adjusting for ag. V In tie interview, questions were asked about nervousness, decrease in libido \i and 'impotence. Nervousness was reported more frequently among the older exposed group than the younger exposed group. Recognizing the cultural considerations which influence replies to questions about sexGal. matters, decreased sexual desire was reported more frequently among the exposed than the not exposed. However, it should be noted that there was a 10'year difference in mean age between the exposed and not exposed. When the two groups were compared on the basis of age, the differences in sexual desire and impotence were age related pot exposure related. 10. Reproduction considerations Again recognizing the inherent defects in the method of eliciting clinical information, no significant differences in the frequency of birth defects, stillbirths and miscarriages were found when, the exposed group was compared with the not exposed group. - 11. An unusual problem encountered only among the exposed were three cases fr. of Peyronie's Disease. This problem is characterized :;by induration and subsequently fibrous scarring of the corpus cavernosum of the penis producing distortion on erection. This problem is not known to have any relationship to toxic exposure. -4- LONG TERM HEALTH EFFECTS OF EXPOSURE TO 2,4;5-T ir ANO/OR ITS '.CONTAMINANTS 1 Tf Historical Review In 1949, at a plant of the Monsanto Chemical Company, Nitro, West Virginia, a runaway reaction occurred involving one of three autoclaves in Building 41 in which trichlorophenoO was being prepared. This process accident occurred on March 8, 1949, when under abnormal temperature and pressure conditions the safety valve and pipe connection of this autoclave blew out and the contents scattered throughout the interior of the building as well as the surrounding terrain and, structures.^ The employees who were asked to clean-up the building and repair the damaged autoclave were the first to develop symptoms. They were chemical operators, pipe fitters and other maintenance personnel. The acute symptoms in cluded irritation p,f the. respiratory, tract and eyes,,headache, .dizziness, nausea and a severe irritant reaction of the exposed skin. After the initial acute symptoms subsided, within 10 days to 3 weeks, an acneform eruption and other health effects were noted. The clinical findings, at that time, were acneform lesions, severe muscle pains affecting the upper and lower extremities, thorax and shoulders on exertion, easy fatigue, nervousness and tability, the com plaint of decreased libido, dyspnea, vertigo, and intolerance to cold. Of the four workers who were hospitalized in Cincinnati in 1949 for study, all had e n larged and tender livers, one manifested sensory loss in a foot. Clinical lab oratory findings included severely delayed prothrombin time and a slight increase in total serum lipids. Histological examination of a nerve biopsy in the worker with pedal sensory loss demonstrated myelin degeneration. Follow-up examinations were conducted in April of 1950* and subsequently, in 1953.2 In April of 1950 the original four and five additional subjects were seen. In 1953 -5- 237745 a tpt?l of ^ subjects were examined, seven of those having been seen in 1949 and 1950; \ Aftigr four years had elapsed, those symptoms and findings observed Initially which were referrable to the nervous system and liver were no longer found. Those with moderate to severe chloracne had improved considerably; there*, was some persistence of the acne in those so affected. In 1977-78 a mortality study^ of the group known to have been exposed to the material of the runaway reaction was conducted. This population consisted of 121 males. The follow-up was 1003S complete for death certificate information. Standard mortality ratios for all causes of death were determined. In 1957, 2,3,7,8 tetrachlorb-dibenzo-p-dioxin (TCDD) was identified as a toxic and acnegenic contaminant of trichlorophenol synthesis.4 ^ Since that time a considerable amount of scientific data has accumulated about the toxi cologic and clinical effects, associatedwith..e&pgsure to Ahe...trlchlorpphenol _ process, to 2,4,5-T synthesis and to the:contaminant TCDD in man, experimental -------- ' ' ................. . T-- ....................- animals, and in vitro research^rodels. .^|he sub-acute effects which have been repeatedly observed in human^iidclude chloracne, the most consistent clinical ^ . -- . - . r , * marker, liver function impalmiot, peripheral neuropathy, nervousness and irri tability, sensation impairment, personality changes, porphyria cutanea tarda as '- V well as hypertrichosis1 and hyperpigmentation.' Elevation of serum cholesterol, triglycerides, GGPT and alkaline phosphatase have been reported. Toxicologic examination of TCDD has revealed hepatotoxic- effects and teratogenic effects in rodents, suppression of cell-mediated immunity in mice and guinea pigs and the induction of hepatic carcinoma in rats. -6- 237746 Some" of the clinical reports appear to have' been concerned with the asso- ciatioty'of`cj^jdioVascular disease and an increased risk for cancer in populations expofed? to TC|)D. '1 OBJECTIVES OF STUDY i* `I* The objectives of this clinical study were to determine and identify the possible long-term -Health effects of chemicals associated with the making of 2,4,5-T, including 2,3,7,8 tetrachloro-dibenzo-p-dioxin (TCDD) and to determine the increased risk for those adverse health effects which have been observed in the sub-acute phase or sub-chronic state in man as well as the effects which have been noted to occur in experimental animals including rodents, rabbits and sub human primates. The objectives of the study, as reflected in its design, were to determine increased risks for cutaneous, pulmonary, cardiovascular, gastroin testinal, hepatic, .renal ,.neurobehav.ioral problems, reproductive problems., birth defects, endocrinologic effects, effects on lipid metabolism, susceptibility to infection and the possibility of increased risk for cancer. A most appropriate population for this study consisted of the employees and former employees of Monsanto Industrial Chemicals, Inc., Nitro, West Virginia who were involved in the manufacture of 2,4,5-T from 1948 to 1969. STUDY DESIGN The study design originally proposed the following three cohorts on the basis of company medical records. (1) Current employees, former employees and retirees who were exposed to the 2,4,5-T process during the period 1948 to 1969 and who were kr.own to have chloracne and/or other clinical manifestations attri buted to the process. 237747 -7- (2) >Y" ; /&- Current employees, former* employees or retirees who were exposed t o '' -'-C;. ; V1 ' * .. 2^,5-T. process during the period 1943-1969, but who were not known toldevefop chloracne or other clinical manifestations attributable 1 'ft to the process. (3) Current employees, former employees and retirees who were not exposed to 2,4,5-T process and have no record of chloracne. In the planning stage, it was anticipated that approximately 150 subjects would be available for the study in each of the above cohorts. It was estimated a total of 400-500 employees and former employees would participate in the study. Both management personnel and members of the Medical Department of the Monsanto/ Nitro plant did an excellent job of recruitment of subjects for the study. However, 1t was recognized during review of questionnaires, including work history interviews as Wei 1^ as 'pt^sician interviews and physical examination, that the participating ,subjects-did-not fulfill-,some of the-criteria for-these original cohort requirements. Selection bias was a factor in voluntary participation. No subject was excluded from the examination. The number in the exposed cohort e x ceeded the number in the'non-exposed cohort. Subjects, who had never claimed or *. been observed to have had chloracne, were found to have history and/or residual chloracne upon examination. There was disagreement with regard to definition of exposure, i.e., exposure and intermittent exposure. On the basis of information gleaned from questionnaires and company work records t'*ee cohorts were consti tuted: 1) those who were clearly exposed to the 2,4,5-T process or its components; 2) those who were clearly not exposed and 3) those who were questionably exposed. 23774S -8- METHODS AND MATERIALS . . . 'Ttfe design of the study and on-site survey were directed by Or. Raymond R. Suski/id. The| interviews and clinical examination of 436 subjects were conducted during the week, June 11-18, 1979, in the Putnam County Health Center in Winfield, West Virginia by members of the Departments of Environmental Health and Dermatology of the University of Cincinnati Medical Center. Personnel from the above departments irfcluded physicians, nurses, graduate students in the De- partment of Environmental Health, technicians and interviewers. An average of 60 persons were examined per day during the study, with fewer persons seen on Sunday. The questionnaire and study program is found in Appendix I. The complete examination included: an interview, clinical laboratory examination of blood and urine, pulmonary function tests, ECG, chest x-ray, nerve conduction velocity measurements, and a complete physical examination, including, skin examination b y dermatologists. During- the dermatological examination*biopsies, skin.scrapings, and photographs were taken as deemed appropriate by the physician and when agreed to by the participating subject. The nerve conduction velocity measurements were obtained for most subjects from two nerves: the m x peroneal nerve fo r motor function and the sural nerve fo r sensory function. In a relatively small number of subjects (22) nerve conduction measurements were made on the ulnar nerve. Clinical laboratory analyses included: blood calcium, phosphorous, BUN, creatinine, BUN/creatinine ratio, uric acid, glucose (CS), total protein, albumin, globulin, albumin/globulin ratio, total bilirubin, direct bilirubin, Transaminase SGO, Transaminase SGP, alkaline phosphatase, LDH, cholesterol, Iron, total lipids,' sodium, potassium, chloride, G-glutamyl transpeptidase, triglycerides, CBC and differential, thyroxine R.I.A., thyroxine binding glo- -9n *1 bulins. Urinalysis included routine examination as well as coproporphyrins and p ,t uroporphyri^ * Blood serum lipid fraction examination completed by the Lipid Research Division of the University of Cincinnati Medical Center included: serum ' \ f c h o l e s t e r o l s e r u m triglyceride, serum HDL, and estimated serum LDL. Chest x-rays were read by members of the staff of the Department of Radiology, University of Cincinnati Medical Center, under the supervision of Dr. Jerome Wiot. E C G 's were interpreted by Dr. Te-Chuan Chou of the Cardiology Division, Department of Internal Medicine, University of Cincinnati Medical Center. Skin biopsies were prepared and read by Dr. Daniel Richfield, Dermatopathologist. Skin scrapings were cultured at the University of Cincinnati. Dr. A. Blair Smith initiated the examination of the data. Dr. Vicki Hertzberg, of the Biostatistics Division, performed all of the statistical analysis described in this report. Data from the questionnaires were coded in the conventional m anner by assigningtiumerical values to the responses given. Normal and abnormal findings noted by,,the examining physicians and salient details of each s u b j e c t 's medical history were coded using the SNOMED system. Chest x-rays were coded by obser- f' ' vations and conclusions drawn by members of t h e ^ R a d ^ a T ^ y ^ p a r t m e n t . , ECG ^ findings were entered as interpreted by members of the Divisidlt'Of Cardiology, Department of Internal Medicine. ECG findings were reviewed with the s u b j e c t 's history, clinical findings, and chest x-rays in order to determine significant ECG findings. Laboratory data were entered as received. All data were keypunched and verified by the University of Cincinnati Computer Center. The keypunched data were then compared to the original files of the study subjects for accuracy. A computer file containing information from the questionnaires, * Values for urinary coproporphyrin and uroporphyrin were determined from a single void sample rather than the required 10 ml. aliquot of 24 hour volume; hence levels outside of the normal range cannot be regarded as significant. -10- 237750 physical examinations, laboratory findirigs-Was created. All extremely large or small values.for each variable were checked against the original values in the study subje'^'s files. Following this second edit for accuracy of entered data, analysis of |ata began, using statistical programs-available through.the Statis tical. 'nalysis System (SAS) computer package. The statistical procedures used are briefly described in Appendix II. As the analysis progressed, periodic editing continued so as to insure the greatest possible accuracy. THE SUBJECTS OF THE STUDY The total number of subjects in the study is 436. There- are 419 white males, 5 black males, 1 male of Hispanic descent, 1 male of American Indian*descent, and 10 white females. One subject among the 419 white males was a 36-year old who was exposed to runaway reaction residue in 1949.at the age of 6 v - S i n c e he-was not an employee, he is not included in the data analysis. Since'the number of women and non-white males is too small to be meaningful, they are not included in the remainder of-the analysis'. " A-brief description of these subjects, however, is contained in Appendix III. Information about sex, race, educ'^jbn and range is found in Table 1. ^ The cohorts designated by exposure are defined as follows-: Group I - no exposure: 163 subjects Group II - exposure, without qualification: 204 subjects These subjects were involved in process and production and include pipefitters, electricians and all other employees working in or around the 2,4,5-T process. Group III - questionable exposure: 51 subjects This group includes laboratory and supervisory personnel not in volved in production, but who occasionally handled 2,4,5-T or were around -11- 237751 the product 1cm area for short periods of time; laboratory personnel who worked n labortrie5 where 2 t4,S'T was analyzed,"but*- actually b a n d t e ^ h e material, carpenters who worked in the buildings of 2,4,5-T production occasionally, supervisory personnel whose main work efforts * were in offices, but occasionally toured th buildings where 2,4,5-T was produced, and warehouse employees who described handling all types of finished products, including 2,4,5-T. For a number of the analyses, the clearly exposed group was compared tb the clearly not exposed group. Group III constituted a suspected exposure category. We report comparisons of the latter group with the clearly exposed and not exposed where pertinent. \ I -12- 237752 RESULTS biographic Information biographic data of the 418 white males were examined to determine potential 1 / confounding variables. Data on all subjects are found in Tables 14. The three groups were found to be'significantly different with respect to employment status (p <0.0001) (Table 2) and the highest education level achieved (p< 0.0001) ( T a b l e d ) . *34.8% of the exposed were retirees and therefore in an older age group, whereas only 13.5 of the not exposed were retirees; 9.8% of the exposed were terminated, among them, employees who left the company in the era of 1949-1955 because of the health problem attributed to 2,4,5-T,.whereas only 0.6% of the not exposed group were in the category of terminated. .Tie mean age of the not exposed and* questionably exposed groups was ten years younger than the exposed group (Table 4). B. Alcoiffil and Smoking Use $ History of alcohol consumption was found not to be signif icantly different in. the three group (TabljejS.). When smoking status (current, former, nevem) wasjused as the criteria t h e r S ^ a s no difference between the exposed and 'i* " not exposed group (Table 6). However, when the. two groups were compared for pack years smoked, the exposed group was significantly higher than the not exposed group (Tabled). When present smokers were compared there was an almost 20 pack year difference between the_ exposed and not exposed. There was a difference of 9 pack years between the exposed and not exposed group in the former smokers. This probably reflects the 10 year age difference between the exposed and not exposed group and the heavier consumption of cigarettes in the exposed group. -13- 237753 C. Family History of Illness / Family history of illnesses were compared; the result? are given in Table 7. All of the comparisons are made using^Fisher's exact test f o r '2 x 2 :? tables. Family history of hayfever was foynd to be reported more often in the not exposed group (0.01'<p <0.05). A family history.of'heart attack or angina after age 60 was reported significantly more often in the exposed group ( p c O . Q l ) . D. History of^1Medical Problems The comparison o f 'the exposed and not exposed cohorts for illness elicited by medical history is given in Tables 8-9. Chloracne is reported more frequently in the exposed group: 86.27% in the exposed vs. 0.0% in the not exposed and questionable exposure group. In the not exposed.group, acne vulgaris occurred more frequently than in the exposed group. A history of upper gastro intestinal ulcer occurred almost 4 times more frequently in the''exposed'group .than j n .the^npt. exposed, group ( p < 0.005). When the medical history of the e x posed and not exposed are compared for two (2) age groups, less than 50 vs. 50 years and older, skin cancer and cerebrovascular accidents are only related to age not exposure; exposure, not age, appears to be a factor in the history of upper gastrointestinal ulcer. Hypertension and coronary artery disease occur more frequently in the older age group irrespective of exposure. E. History of Medical Problems in Populations With and Without Chloracne Since chloracne appears to be the .prime clinical indicator of. exposure,- absorption and biological response, the study subjects were further classified into three sub-groups within the exposed populations; i.e., those who never have developed chloracne, those who had a history of chloracne but which was not observed in the 1979 examination and those whose chloracne persisted and -14- 237754 was noted on clinical examination (Tables 10-13). Among the 204 clear.ly exposed, * 1* 176 devfclope^chloracne at some period. Of these, 107. were found to have chlor- acne o n examination. Among the not exposed and questionably exposed, there were none who presented a history of chloracne nor were they found to have evidence of chloracne on examination. In both age groups, history of upper G.I. ulcer is more frequent in those who had chloracne than those who did not have chloracne regardless of exposure (Tables 10 and 11). As in the exposed versus not exposed t group, the history of hypertension, cerebrovascular accident (CVA) and coronary artery disease are clearly age-related, not chloracne related. When the clinical histories of the exposed population only were examined in relation to the history of and/or presence of chloracne, the same observations are made about angina, hypertension and coronary artery disease in that they are age-related not chloracne related. In the exposed group there is no association between the history of-upper G; h-tilc-er--and-chloracne (Table 12)-.' . - - - .Although there...appears. to be a statistically significant relationship between exposure and history of upper G.I. djteer, a closer examination of the temporal sequence of exposure.'"and onset ofmftfer r e v ^ s six subjects in the exposed group and on subject'Tin-the questionable group whose symptoms of ulcer, occurred prior to exposure. XT' ^ In addition, there are four;- subjects famong the *1*1 exposed group whose ulcers were related to- cortisone therapy, ASA therapy r alcoholism. If these 11 subjects are omitted from the analysis there are then 32 of 194 (16.49X) of the exposed group reporting ulcer vs. 9 of 163 (5.523!) among the not exposed subjects. This difference is statistically significant upon age adjustment (p=0.01). However, if the seven subjects reporting history of ulcer before exposure are considered to be not exposed subjects because of the temporal sequence, there are then 32 of 194 (16.493!) of the exposed group reporting ulcer -15- 237755 'V'S. 16 of-170 (9.41%). This'difference is not statistically significant upon age adjustment.,- . | I; Other potentially confounding variables were examined wftTr the following results: (1) There was no relationship found between exposure category and history of ulcer among never smoking subjects. (2) Among formerly smoking subjects, 24.10% (20 of 83) of the exposed group report history of ulcer compared to 6.85% (5 of 73) of the not exposed group (p < 0-005). (3) There was no relationship found between exposure category and history of ulcer among presently smoking subjects. (4) There was no relationship found between exposure category and history of ulcer among the*three categories of alcohol drinking (never, former, and present). -16- 237756 The following is a breakdown of data designating exposure vs. ulcer type and treatment: , -- 'i* *t * History of * + Not' -Exposed {n=i63r #% Exposed ((1=204) #X Questionable * Exposure #* % Gastrointestinal Bleeding 0 50 Upper G.I. Ulcer ^ 9 5.52 42 20.59 6 11.76 Site as Reported Peptic Stomach Duodenal 3 91 4 21 3 2 12 2 Treatment as Reported Did not Report Treatment No Treatment Treated with Diet Only Medical Treatment Ulcer Symptoms Relieved with Hiatal Hernia Repair Gastrectomy Partial Gastrectomy Surgical Treatment of Ulcer, NOS 2 - 7 13 3 4 10 ? -3 1 6 3 2 1. .. Problems Now 3 81 These data have the additional problem of being prevalence figures only as opposed to incidence rates.^ Also, report of upper gastrointestinal ulcer and * designation of ulcer by site is crude at best since neither examination by gastrointestinal x-ray, nor endoscopy diagnosis was available. Hence, while these data are suggestive, conclusions cannot be drawn due to the equivocal findings and the crude nature of the data. -17- 237757  F., Physical Examination Findings ,, .. ' ^ -- - - - - v ~ ~- : -The physics* examination revealed a wide'range of findings. These are -W ' ' ' cited:in a comprehensive list in Appendix IV. The significant clinical findings were those associated with the skin and appendages, connective tissues, and sexual complaints- . The discernable abnormalities referrable to the cardio vascular, pulmonary, hepatic, hematopoietic systems, and nerve function were largely determined b y objective tests in this study. These will be discussed under organ function tests, rad.iography and ciinical laboratory measurements. The significant clinical findings as related to exposure status are found in Table 14 and 15. As was indicated previously 52.45% of those who were clearly exposed still had chloracne. There is a significant difference in the frequency of acne vulgaris among the not exposed as compared to the exposed, 11.66% versus 1.96%. Among the exposed 120 or 58.82% were found to have actinic -elastosis_in contrast to 49 (30.06%) among the not exposed and 39.22% among the questionably exposed. There is a very significant difference in the occurrence ^of actinic elastosis between the exposed and not exposed in both age groups even ^ . t h o u g h age is a factor in this clinical finding. There" is a rather unique find- % ing of 3 cases of Peyrortie's Disease among the 204 exposed. '73;.'- `^chloracne at some time and they were all over 50 years old. All of them had .. .*< Rosacea and rhinophyma occurred in 4 exposed persons over the age of 50 and 2 questionably exposed. The complaints of loss of libido occurred more f r e quently in the exposed than'the not exposed. When age was considered, the largest number of complaints of loss of libido or impotence occurred in the older age group but there were greater percentages of complaints about these problems in the exposed group as compared to the not exposed. 237758 -18* Tables 16 and 17 identify the clinical findings in the exposed popu lation in relation to chloracrid status.- In these groups, actinic elastosis is '1 ' found predominantly in the subjects with persistent chloracne (74.8) but it also occurs significantly in those with a history of chloracne only (46.38%). 11 cases of malar hirsutism were found among tfjp exposed in the persistent chloracne cases (10.28%). 9 of these 11 cases occur in the group over 50. Table 18 lisjts the sites and severity of chloracne observed on physical examination. The distribution and severity of actinic elastosis is presented in Table 19. Since there appears to be some association of actinic elastosis with chloracne further analysis of this relationship with respect to severity of chloracne and severity of actinic elastosis was carried out. These analyses are presented in Table 20 and Table 21. Of the 67 mild cases of residual chloracne, 44 had actinic elastosis. Of the 36 with moderate chloracne 33 had actinic elastosis. 33%-of the .actinic elastosis cases found with chloracne were severe. It would appear that the percentage of persons with severe actinic elastosis^is related to presence of chloeacne (Tab'T e I n otjhjr words, aclinic elastosis appears to be more severe when found wittL.chloraciii * . ' Blood pressure findings at-the time of examination ihvthe-exposed,.. questionably exposed and not exposed groups are presented in Table? 22. When the criterion for abnormal systolic and diastolic use different values for the groups < 5 9 (140/90) and ^60 (160/95), there appears to be no significant differences between the exposed and not exposed group. These criteria are stricter than that proposed by the National Heart, Lung and Blood Institute. 237759 -19-  G. Clinical Laboratory and Other Test Findings '. ^ Lipids ^ ? Because of the suspected relationship of serum lipids to exposure to 2,4,5-T and its contaminants, serum lipids are analyzed in several ways. Serum cholesterol, triglyceride, LDL, and HDL were classified by age-specific per centiles as determined by the National Lipid Research Project data^ (See Appendix V). Serum cholesterol, triglyceride, and LDL were classified as "abnormal" if they fell above the 90th percentile, and were classified as "normal" otherwise. Serum HDL was classified as "abnormal" if it fell below the 10th percentile, and was classified as "normal" otherwise. The frequency of abnormal lipid findings are compared for exposed,, not exposed workers and questionably exposed (Table 23), and no significant differences are noted in cholesterol, triglycerides, HDL and LDL,levels. These relationships are further investigated within each level of smoking.status (present, former.and never) (Table. 24).- Smoking-.appears -ta~have no effect on the differences.between these groups. Logfstic regression was used to detect associations of abnormal serum lipid findings with exposure. After:. adjusting for pack years smoked, no significant differences betSpert exposed- and not exposed could be found. * Multiple linear regression was also used to detect possible "Shifts" in the mean serum lipids for the two exposure groups after correcting for the effects of age, tobacco, and alcohol. In all cases (cholesterol, triglycerides, HDL, LDL) there are no significant differences between the two exposure groups after adjusting for the effects of age, tobacco and alcohol. When the lipid levels of the exposed group only are examined in relation to the occurr&nce of chloracne, the HDL levels in those with persistent chloracne are found to be somewhat lower than of the group with a history only of i .chloracne or those who never had chloracne (Table 25). Those who had a history J * t only of chloracne have somewhat higfter LDLr levels than those who had chloracne i' . * currently or^never. Among former smokers in the exposed group there is a greater frequency of abnormal triglyceride and HDL levels among those with residual chloracne than those with no history or only history of chloracne (Table 26). Among those who smoke presently there is a higher frequency of abnormal or high LDL levels for those^who have a history only of chloracne. 2) Pulmonary Function Status * Both FEVj and FVC were considered abnormal if the observed value was less than 80% of the predicted value of Morris. A FEV^/FVC ratio of less than 0.70 (70%) was considered to be abnormal. The 75 was considered abnormal if the observed value was less than 70% of the predicted value of Morris. Consistent differences in abnormal pulmonary function values (FEV^, F.YC, FEVj/FVC and ^ 2 5 - 7 5 ? are ^ound between the exposed and not exposed .grdups (Table 27). Among present smokers there was a significant difference in' ; ^pulmonary function values between the exposed and not exposed group. There are *'';.^.-<Jifferences in pulmonary function values, among persons who have formerly Smoked or among persons who have never smoked when the exposed were compared to the not exposed (Table 28). . * 'i Logistic regression was used to detect associations of abnormal pulmonary function parameters with exposure. The model and coefficients are given in Appendix VI. The significance and odds ratios for abnormal pulmonary function for exposed vs. not exposed after adjusting for pack years smoked are -21- o^irrypA as follows: Pulmonary Function Parameter FVC FEVj/FVC FEF25-75 Significanci of Exposure 0.02 0.05 0.01 0.06 Odds Ratio 2.71 2.10 2.89 1.82 The association of exposure and abnormal pulmonary function among current smokers was investigated further for the confounding by prior coal mining exposure. None of the not exposed subjects who currently smoke had prior coal mining exposure. Of the exposed subjects who currently smoke, 13 individuals had prior coal mining exposure. Among the current smokers who had no coal mining exposure the exposed were found to have a higher frequency of abnormal FEV^, FVC .and MMEFR..than the. not exposed as,,follows: Pulmonary Function Parameter FEVi FVC FEVj/FVC FEF25-75 Non-Miners-Current Smokers Miners Current Smokers Exposed "(n-60)'* # Abnormal Not Exposed (n=44) # Abnormal Proba tuli ty Exposed {" 13) # "Abnormal 12 20.00 14 23.33 10 16.67 18 30.00 2 4.55 1 2.27 3 6.82 5 11.36 0.039 0.003 N.S. 0.039 7 53.85 4 30.77 8 61.54 8 61.54 237762 *L Tables 29 and 30 lis t the o c c u r r e d of ECG findings in the e x posed and not exposed a s w e l l * a s the differences in occurrence when the exposed and not exposed are ^separated into < 5 0 and ,50 year groups. Number of persons with P V C 's and PA'C's appear to be only age-related. Tables 31 and<32 list the x- ray findings in the exposed as compared to th not exposed as well as the comparison between the <50 and >50 year old groups. Significant x-ray findings findings appear to be age-related but not exposure related. 4) Clinical Laboratory Findings In Table 33 the relative frequency of out of normal range values is compared in the exposed vs. not exposed group. ; In Table 34 the relative f r e quency of out of normal range values of clinical laboratory findings for the exposed group in relation to chloracne status is presented. Analysis covariance was,, used, to compare.. the_laboratory. f i n i n g s in the exposed group to the not exposed groo^.af jus ting for the effects of age, smoking status and alcohol status. The'iame procedure wa s us ed to compare the laboratory f i n i n g s for t h r e e - c h l o r a c n e ^ ^ p s a f ^ ^ p ^ a d j u ^ i Fg for the effects of age, smoking and alca atus- -ip Fisher's exact test was used to determine the existence; of differences between the exposed and not exposed with respect to the relative frequency of out-of-reference-range laboratory findings. The X test for independence in contingency'or two way tables was used to determine differences in the three (3) chloracne groups with respect to the relative frequency of out- of-reference-range laboratory findings (Appendix VII). 2377F3 In several other incidents involving the synthesis of trichloro- phenol and 2,4',5-T, porphyria cutanea tarda was observed with uroporphyrinuria. ' Th e x p o s e d ^ n o j exposed and questionably exposed in this study were compared for percentage of abnormal levels of coproporphyrin and uroporphyrins. No signi ficant differences .were found (Table 35). 5) Nerve Conduction Velocity Data Prior; to the beginning of the data analysis, all subjects reporting diabetes or reporting a high weekly alcohol intake were eliminated from the analysis (high weekly alcohol intake was-defined as greater than or equal to 35 oz. alcohol per week). After these deletions, there were 319 with complete sural examinations and 336 with complete peroneal examinations remaining. The data from the nerve conduction velocity tudy were adjusted according to the. deJesus method (Appendix VIII). Tables 36-38 lists all nerve conduction parameters considered. The following methods of analysis were then used: a) Analysis of covariance was used to compare the nerve conduction parameters after adjusting f o r age differences in the groups. No -in significant differences were, found .betwen the.:expfesefi'and riat.ii&ftsed. Also ' ** *' . 3 ^ within the exposed only* ho- relation to ch 1or - adjusting for age. tu w a t t found after Jr - b) Using the normal ranges provided by NIOSH (see Appendix IX) the exposed group has a significantly higher proportion with abnormally low peroneal nerve conduction velocities than the not exposed group (74% vs. 49%, p <0.0001). No differences were found with respect to abnormal sural nerve conduction velocity or peroneal distal latency (not enough ulnars were tested to say anything about them). However, 8oyd points out that these ranges are not age- adjusted. Moreover, temperature has a great effect on velocity and latency, and -24- 2377G4 temperatures were not reported with these ranges. .(Ranges and appropriate references are etached}. Among young men (age-<50) no difference between exposed and'not exposed with respect to any of the parameters was observed. Among the older men-(age>50) a significantly higher proportion of the exposed group have abnormally low peroneal nerve conduction velocities as compared to the not exposed group (793S vs. 573, 0.01 < p <0.05). c) The z-score method (as used by Selikoff ) was als.o used to detect differences in the nerve conduction parameters between exposed and not exposed. No significant differences were found using this method. H. Reproduction and Birth Defect Data In presenting this information it should be understood that all of the data depended on personal recall of such information as number of pregnancies, miscarriages, live births, stillbirths, infant mortality, etc. Confounding factors which effect accuracy are obvious - e.g. interviewed male subject only, agf"of subject, and change in cultural-attitudes about;reproductive matters. It i ^ p g j s o confounded'by the fact that the pregnancies.iiniscarriages' stillbirths riot designated in a time period related to exposure e.g., pre, during, post exposure. Fisher's exact test was used to compare the reported occurrence of r e productive anomalies in the exposed group to the not exposed group. The data on reproductive and birth defect findings are found in Tables 39 and 40. No sig nificant differences were found in miscarriages, in rate of birth defects or stillbirths. 2377C5. The TCP Runaway Reaction 'Group- i> . * :'* i r Among th^ 204 persons who were known to be exposed to the.2,4,5-T synthesis and preparation, there were 55 who were involved in a trichlorophenol process runaway reaction, which occurred on March 8, 1949, described in the historical introduction- There were a total of 122 persons known to have been involved in this accident and with its products. This group of 55* was matched as closely as possible in age with an unexposed group and a third group of exposed but not to the accident material. Since the range of the runaway reaction group was 49 years and older, the matched unexposed control group consisted of persons in the same age range, e . g . , ^ 4 9 years. There were 65 persons in the unexposed group. Because of the percentage of aging r e - ' tirees in the TCP run away group-there was still a four year age difference between the accident group and controls. There was no difference in age between the TCP "accident group and the other exposed, - In .the other_expcsecL group, the age r a n g e - w a s ^ 55. . There we_re 77 subjects^in. that group.: Data comparisons were made between the accident group and the tinexposed group^as'well as the accident group and the other exposed group. The following were the significant findings elicited: * 1) When smoking status was considered, the average number of pack years was significantly greater in the'TCP runaway group than the not exposed group' by 10 pack years. The difference between the TCP group and other exposed group was 9.5 pack years. 2) All of the TCP exposed had a history of chloracne; none of the not exposed had chloracne; 83% of the other exp.osed group had a history of chloracne. In the TCP group examined, 58.2% still had chloracne; in the other exposed 55.8% still had chloracne. -26- 2377E6 3) There were no significant differences in the current findings of chlor- acne and aclinic elastosis between the TCP and the other exposed group. 4) The percentage of persons with actinic elastosis was significantly larger in the TCP group than the not exposed group (75% vs. 49%). The difference in the occurrence of actinic elastosis between the TCP and other exposed group (75% vs. 68%) was not significant. 5) There was**a significantly larger number of systolic murmurs found in the TCP as compared to the other exposed group (12.7% vs. 1.3%, 0.01 p 0.05). Although there appears to be a statistical difference between the percentage of abnormal BUN'S found in the TCP group (9.43% or 5/55) as compared to the other exposed (1/73 or 1.37%) there was no difference between .the TCP group and not exposed group (9.43% vs. 6.35%). 6) The complaints of nervousness were greater in the TCP runaway group "than the not "exposed controls" {'21.8% vs. 7.7%). In the other exposed controls'the occurrence-of nervousness was 16.9%. - - - 7) Logistic regression was used to compare TCP runaway group to the not exposed and to the other exposed group, for abnormal serum lipid findings, while simultaneously adjusting for pack years smoked. No significant differences were found in either comparison. ; r B) In the pulmonary function tests, the number of persons with abnormal mm FEVj/FVC test findings in the TCP group was significantly greater than among the not exposed. This same difference was seen among current smokers in the two groups. There was also a slightly greater percentage of persons with abnormal MMEFR among the TCP group who are currently smoking than the not exposed who are currently smoking. ": :: 237767 9) ' *' Logistic regression was used to'"Conipare*TCP runaway-exposed to not - *. 4 4 exposed for abnormal pulmonary f u n c t i o n a l l e adjusting for^pack years smoked. No relationship was found between TCP runaway exposure and abnormal FEV^, FVC or MMEFR. However, a significant relationship was found between this exposure classification and abnormal FEV^/FVC, with a greater frequency seen in the TCP runaway group (p=0.01). The odds ratio is 4.02 for abnormal FEV^/FVC, given pack years smoked, for TCP runaway exposure vs. not exposed. Logistic regression was used to compare' TCP runaway exposed to other exposed for abnormal pulmonary function while adjusting for differences in pack years smoked. No significant differences were found. -28- 2377C8 Other Work Exposures*'flf the Study Population jV ' " ..Other;,,TtHan 2,4,5-i, its intermediates, and/or contaminants, all'.of the workers in the study were exposed to multiple chemicals and industrial processes under'varying working conditions, at Monsanto/Nitro and/or other industries, in a period roughly between the late 1930's and 1979 (a few subjects were born before 1900). Chemicals identified by the subjects as other Monsanto/Nitro exposures in clude: rubber processing chemicals, i.e., diphenylguanidine; a group of benzo- thiazolesulfenamide accelerators (Santocures); antioxidants, i.e., derivatives of m-cresol; antiozonants (Flectols and Santoflexes), most of the Santoflexes are p phenylenediamine derivatives; vulcanizing agents such as 4,4-dithiodi- morpholine; insecticides, i.e., ethyl and methyl parathion. Additional chem icals named were phthalic anhydride, o and p dichlorobenzene, o and p nitro- phenal,.. toluene.-and mercaptobenzothiazole,-derivatives- of-- TMTD and-t r i m e t h y l* " quinoline components. One of the products, Sopanox, an antioxidant,.was alleged^... by the study subjects to be associated with the occurrwcie of kidney stones. The *'* -* * -* reporting of the H/0 kidney stones is nfl^i$t4tS,sticatft;f?erentiamong the three^- groups. * v S*-/1 N One of the chemical hazards to which the study population was. exposed was p aminobiphenyl, a known bladder carcinogen. Ninety-five subjects, or 22.7225 of the 418, reported exposure to p aminobiphenyl during the course of their e m ployment. To the best of our knowledge these subjects are periodically examined in a monitoring program established by the Medical Department of the Company. Subjects reported the exposure and the monitoring program during the interview and to the examining physician. The data analyses relating to this exposure and -9Q- exposure tp, 2> A \-X is as- follows: ***'* , . * ""***-'* f paminobiphenyl Exposure Not Exposed Cn=163) #X 8 4-91 H/0 Bladder Tumor 1 0.-61 H/0 Bladder Cancer 1 0.61 * Figures are mutually exclusive 2,4,5-T Exposure Exposed (n=204) #% 71 34-80 7 3.43 2 0.98 Questionable Exposure ** C*t> 16 31.37 0 0.100 2 3.92 The degree to which each subject may have been exposed to other chemicals both at Monsanto/Nitro and other places of employment varied markedly- This fact must be considered in attempting to interpret biological significance to statis tical differences based on small occurrences. v j. '* -30- 237770 Discussion and Conclusions , 'The strengths art^y/eaknesses of this study must be considered critically in any appraisal of the medical significance of the data elicited from it. 'Data gleaned from interviews and clinicai examinations are useful but have *% some inherent weaknesses. Medical, work and personal historie^s depend on memory; findings of physical examinations carried out by different physicians are based on judgments which fnay differ somewhat with the examiner. Nevertheless these. constitute an important data base for making clinical judgments. Since quantitative TCDD levels for any period during the manufacture of 2,4,5-T were not available, it was only possible to relate effects found, to the *\ history of definite exposure, no exposure, and questionable exposure to any aspect of 2,4,5-T production. Since chloracne*'f s the hallmark of biological activity of TCDD it was possible in this study to~ refute al 1" systemic anti other CtltTrteous' effects' td the presence or absence .of. chloracne in the'expbseikgroup 'ai^we'tl-to exposure itself. From this study obe can c o n c l u d e ^ h a t fprjEbose excised to TCDD the-dramatic ._ _ long term effects are cutaneous-"and-lri^lve*'tl8f^ptlos"eStceous apparatus and the os '* ' .. - elastic tissues of the dermis.; O M o r a c n e may persist forces long as.3Cgyears after onset and is associated with 'elastic tissue changes irt the dermis and some persistence of malar hirsutism. There are suggestive but inconclusive data indicating that for the exposed there was an increased risk for upper gastrointestinal ulcer; t h a t `among those in whom chloracne persists there is (although numbers are small) an association with low HDL levels and further for those with only a history of chloracne there is an association with higher LDL levels. In those who currently smoke and have been exposed to the 2,4,5-T process -31- 237771 materials there appears to be an association of abnormal pulmonary function .'* findings when compared to those who smoke but who were never exposed. - t. There is no evidence from this study that there are identifiable long term effects on the cardiovascular system, on the liver, on the gastrointestinal tract, on peripheral nerve function. There is no indication, on the basis of historical Information (recognizing all of its weaknesses) that there was a greater risk for miscarriages, stillbirths, or birth defects among the families in which the male i parent was exposed. 'M . -32- 237772 References 1. -Ashe, tfilTiam and Suskind, Raymond R.: Report of the Kettering Laboratory, University of Cincinnati, 'Chloracne Cases of the Monsanto Chemical Company, Nitro, West Virginia, October 194JL;.Apri 1*1950. 2. Suskirfd R.R.: Report of the Kettering Laboratory University of Cincinnati, A C T imcal ana tnvironmentaI Survey, Monsanto Cnemical Company, Nitro, West Virginia, July 1953. 3. Zack, J.A. and Suskind, R.R.: The Mortality Experience of Workers Exposed to Tetrachlorodibenzodioxin in a Trichlorophenol Process Accident, Journal of Occupational Medicine, 22:11-14, 1980. 4. Kimmig, J. and Schulz, K.H.: Chlorinated Aromatic Cyclic Ether as a Cause of So-Called Chloracne, Naturwissenschaften, 44:337-338, 1957. 5. Kimmig, J. and Schulz, K.H.: Occupational Acne Caused by Chlorinated A r o matic Cyclic Ethers, Dermatologica, 115:540-546, 1957. 6. Systematized Nomenclature of Medicine, College of American Pathologists, Second Edition, 2 Volumes, April 1979. 7. Kurata, J.H., Elashoff, J.D., Grossman, M.I.: Inadequacy of the Literature of the Relationship Between Drugs, Ulcers, and Gastrointestinal Bleeding, ,Gastroenterology, 82:373-82, 1982.... ...... .................... - 8. Report of Jojnt National Committee on Detection, Evaluation and Treatment of Hfgh Blood-pressure,--National-Heart-,- Lung and Blood Institute,- 1 9 7 7 ... 9. The Lipid Research Clinics Population Studies B o o k : Volume I: The P r e valence Study, USOHHS 11980). . -33- 237773 I * TABLES M 237774 TABLE 1 ~ nrMfirtnaPHir h a t a ALL STUDY SUBJECTS Males Females Total f Total 426 10 436 White 419 10 Black 45 0 Other 2 0 HIGHEST EDUCATION LEVEL ACHIEVED 0 - 8**1 Grade 9 - 12th Grade" . > 1 2 ^ Grade.; Total, - % 25 266 ._ ; 12 % i *' *4: . `tI 41*6*-1V ' 237775 TABLE 2 s PRESENT EMPLOYMENT STATUS Not Exposed Number Percent Active 140 85.89 Retired.. - 22 13.50 ` Terminated 1 0.61 * Total 163 Questionable .... -Exposure -.... Number ^ 38 Percent 74.51 11 21.57 -......... -----. 2 3.92 -51 Exposed Number Percent 113 55.39 71 34.80 20 9.80 204 Total ' 291 104 23 418 p < O.OOOl it- '. 237776 35- TABLE 3 EDUCATION LEVEL 0 - 8th Grade Not Exposed Number 2 - P e r c e n t -....... 1.23 9 - ,12th Grade ` 92 .... 56.79 Questionable Exposure Number Percent * 4 . 7.84 27 52.94 >12th Grade Total 68 41.98 - 162 ..... .. . . .. 20 39.22 51 Exposed Number /. Percent Total .19 9.36 ' 25 147 72.41 266 37 18.23 125 203 416 p < O.OOOl -36 237777