Document 6YZdBMENQ3EgKEMm9bExJYm4

/*i:inihiw r/if/ //nr/Z/i /VrN/i7irr> l M l. //(. J t f . . / . V '. /.'W.* Effects of PCB Exposure on Biochemical and Hematological Findings in Capacitor Workers by Richard W. Lawton,* Marie R. Ross,t Joseph Feingold* and John F. Brown, Jr.* Certain former operations in capacitor manufacturing resulted in extensive direct contact of the workers with electrical grade polychlorinated biphenyls (PCBs). A study group of 194 such individuals, all exposed to Arodor 1016 and manyjireviously exposed to Arociors 1242 and/or 1254, was examined before (1976) and after (1979) discontinuance of PCB use in the operations (1977). At the two examinations, the ap proximate geometric mean serum levels (in ppb) and 5 to 95% ranges were for lower PCBs (LPCB), 363 (57-2270) and 68 (12-392); and for higher PCBs (HPCB), 30 (6-142) and 19 (4-108), respectively. The statistical associations among 42 measured clinical chemical and'hematological parameters, five different measures of PCB exposure, and seven confounding variables observed in the two examinations were determined by three regression procedures. Similar regressions were performed with DDE, which was present at background levels. The principal statistical findings were a depression in serum bilirubin and elevations in serum GGTP and lymphocyte levels at the time of the first examination, and only an elevation in monocytes at the second. Appraisal of the results suggested an induction of microsomal enzymes which appeared to be subsiding after the cessation of direct exposure to PCBs. The statistical association between serum levels of PCBs and lipids reported by others was confirmed, but shown to be explained by the partitioning behavior of PCB in the body, rather than to changes in liver function. No evidence for health impairment related to PCBs was found, despite the high serum levels of PCBs in the study population. \ Introduction Concern over the human health effects of the poly chlorinated biphenyls (PCBs) arises because of the en vironmental persistence of these substances, which were widely used in the U.S. for nearly 50 years (1929-1978). Several hundred million pounds still remain in exis tence, either in electrical equipment or in environmental compartments such as landfills, spillage sites or aquatic sediments (i), so that the potential for human exposure continues. The main growth of PCB usage occurred only after early toxicological studies and extensive industrial ex perience indicated these materials to be of relatively low toxicity (2-5). In the 1970s, however, PCBs were \ linked to an outbreak of severe chloracnegic disease in southwestern Japan in 1968. The outbreak was traced General Electric Research and Development Center, P 0. Box 8, ^ ^rhenectady, NY 12301. i 'General Electric Company, Capacitor Products Department, V .,, ndson Falls, NY 12839. Present address: 31 East Street, Fort Edward, NY 12828. to the ingestion of rice oil (Yusho) that had been con taminated with thermally decomposed PCB. The toxic agents responsible for "Yusho disease" were ultimately identified as 2,3,4,7,8-pentachIorodibenzofuran and other polychlorinated dibenzofurans (PCDFs) (5), which are formed from PCBs only at very high temperatures (7). Despite contrary evidence, concern has remained that ordinary, unpyrolyzed PCBs, even at residual environ mental concentrations, might also produce some human health effects. In order to identify such effects, a number of epide miological investigations have been performed on oc cupationally (8-19) or environmentally (20-22) exposed populations. All such studies have indicated PCB-related health effects in the human to be uncommon, and the only reasonably consistent findings to be statistical associations between the levels of serum PCBs and those of serum lipids, e .g ., triglycerides and cholesterol (13f15t21f22). These associations have been interpreted as suggestive of otherwise asymptomatic alterations in liver function. This interpretation did not recognize that in the body P L A IN T IF F 'S EXHIBIT 166 LAWTOS E T AL. PCBs are partitioned between adipose tissue (the major reservoir) and blood serum in proportion to the lipid content of the serum (,4). Therefore, statistical associations of serum PCBs with serum lipids and any covariant parameters could arise solely from such par titioning. In order to determine whether PCBs actually do have any effect upon serum lipid levels, it would be necessary to examine the dependence of the latter, not upon the levels of PCB in the serum, but instead upon their levels within the body lipids. For lipid-soluble ma terials, the concentration within the body's lipid phases has long been recognized as the determinant of the phar macological response (25). In addition to the consistently observed association between the levels of PCBs and lipids in the serum, there have been isolated reports of hypertension (22), hepatomegaly (14), chloracne (U,22), other dermal ef fects (12-14), oxidase induction (16), restrictive lung disease (IS), increased levels of serum y-glutamyl tran speptidase (14,15,22), serum glutamic oxalacetic trans aminase (13-15), serum glutamic pyruvic trans aminase (12,14), serum ornithin-carbamoyl transferase (14), or serum pseudocholine esterase (14), increased bromsulphthalein retention (12), or reduced plasma HDL cho lesterol (15) in the groups studied. Most such reports, however, either lack consistency with the findings of other investigators, or were observed only in groups that were small, lightly exposed to PCBs, and/or ex posed to PCBs in combination with other chemicals. An opportunity to study the effects of unpyrolyzed, electrical grade PCB exposure has been provided at two capacitor manufacturing plants located within a mile of each other along the Hudson River in upstate New York. During their period of PCB use (1946-1977), these plants are believed to have been the largest single U.S. users, consuming about 15% of all U.S. production, and to have had one of the largest PCB-exposed work forces. Be tween 1946 and 1975, approximately 9500 different em ployees worked there, and the employment during the 1976-1979 period of this study was between 1500 and 1800. Previous investigators of this plant population have found its mortality experience (reported as that for "Plant 2" by Brown and Jones) (19) to be normal for all major causes of death, and have indicated a general paucity of significant clinical effects in a volunteer sam ple drawn from the plant population as a whole (17). However, only 5 to 10% of the plant employees worked in areas or jobs providing direct, continuing exposure to the dielectric fluids. Thus, it appeared probable that this exposed subpopulation, still a sizeable group, might exhibit considerably greater blood PCB levels and phys iological effects than the remainder of the plant population. Accordingly, a study population was selected, using ^/criteria that were broad enough to include the entire directly exposed group, as well as some less directly exposed individuals, and detailed clinical examinations were performed approximately 16 months before and 29 months after the discontinuance of PCB use in the plant. No attempt was made to establish a comparable age- and sex-matched control cohort, but since the study group itself exhibited a distribution of PCB levels cov ering over two orders of magnitude, analyses within the group offered adequate prospects for identifying and quantifying physiological effects attributable to the PCBs. Since it was not possible, initially, to decide which of the many clinically measured varaiables might be af fected by PCBs or might confound the study, a three tiered analysis was used to identify statistically signif icant associations. In order, these analyses were: (a) a simple linear regression, intrinsically bias-free, for each measured variable or its log transform against each measure of PCB exposure; (b) a multiple linear regres sion for each of the measured variables against the se lected measures of PCB exposure and a series of other independent variables, e.g., sex, age,.smoking history, body mass index, etc., believed to have effects on one or more of the clinical variables and (c) a backward stepwise linear regression to remove non significant in dependent variables. For each regression, statistically significant associations (p < 0.05) with th measures of PCB were identified. In addition, the 95% confidence limits of the regression coefficient were used to estimate the significance of the-association over the observed range of exposure. Similar regressions were performed for the same measures o fp ,p '-DDE, which shares with PCBs partitioning behavior in lipids. Materials and Methods t Description of Environment In the*process used for manufacturing liquid dielectric capacitors, metal foil strips separated by paper layers were wound into tight coils and inserted into metal cans. A cover assembly, having electrical connectors and a port to admit the dielectric fluid, was then sealed to the can. For Ailing, small capacitors, such as those used in fluorescent light ballasts, were packed onto racks that were placed in large autoclaves, warmed, degassed un der vacuum, and then flooded with the dielectric fluid. The racks were then drained, removed, and transported to sealing stations where the port was manually sealed with solder. Large power capacitors were originally filled manually; subsequently, they were filled by.an elec tronically controlled manifold system which 'reduced spillage and dermal contact. Following sealing, both large and small capacitors were cleaned, degreased; tested, repaired if appropriate, labeled, and packed for ship ment. The dielectric fluids were blended with stabiliz ers, treated with fuller's earth to rem ove polar impurities, and tested for dielectric properties, thereby producing an "electrical grade" product. The jobs which provided frequent or continuous direct exposure to the dielectric fluids (i.e., direct dermal contact and/or close proximity to a source of vapor) were those associated HH'itKMiCM. ASH HKMATUUK'M'.' ' * /.v ( W i'M 'iT o a * hit with the fluid processing, the handling and sealing of wet capacitors, capacitor salvage and repair, and the - v quality control laboratory. Aroclor 1254, a commercial mixture of PCB isomers and homologs having the average molecular composition of a pentachlorobiphenyl, was used initially in the two plants (1946 fT.). During 1953-1955 a transition was made to Aroclor 1242, a PCB mixture having the average composition of a trichlorobiphenyl,. for all but a few spe cialty capacitors. In 1971, all use of Aroclor 1254 was phased out, and Aroclor 1242 was replaced by Aroclor 1016, a PCB mixture that also had the average compo sition of a trichlorobiphenyl, but which had been redis tilled by the manufacturer to remove all homologs containing more than four chlorines per biphenyl. Fi nally, on June 30, 1977, all PCB use ceased. The sta bilizer used with the Aroclor 1254 mixture was pchloroanthraquinone; that most commonly used with Arodors 1242 and 1016 was Union Carbide epoxide 4221. In late 1975, PCB area air levels (reported as Aroclor 1242) around the capacitor filling and sealing area were found to be in the range 200 to 2900 pg/m (geometric mean, 690 pg/m). Subsequent to the cessation of PCB use they dropped steadily to a mean of 16 pg/m in April 1983. The TWA levels indicated by personal air sam plers on directly exposed personnel in these areas in early 1977 were reported by Brown and Jones (79) to average 168 pg/m (range, 24-393 pg/m), or about half the mean area levels that we measured at that time. v These investigators also measured the levels of trich loroethylene (used in degreasing); tin, lead and zinc (sol dering operations); aluminum and iron (welding operations); and toluene and methyl isobutyl ketone (painting operations): all were generally within OSHA standards. Unfortunately, there are no data regarding PCB air levels during the 1946-1974 period. Major changes in plant ventilation and changes in the Aroclors used made it impossible to estimate those levels. Study Population The study population was selected in early 1976 to include all those workers (a) whose jobs required direct occupational contact with PCBs in high air level zones, such as those in handling and seating wet capacitors, and capacitor salvage and repair; (b) whose jobs did not involve direct PCB contact, but were at the periphery of the high exposure zone; and (c) whose duties required brief high exposure and contact, such as maintenance men. At the time of the 1976 examination, this study population included 152 males and 42 females. Of these, 41 had service times beginning during the period when the predominant dielectric fluid was Aroclor 1254, 103 began service when it was Aroclor 1242, and 50 had jeen exposed only to Aroclor 1016. The mean service L. >imes and ages in the study population were 17 years (range, 2-35) and 40 years (range 20-65), respectively. Thirty-three of the 194 were also members of the 326- member volunteer group examined by Fischbein et al. a little later in 1976 (IT). By the time of our 1979 ex amination, our directly exposed study group was re duced to 174 workers due to relocation, although still including 14 retired and/or relocated workers who re turned for the reexamination. At the time of the initial population selection an es timate of the relative PCB exposure was made by one of us (M. R.) on the basis of the individual worker's job activities. The estimated exposure categories (abbrevi ated as EXE below) generally corresponded to those used for selection, e.g., "low" for peripherally exposed jobs; "medium" for those with intermittent exposure; and "high" for those with continuous direct exposure. After the serum PCB data became available, analysis showed that the "low" and "medium" categories were statistically indistinguishable in terms of serum PCB means. Clinical Measurements The clinical examinations were conventional but rel atively complete. They included a medical history, phys ical examination by the plant physician (J.F.), ECG, chest X-ray, spirometry, 26-parameter biochemical an alysis, hematology, and urinalysis. Body weight was measured in light indoor clothing and height measured without shoes. Systolic and diastolic blood pressure were measured in the seated position to the nearest 5 mm. The participants were instructed to be fasting for the preceding 12-hr period with water ad libitum. The decision to measure serum PCB levels in 1976 was made during the course of the study, so that only 41 workers had the blood samples required for the PCB analyst drawn while fasting; the remainder were re called and sampled in the nonfasting state. All of the 1979 measurements were made on fasting sera. In 1979, the physical examination, ECG, and chest X-ray were repeated only in special cases, but the med ical history was expanded and improved, particu larly in relation to smoking habits. Documentation on alcohol consumption remained incomplete and generally unreliable. The chest X-rays were read by a local radiologist (R. C. Batt) who coded the findings according to the ILO/UC criteria (CRC/NIOSH (M) 2.8, Rev. 4/80). ECG's were interpreted by one of us (J.F.). Clinical chemical analyses were performed on the serum sam ples using an SMA-26, and hematological analyses on EDTA venous blood specimens, by Medpath (Teterboro, N.J.). The total serum lipids were measured in 1976 by using the phenol turbidity method, but not in 1979; the method has been criticized (26) as unreliable for total serum lipids. In 1979 serum triglycerides were reported numerically up to 400 mg/dL and thereafter as 5 400 mg/dL. WBC differentials were performed manually in 1976 and with the Technicon Hemalog D in 1979. Serum osmolarity was calculated from the elec trolytes and crystalloids (BUN, blood glucose) by con- 168 MUTCW ETAL. ventional procedures. The mean corpuscular hemoglobin was calculated from the reported red blood cell count and hemoglobin values in 1976. Serum PCB and DDE Measurements The PCB determinations upon both the 1976 and 1979 serum samples were performed by Hazelton Raltech (Madison, WI) using conventional PCB extraction, packed column gas chromatographic, and data reporting, procedures, which are discussed in detail elsewhere {27). These procedures also provided data on the levels of DDE present The observed distributions of PCB iso mer peaks in the capacitor workers' chromatograms are also described elsewhere (27); they differed markedly from those in the Arodor 1016,1242 and 1254 standards because of metabolism and elimination of all of the monoand dichlorobiphenyl isomers, most of the tri-and tetrachlor, and many of the pentachloro isomers'as well. The serum PCB levels were reported by the analyst in 1976 as levels of Aroclors 1242 and 1254, and in 1979 as Aroclors 1242, 1254, and 1260. These were calculated from the observed chromatograms by ratioing the sums of certain selected peak heights to those of the corre sponding peaks in Arodor standards. The selected peaks were, for Arodor 1242, those having retention times (relative to that for DDE, times 100) of 37, 70 and 84; for Arodor 1254, those at 125, 146, 160 and 186; for I Arodor 1260, those at 125, 146, 160, 186, 197, 294 and 372. Because of the differences in PCB composition be tween the serum samples and the standards, such values overstate the amounts of PCB actually present. The latter are now more commonly described in terms of values for LPCB ("lower PCBs," i.e., all isomers having gas chromatographic retention times on silicone less than that of DDE, which comes in the middle of the pentachlorobiphenyl range), and HPCB ("higher PCBs," i.e., those with retention times > DDE) with the sum of LPCB and HPCB representing the total PCB present (TPCB). We have shown elsewhere (27) that for our more heavily exposed subjects the factor for converting a reported serum Arodor 1242 value to LPCB is 0.237; for Arodor 1260 to HPCB, 0.71. The factor for con verting the 1976 Arodor 1254 values to HPCB was es timated from the chromatograms as 0.35. In 1980 it was found that the individual 1979 values of Arodor 1242, Arodor 1254 and DDE, relative to the population means, were highly correlated with the cor responding values in 1976, but that the 1979 values were generally several times higher, despite the limited ex posure to either PCB or DDE during the intervening period, indicating that one of the data sets contained a systematic error. Review of the methodology and tests on control samples confirmed the validity of the 1979 procedure (27); however, the standardization error in ^ the 1976 data could not be identified after the 4-year ' time lapse, and the analyst could not verify the 1976 serum PCB and DDE measurements. Accordingly, we have reported here all such data as the originally re ported numerical quantity times a correction factor x. Determination of the value of x was not required for the statistical association studies, but was for estimates of actual PCB levels and clearance. To estimate x, we used the DDE peak on the chromatograms as an internal standard; there had been no more than background ex posure of the study group to this substance, which is highly persistent in the body, during the 1976-1979 pe riod. The geometric mean DDE value for 1979 was cal culated directly from the reported data; that for 1976 using the maximum likelihood estimate for censored data (25), since 87 of the 140 1976 values for p.p'-DDE had been reported as ^ 1 ppb. The resulting values (10.2 and 1.18 ppb, respectively) indicated the mean value of x to be about 10, suggesting a possible decimal point error in 1976. Lipid PCB Determinations Serum cholesterol lipids (free cholesterol plus cho lesterol esters) were estimated from normal values (26) as 1.50 times the clinically reported serum cholesterol (i.e., cholesterol equivalent) values. Total serum neutral lip ids were calculated as the sum of the clinically reported serum triglycerides plus the serum cholesterol lipids. Serum lipid PCBs were calculated as the gross serum PCBs divided by serum neutral lipids. It is shown else where {23,21) that PCBs partition between-serum and adipose tissue fat in direct proportion to the content of neutral lipids as specified by this procedure; thus, this measure of the serum lipid PCB level is approximately equal to the PCB level in adipose tissue fat. However, the regression results were not particu larly sensitive to the exact procedure used for calcu lating serum lipid phases and comparable findings were obtained when they w&re calculated instead as simply the sum of the serum triglycerides and the serum cholesterol. Statistical Analyses Statistical analyses were performed on a Honeywell 600/6000 computer, using STATPAC {23), a versatile general statistical program with many of the capabilities of the more widely used BMDP series {29), for the dis tribution plotting; simple, multiple and stepwise linear regression; and maximum likelihood estimates for cen sored data. To evaluate the distributions of the dependent vari ables, probability plots of both the measured and logarithmatically transformed data were prepared and compared for linearity. The choice for a number of the distributions seemed equivocal, with the arithmetic and geometric mens approximately equal. We generally selected the distributions yielding the highest F-ratios in the regressions, or the distributions which tended to produce normal residuals. Partial residual plots were prepared and examined in cases where there was a sig nificant PCB association. Outliers were not removed. The residuals from the regression appeared to be nor mally distributed with only a few exceptions (blood, glucose, red cell indices), in which cases the distribu tions were not improved by data transformation. /: mM . A X P Hi:.\tATOtSX:ttV.! :xn.< ix im Mi'n't U! inwkkus irer Tlie independent. variables used in the multiple and stepwise regressions* as identified from preliminary studies, were as follows: Sex (SEX). Workers were coded as males = 1 and females = 2 so that a positive association repre sented an attribute of females. Age (AGE). The numerical age to the nearest year in 1976 was used. Service time and age covaried (r = 0.75). Urinary Specific Gravity (SPG). A number of var iables appeared sensitive to the degree of hydration in these fasting subjects, principally the crystalloids (blood glucose, blood urea nitrogen). Disease Status (DS). Thirteen workers were iden tified as having significant conditions bearing upon the analysis. They included diabetics (5), those with alcohol problems (3), nonfasting at examination time (3), one worker who was post-pituitary surgery, and one worker with epilepsy taking Dilantin. A dummy variable was created (disease absent = 0, disease present = 1) to identify these cases. Nonfasting PCB Measurement (FNF). In the 1976 ' - data individuals whose serum FCB levels were ob tained in the nonfasting state were coded as 0; fast ing measurements were coded as 1. Smoking (SC). Smokers and ex-smokers were coded as 2, nonsmokers as 1. Job Status (JS). For the 1979 analysis, workers "continuing in employment were coded as 0, sepa\ rated or retired workers as 1. Body Mass Index (BMI). The body mass index (W/ H) was calculated for each individual in 1976 and 1979 and used as an independent variable. In the stepwise regressions, the criterion for adding an independent variable was an F value of 3.0; that for removing a variable was 4.0. The numbers of subjects in the stepwise regressions were only 159-166, rather than the 189-194 used in the simple regressions, be cause of incomplete data. In addition, 41 cases had miss ing hematological data in 1976 (Table 1) yielding an N of 128 for these analyses. Because half of the 1976 gross serum DDE values were reported only as lx ppb, a dummy variable (=s la: ppb = -1 , > lx ppb = +1) was used in the regres sions, and no attempt was made to estimate serum lipid DDE values. The 1979 gross serum DDE and serum lipid DDE levels were expressed in the same terms as the corresponding measures of PCB. Results Clinical Characteristics of the Study Population The medical history and physical examination of the \u d y population revealed a variety of clinical conditions , ^fid health-related practices. Sixty-five percent of the population were smokers or ex-smokers, and 26 workers (17%) showed obstructive patterns on repeated spirometric testing. There were six chest X-ray reports of possible emphysema. Fifty percent of the population was overweight [> 1.2 times the "desirable" weight by 1959 standards (.?0)J and many were frankly obese. Hypertension histories were obtained on 11.3%. Elevated diastolic pressures (>90 mm Hg) were found in 19.6% of the population at examination in 1976 as compared to 14.7% at the time of employment. Reported urinary tract problems .in cluded renal stone or "gravel" in 8 subjects; surgically corrected congenital obstruction in one; bladder, kid ney, or prostatic infection in 14; and one benign bladder tumor. Forty-five workers showed one or more elevated fast ing blood glucose values between 1976 and 1979. Of these, 5 were known diabetics, 30 were found to give normal postprandial values, and the remainder showed normal fasting values when retested. A review of the medical dispensary records of all PCBexposed employees between 1960 and 1975 revealed 49 visits for contact dermatitis attributed to the dielectric fluid. There were an additional 16 visits for nausea, dizziness, or eye or nasal irritation following short initial exposure to the workplace. In all cases the conditions subsided with removal from exposure. Chloracne was never observed in the plant work force. Clinical Laboratory Findings The distributions of clinical chemical and hematolog ical values observed in the study population are de scribed in Table 1, along with the normal ranges unadjusted for sex and age, and asterisks indicating variables having values outside these rough reference ranges.! The numbers of individuals reported by the clinical laboratory to be outside its internal age- and sex- ad justed standards (based on 2 standard deviations from mean values) are listed for each reported param eter in Table 2. If the parameters were normal and normally distributed, one would expect four or five cases to fall along each tail of these distributions, making a total of eight to ten. Table 2 indicated substantial numbers of individuals with elevations in serum triglycerides, cholesterol, and SGPT in both 1976 and 1979, and in blood glucose and albumin/globulin ratio in 1976. Some elevations in BUN and chloride were found in 1979. In 1976 there were some elevations in total WBCS associated with de creased PMNs and increased lymphocytes, monocytes, and eosinophils. In 1979 there were marginal increases in monocytes and eosinophils, but the WBCs were near normal. Urinalysis in both 1976 and 1979 showed ele vated cell counts, but no attempt had been made to obtain a clean catch. Urinary specific gravity and cal culated serum osmolarity were both elevated in 1976 and 1979 (Table 1). The observed distributions of serum triglycerides and serum cholesterol are shown in greater detail in Figures 1 and 2. The cross-hatched bars represent values that 170 LAWTON ETAL, Table 1. DUtribution o f clinical l a b o r a t o r y finding in atudjr population of PCB-exposed capacitor worker. Cliniml mfH.turvnu-iUf.* 1976 (N - 194) 1979(N - 1741 Sl<l. AM ar Percentiles Mewing AM or IVrcenlile? Miaatng Unite !Dial. nunn*1' Mean sn GM 6th 95th valuee Mean 5D GM 5th 95th valuer Serum triglycerides Senun total cholesterol Serum neutral lipids'1 SCOT SGPT SGGTP Serum alkaline phosphatase Serum lactic dehydrogenase Serum total protein Senun aibumnin Serum globulin Albuming/globulin ratio Blood urea nitrogen Serum creatinine Bun/creatinine ratio Serum total bilirubin Senun direct bilirubin Blood glucose Senun uric acid Serum Na Serum K Serum C m Senun Mg Serum Cl Serum P . Senun osmolarity . Specific gravity (urine) `'Serum iron RBC HG HCT MCV MCH' MCHC WBC PMNS # PMNS Lymphocytes % Lymphocytes Monocytes % Monocytes Eosinphils % Eosinphils Basophils % Basophils mg/dL LN 50-200 2.133 0.224 135.8 58.9 310* 1 1167 0.225 143.7 61.3 336.9* mg/dL N 125-300 251.3 55.2 251.3 169.2 3415* 1 238.1 50.1 238.1 155.7 320.4* mg/dL N 2.709 0.127 5111 316 829 1 1702 0.119 603.3 321 789 lU/L LN 1-50 1.374 0.152 13.3 23.7 41.9 1.448 0.144 28.0 16.3 48.4 1U/L LN 1-55 1.460 0.189 28.9 14.1 58.8* 1 1.537 0.164 34.4 18.5 63* U/L LN 1-70 1.091 0.266 113 4.50 33.8 2 1.198 0.263 15.8 5.82 417 IU/L LN 15-55 1.419 0.142 26.2 16.8 43.4 1.466 0.122 29.1 18.4 46.3 IU/L N 57-250 160.1 28.2 160.1 114.6 205.4 173.6 31.3 173.6 1211 225.1 g/dL LN 12-8.3 0.858 0.079 7.21 6.46 8.04 0.864 0.028 7.30 6.68 111 g/dL LN 3.8-5,1 0.648 0.032 4.44 192 102 1 0.644 0.036 4.41 184 6.05 g/dL N 11-4.0 2.77 0.51 177 1.93 160 3 192 0.43 192 121 3.63 LN 1.2-2 0.213 0.099 1.63 1.12 138* 3 0.182 0.082 1.52 1.11 107 mg/dL LN 7-2S 1.185 0.119 15.3 9.72 24.0 1 1.227 0.115 16.9 10.9 26.0 mg/dL N 0.8-LB 1,30 0.21 1.30 0.96 1.64 1 1.32 0.20 1.32 0.99 1.65 LN 1.078 0.124 110 7,45 19.10 2 1.112 0.127 110 100 20.9 mg/dL LN 0.1-1.4 -0.339 0.193 0.46 0.22 0.95 -0.218 0.168 0.61 0.32 1.14 rnd/dL N 0-0.4 0.15 0.08 0.15 0.02 0.27 1 0.14 0.07 0.14 0.02 0.26 mg/dL LN 65-130 2.010 0.057 1014 82.8 126.5 1 1.995 0.056 98.8 79.8 122.3 mg/dL N 3.3-8.9 5.74 1.25 5.74 3.65 7.80 1 6.01 1.30 6.01 3.88 115 mmole/L N- 134-146 139.5 167 139.5 135.1 143.9 2 139.4 132 139.4 135.6 143.2 nunole/L N 3.4-5.4 4.26 0.44 4.26 154 4.98 2 4.42 0.46 4.42 3.66 5.17 mg/dL N 8.7-10.5 9.87 0.49 9.87 9.06 10.7* 1 9.67 0.42 9.67 197 10.4 meq/L N 1.6-2.3 1.91 0.15 1.91 1.66 115 mmole/L N 96-109 10SL0 133 1010 97.1 106.8 2 103.8 3.27 103.8 98.4 109.1* mg/dL ' N 1.9-4.3 3.01 0.44 101 128 174 1 185 0.44 185 112 3.58 mosmdtg N 280-295 299.0 6.06 299.0 288.9 309.0* 2 ' 299.5 5.37 299.5 290.6 304.3* g/cm* N 1.026 0.005 1.026 1.017 1.035* 5 1.026 0.005 1.026 1.018 1.03* ug/dL N 55-200 1)2.1 36.1 1111 53.1* 170.6 2 108.2 29.1 108.2 60.3 156.2 HP/mm* N 4.3-6.4 4.94 0.42 4.94 4.24* 5.63 1 ' 5.05 0.39 5.05 4.39 5.70 g/dL N 13.0-18.5 14.8 1.67 14.8 115* 17.0 1 15.4 1.18 15.4 13.4 17.4 % N 39.5-57.0 46.2 3.55 46.2 40.3 52.1 n* N 80-103 91.6 5.18 91.6. 83.1 100.0 N 26-33 30.0 1.89 30.0 26.9 33.1 l 30.6 1.61 30.6 27.9 33.1 % N 30.6-36.0 33.5 105 33.5 31.2 35.7 lO'/mm* N 4.1-11.9 7.02 1.74 7.02 4.21 9.81 2 6.90 1.87 6.90 3.85 9.92 HP/mrn* N 1.65-8.33 4.20 1.32 4.20 102 6.37 41 4.12 1.46 4.12 1.77 6.45 % WBC N 45-77 59.7 9.57 59.7 44.3 74.9 41 59.0 8.65 59.0 44.6 73.2 ltf/mm1 N 1.05-3.58 2.33 0.76 133 l.b7 3.57 4) 109 0.64 109 1.03 3.14 % WBC N 16-45 33.8 9.64 33.8 18.2 49.2* 41 30.9 7.84 30.9 17.9 43.9 lOVmm5 N 0.06-0.93 0.30 0.17 0.30 0.02 0.59 41 0.48 0.22 0.48 0.12 0.83 % WBC N 0-8 4.38 114 4.38 0.87 7.95 41 6.90 1S7 6.90 160 11.2" KT/mm* N 0.04-0.42 0.13 0.13 0.13 0 0.32 41 0.16 0.12 0.16 0 0.33 % WBC N 0-4 1.84 1.73 1.84 0 4.34* 41 134 1.78 134 0 4.8* lOVmm1 N 0.01-0.15 0.046 0:026 0.046 0.003 0.090 % WBC N 0-2 0.68 0.38 0.68 0 1.30 2 2 - *All blood chemical determinations made on senun: hematology measurements in EDTA venous blood. Abbreviations: SCOT, serum giuUmic-oxaiacetic transaminase; SCOT, serumglutamic-pyruvic transaminase; SGGTP, serumfgiutamyltranspeptidase; LOH, serum lacticdehydrogenase; A/GRatio, albumin/globulin ratio; BUN, blood u m nitrogen; RBC, red blood cell count; HG, hemoglobin; HCT, hematocrit; MCV, mean red cell corpuscular volume; MCH. mean red cell corpuscular hemoglobin; MCHC, mean red cell corpuscular hemoglobin concentration; WBC, white blood cell count; PMN, polymorphonuclear white cells. `laboratory standard ranges unadjusted for sea and age. Asterisk () indicates values outside these ranges. Arithmetic (AM) or geometric (GM) means given based on selected distribution (N ** normal, LN lognormal). fell outside the 95th percentile for age from the Lipid Research Clinics Prevalence Study (Si). By theL.RCP study criteria, the portions of the male population ex hibiting elevations in triglycerides in 1976 and 1979 were 5.7% and 12.6%, respectively. This elevation was sig nificant at the 1% level by paired -test. For females, the elevation in triglycerides was significant at the 5% level but was found in the younger women and appeared to be related in part to obesity and estrogen medication. Of unknown significance were the differences in ana lytical methodology (fluorimetric method with Techni- con Autoanalyser used by LRCPS; enzymatic method with discrete analyser used here). The serum choles terol values (Fig. 2) were normally distributed. By the LRCPS criteria, the portions of the entire group above the 95th percentile in cholesterol in 1976 and 1979 were 29.4% and 23.3%, respectively. Measures of PCB Exposure Table 3 shows the means and ranges for serum PCB concentrations described in four ways: first as the Ar- HlttrilKMK'Ah AS'h HKy.XTnlJC.fM, H X PIXCS IX ( \ U \ l t T(Hi WtiRKKICS ITI Ibble 2. Number or study jrroup member reported by clinical lalHiraUny to have clinical variables outride normal (95^) ave- and sex-idjusUri mnees. Clinical variables' Trigylcerides Tbtal cholesterol SGPT SCOT GGTP Alkaline phosphatase LDH Blood glucose Uric acid Tbtal bilirubin Direct bilurubin Tbtal protein Albumin Globulin A/G ratio BUN Creatinine Na K Cl Ca P Mg Iron EBC Hemoglobin Hematocrit MCV MCH MCHC WBC Differential PMNs Lymphocytes Monocytes Eosinophils Basophils Urinalysis Albumin Acetone Cells 1976 (N - 194) Males Females Tbtal 36 27 15 6 6 1 3 25 2 0 1 4 0 9 12 4 1 3 3 1 I0 4 7 0 1 0 1 0 4 1 0 0 0 0 3 2 2 0 2 0 0 1 0 40 34 15 7 6 2 3 29 3 0 1 4 0 . 12 14 6 1 5 3 1 2 0 5 16 2 13 303 10 1 + /- Split1' *38/2 '33/1 ISA) 7/0 6/0 2/0 3/0 *29/0 2/1 0/0 1/0 4/0 0/0 7/7 13/1 6/0 1/0 4/1 1/2 0/1 0/2 0/0 5/1 1/2 3/0 1/0 11 2 13 9/4 7 4 11 29* 8 5 13 11/2 7 2 9 9/0 6 1 7 7/0 * 3 2 5 5/0 1 1 2m 12 12 24 24/0 Males 38 20 12 3 3 5 2 3 3 2 1 2 8 1 6 6 1 2 5 9 3 1 1 6 '1 2 4 4 10 1 8 1 8 3 4 1979 {N - 174) Females Tbtal 3 41 3 23 1 13 03 03 27 13 3 .6 03 24 12 2 .4 2 10 23 28 39 01 02 16 4 13 25 01 01 06 01 02 04 04 1 ' 11 01 3 11 01 2 10 36 26 01 1 21 3 48 12 + /-^plith 41/0 22/1 13/0 3/0 3/0 4/3 3/0 6/0 3/0 4/0 2/0 3/1 1/9* , 2/1 4/4 9/0 1/0 0/2 5/1 11/2 3/2 0/1 0/1 2/4 0/1 1/1 5/1 3/1 *8/3 0/1 7/4 1/0 5/5 6/0 6/0 1/0 3/0 12/0 'See Tbble 1 for abbreviations used. hAsterisk (*) indicates significant probability (95%) of increase or decrease assuming a true binomial proportion of 0.5. oclor levels reported by the analytical laboratory, based on measurements of selected peak heights; second, as LPCB or HPCB values, representing total PCBs ac tually present in the specified ranges, determined by applying appropriate conversion factors to each re* ported Aroclor value (7); third, as total PCBs (TPCB), the sum of the LPCB and HPCB values obtained from the Aroclor 1242 and 1260 values, respectively; and fourth, as serum lipid PCB values, obtained by dividing the serum LPCB, HPCB, or TPCB values by the total serum neutral lipid. Also shown are the corresponding data for p,p'-DDE, which was present at background *, levels. 1 . ' Reference values were obtained on 25 office workers a Connecticut office in 1979. Their geometric mean analytical values were for serum Aroclor 1242, 6.6 ppb; for Aroclor 1254, 14.4 ppb; for Aroclor 1260, 8.3 ppb; for p,p'-DDE, 11.3 ppb. Thus, taking the correction factor x for the 1976 data as 10, the mean xenobiotic levels in the study population in 1976 and 1979 were, for Aroclor 1242, about 220 and 40 times the background levels, respectively; for Aroclor 1254 and 1260, about 4 -6 times the background; and for DDE, about the same as background. Also for r = 10, Table 3 indicates that in the 44 to 45 months between the 1976 and 1979 examinations, the geometric mean level of serum Aroclor 1242 fell from 1470 to 277 ppb, or 81%, and that of serum Aroclor 1254 from 84 to 55 ppb, or 35%. Both calculations must be taken with caution, given the imprecision of the ana lysis, since they are uncorrected for changes in the composition of the study population and highly depen- 172 LAITON ETAL. t m SCRUM TRISLYCERIO ES (MGMSfOU F igure 1. Serum triglyceride levels (mg/dL) in the study popula tion. Cross-hatched area indicates values falling above the 95th percentile for age from the Lipid Research Clinics Prevalence Study (iJ). dent upon the estimated value of the correction factor, x. The implied extent of clearance of retained LPCB is less sensitive to the estimated value of x, but more difficult to interpret, since some direct exposure to Aroclor 1016 continued during the first third of the interval between the examinations. Investigation of the rates of clearance of individual PCB isomers in this population is continuing. Because of the analytical uncertainties in the 1976 serum PCB data, we also examined two other measures of PCB exposure, namely, the estimates of relative ex posure based on analysis of the individual job activities, and the measurements of PCB air levels in the various working areas, made at the time of the 1976 examina tion. The correlations of these measures with the 197f>` and 1979 serum LPCB and HPCB values are shown in Table 4. The measurements of area air levels proved to be a N^oor estimate of exposure, and were dropped from fur ther consideration in the regressions against clinical variables. The relative exposure estimate showed a high correlation with serum LPCB in 1976 (r = 0.73, de clining to 0.51 in 1979), indicating that it reflected es timated current exposure to Aroclor 1016, which was 1979 SERUM CHOLESTEROL (MGMS/DL) F icure 2. Serum total cholesterol levels (mg/dL) in the study pop ulation. Cross-hatched areas indicate values falling above the 95th percentile for age from the Lipid Research Clinics Prevalence Study (31). in use at the time. The correlations between the ex posure estimate and the serum HPCBs were lower but equal in 1976 and 1979 (r = 0.36). Relationships of Independent Variables to PCB Exposure The correlation coefficients for the association of the individual independent variables selected for the re J l/nciiKMlt'M. .l.v/i UKM.\Tu.>h;i V.O/.V(,>* /.Y C. W .W 77/JA- n'OA'A7;/;v 17:1 Table 3. Distributions, of serum PCB and DDE in stud; population of PCH-expoed capacitor Wirkers. Descriptor of serum parameter log mean - log r* s SD 1976 Percentiles <* 5 95 Measures of lower PCBs Arcelor 1242, ppb LPCB. ppb Lipid LPCB, ppm Measures of higher FCBs Aroclor 1254, ppb Aroclor 1260, ppb HPCB (1254), ppb HPCB (1260), ppb Lipid HPCB (1254), ppm Lipid HPCB (1260), ppm Measures of total PCBs TPCB (1242, 1260), ppb Lipid TPCB, ppm Measures of p.p'-DDE DDE, ppb Lipid DDE, ppm 2.167 0.486 147. lx 23.4x 926.2x 1.560 0.481 36.3x 5.7x 226.9x 0.96S 0.474 9.3x 1.5x 56.3x 0.924 0.414 0.472 0.411 -0.120 0.396 8.4x l.Sx 3.Ox 0.6x 0.8x 0.2x 40.3x 14.2x 3.4x 0.033 l.OSx * GM = geometric mean; MV = missingjralues. k Most likely value of log z = 1.0; z = 10. 1979 log MV' mean z SD ' GM' Percentiles 5 95 9 2.443 0.463 277.1 48.1 1598.1 9 1.832 0.463 67.9 11.8 391.6 9 1.245 0.432 17.6 3.4 90.3 9 1.740 0.442 54.9 10.3 292.4 1.543 0.447 34.9 6.4 190.0 9 1.285 0.442 19.3 3.6 107.7 1.375 0.447 23.7 4.4 129.2 9 0.695 0.400 5.0 1.1 22.7 0.785 0.406 6.1 1.3 28.6 1.979 0.437 95.2 18.2 499.0 1.392 0.405 24.6 5.3 114.4 27 1.005 0.318 10.1 3.0 33.7 0.305 0.284 2.0 0.7 6.0 MV" 1 1 1 1 30 1 30 1 30 30 30 1 1 Table 4. Correlation coefficients for associations between serum PCB level*, area air levels, and the job analysis exposure estimate in 1976 and 1979. % H 8 LPCBs (1242) Gross serum concn. Serum lipid concn. HPCBs (1254) Gross serum concn. Serum tipid concn. HPCBs (1260) Gross serum concn. Serum lipid concn. Exposure estimate 1976 Exposure estimate (N - 181) Area air level <N 180) 0.731 0.727 0.368 0.360 0.144 0.167 t r 0.080 0.102 1.0 -0.084 Exposure estimate W -> 172) 1979 Area air level 0.506 0.516 - 0.014 0.029 0.360 0.369 -0.029 0.010 0.359 0.363 1.0 -0.025 0.014 -0.164 gression study with the various measures of PCB ex posure, as obtained by simple regression, are shown in Table 5. The tabulated data indicated strong associations of age and/or service time with HPCBs in both 1976 and 1979, and with LPCBs in 1979, probably reflecting a time-dependent accumulation of the more slowly me tabolized PCB isomers. There were significant negative correlations between the body mass index and the serum, lipid LPCB in 1976 and significant positive correlations with the gross serum LPCBs and HPCBs in 1979. In 1979, LPCB values were significantly higher in females than males (positive sign of coefficient) but lower in "Tnokers (negative sign), probably reflecting oxidase in duction (52). 1979 PCB levels in retirees were not sig nificantly different from those in workers who continued in employment. Urinary specific gravity (SPG) and the presence of diabetes or alcoholism showed no association with serum PCB levels, but SPG was the one inde pendent variable showing association with the exposure estimate in 1976. Higher levels of LPCBs were found in nonfasting individuals in 1976. Relationships of Clinical Laboratory Findings to PCB Exposure Tables 6, 7, 8 and 9 show the results of applying the three regression procedures to the clinical laboratory findings using as independent variables the 1976 and 1979 levels of LPCB, HPCB, and DDE (each expressed as both the log gross serum and log serum lipid value except for DDE in 1976), and the PCB exposure esti mate (EXE), respectively. In each case, we have tab ulated the regression coefficients for the associations with PCB or DDE, and their confidence intervals for those clinical parameters that showed a significant 174 LAW TON E T AL. Dependent variable Grew aerum PCBs LPCBs, 1976 LPCBs, 1979 HPCBs, 1976 HPCBs, 1979 Serum lipid PCBs LPCBs, 1976 LPCBs, 1979 HPCBs. 1976 HPCBs, 1979 Exposure estimate 1976 1979 Statistical significance < 5%. Ifeble 5. Correlation coefficient! relating measure* of PCB exposure and the elected Independent variable*. Independent variables Age Ser Sex SPG BMI DS SC FNF JS 0.107 0.259- 0.4210.516* 0.073 0.011 -0.121 0.191* 0.160- 0.084 0.394- -0.091 -0.111 0.549* -0.026 0.037 -0.090 -0.045 0.208* 0.096 - 0.013 -0.0S6 0.189* 0.141 -0.132 - 0.188- - 0.173- - 0.015 0.009 -0.021 -0.050 0.005 0.028 0.209* 0.340* 0.493* -0.007 0.030 -0.111 0.140 0.185* 0.077 0.314* - 0.014 -0.099 0.535* 0.014 0.026 -0.172* -0.121 0.136 0.061 -0.109 -0.001 0.114 0.109 -0.172* -0.211* -0.199* -0.008 -0.037 -0.045 -0.069 0.014 0.004 -0.009 -0.040 - 0.168* -0.106 -0.115 -0.109 -0.138 0.004 -0.009 -0.040 -0.062 -0.036 -0.115 -0.109 0.084 Table 6A. Regression and correlation coefficients for 1976 and 1979 clinical variables showing at least one significant association with a measure of serum LPCB. as determined by three regression procedures: log gross serum LPCBs. Dependent variable* 1976 Log triglycerides Cholesterol Log SGFT LogGGTP Log total bilirubin Direct bilirubin Uric acid Log blood glucose Phosphorus RBC Lymphocytes X 2.14 252 1.46 1.10 -0.34 0.15 5.73 2.01 3.01 4.94 2.33 __________ Log grass serum LPCBsb_____ Simple regression Multiple regression . Confid. limits Confid. limits 9.* 5 95 . 5 95 0.07 24.90.03 0.07 -0.09* -0.03* 0.08 -0.01 0.14* -0.07 0.03- 0.01 8.48 -0.01 -0.01 -0.14 -0.06 -0.31 -0.03 0.01 -0.20 0.04 0.14 41.4 0.09 0.15 -0.03 -0.01 0.46 0.01 0.27 0.06 0.55 0.10* 0.03 24.3* 7.15 0.07* 0,02 0.12* 0.04 -0.08* -0.15 -0.04* t -0.06 0.31 , -0.01 -0.001 * -0.02 0.20* 0.04 -0.07 -0.18 0.42- 0.12 0.17 41.5 0.13 0.21 -0.02 -0.01 0.64 0.02 0.35 0.05 0.71 ______Step regression Confid. limits . 5 95 0.10* 25.7* 0.08* 0.13* -0.09* -0.04* 0.35* -0.001 0.16* -0.08 0.35* 0.03 9.49 0.02 0.05 -0.15 -0.06 0.04 -0.02 0.01 -0.19 0.07 0.16 42.0 0.13 0.21 -0.03 -0.01 0.66 0.01 0.30 0:02 0.63 1979 Log triglycerides Cholesterol LogGGTP Log total protein Globulin Hemoglobin % Lymphocytes Monocytes % Monocytes 2.16 238 1.20 0.86 2.92 15.4 30.9 0.48 6.96 0.18* 34.3* 0.11* 0.01* 0.19* -0.04* -0.49 0.07* 1.18* 0.11 18.8 0.03 0.002 0.05 -0.78 -3.02 0.01 0.38 0.25 49.8 0.20 0.02 0.33 -0.02 2.04 0.14 1.99 0.15* 25.6* 0.10* 0.01 0.10 -0.20 -1.92 0.09* 1.41* 0.08 8.81 0.02 -0.001 -0.05 -0.56 -4.69 0.02 0.54 0.22 42.4 0.19 0.02 0.24 0.16 0.85 0.17 2.27 `See Table 1 for abbreviations. Asterisk (*) indicates 95% confidence limits did not include zero. bN (1976) for simple regression 189-194. N for multiple regresin and step regression = 159-166 except hematology where there are 41 missing values flhble 1). v 0.14* 24.0* 0.12* 0.01 0.13 -0.20 -1.44 0.08* 1.41* 0.07 7.90 0.04 -0.001 -0.005 -0.54 -4.31 0.01 0.58 0.21 40.1 0.20 0.02 0.26 0.13 1.00 0.15 2.23 association in one of the six regressions performed. The test of significance was that the 95% confidence interval for Pi not include zero; statistically significant associa tions are indicated by asterisks. In addition, in Tables 6 - 8 we have listed the partial r values for the backward step regressions against the log serum lipid xenobiotic (e.g., LPCB, HPCB or DDE) and the other independent variables. The partial r values listed are generally those where the regression coefficient was found to be of sta tistical significance, which, depending on the number of degrees of freedom, generally corresponded to a partial r > 0.15; in addition, we have listed some borderline associations, as indicated by partial r values of 0 .1 4 0.15. A comparison of the pt values across these various regressions indicates the sensitivity of the relation to confounding by the other independent variables, and / KMICAL AX!) ti EMA `U nite AL FlXlttML' IX <ACAt ITUti Wt UiiCKUS 17.') T*bk 6U. Regression and correlation coefllcienU for la'fi and 1979 clinical variables showing at least one sign neurit association with a measure of serum LPC8, as determined by three regression procedures lor serum lipid LPCHs. Dependent variable" 1976 Log triglycerides Cholesterol Log SGPT LogGGTP Log total bilirubin Direct bilirubin Uric add Log blood glucose Phosphorus RBC Lymphocytes ____________________ Log serum lipid LPCBs* Simple regression Multiple regression Confid. limits Confid. limito fi. 5 95 fi. 5 95 -0.04 -1.65 -0.002 0.02 -0.08* -0.03* -0.09 -0.02* 0.16* -0.01 0.23* -0.11 -18.8 -0.06 -0.06 -0.04 -0.05 -0.48 -0.04 0.02 -0.23 0.01 0.03 15.5 0.06 0.10 -0.02 -0.01 0.30 -0.003 0.29 0.03 0.54 0.01 1.82 0.05 0.10* -0.09* -0.04* 0.25 -0.003 0.22* -0.11 0.38* -0.07 -16.4 -0.01 0.01 -0.15 -0.06 -0.09 -0.02 0.06 -0.23 0.08 0.08 20.1 0.11 0.19 -0.02 -0.01 0.59 0.01 0.38 0.01 0.69 Step regression Confid. limito fi. 5 95 0.01 1.82 0.05 0.11* -0.09* -0.04* 0.31 -0.003 0.19* -0.12* 0.39* -0.07 -16.4 -0.01 0.02 -0.15 -0.06 -0.02 -0.02 0.04 -0.23 0.09 0.08 20.1 0.10 0.19 -0.03 -0.01 0.63 0.01 0.34 -0.01 0.69 1979 Log triglycrides Cholestrol LogGGTP Log total protein Globuiin Hemoglobin % Lymphocytes Monocytes. % Monocytes 0.08 12.2 0.07 0.007 0.12 -0.59* -1.67 0.08* 1.36* -0.01 -5.20 -0.02 -07003 -0.26 -1.00 -4.38 0.01 0.50 0.16 29.6 0.16 0.02 0.27 -0.19 1.03 0.16 2.21 0.05 2.64 0.07 0.003 0.02 -0.31 -3.26* 0.11* 1.61- -0.03 -15.5 -0.02 -0.01 -0.13 -0.69 -6.14 0.03 0.71 0.13 20.8 0.16 0.01 0.17 0.06 -0.39 0.19 2.52 0.04 2.88 0.07 0.003 0.02 -0.32 -1.95 0.10* 1.58* *N (1979) for simple regression 172-174. N for multiple regression and step regression 165-166. __ -0.03 -14.7 -0.02 -0.01 -0.13 -0.68 ^.65 0.02 0.71 0.12 20.4 0.16 0.01 0.17 0.03 0.75 0.17 2.46 Table 6C. Regression and correlation coefficients for 1976 and 1979 clinical variables showing at least one significant association with a measure of serum LPCB, as determined by three regression procedures independent variables. *------------------------------------------------------------------ _______________Independent variables* ^ ________ ______________ Step regression Confid. limits Dependent variable* fi. 5 95 1976 Log triglycerides 0.01 Cholesterol 1.82 Log SGPT 0.05 LogGGTP 0.11* Log total bilirubin -0.09* Direct bilirubin . -0.04* Uric add 0.31 Log blood glucose -0.003 Phosphorus 0.19* RBC -0.12* lymphocytes 0.39* -0.07 -16.4 -0.01 0.02 -0.15 -0.06 -0.02 -0.02 0.04 -0.23 0.09 0.08 20.1 0.10 0.19 -0.03 -0.01 0.63 0.01 0.34 -0.01 0.69 Age 0.20 0.35 -0.47 0.14 Sex -0.22 -0.29 -0.19 -0.17 -0.58 SPG -0.15 0.20 partial r ______________________________ BM1 DS Log SC JS FNF (LPCB)sl 0.2S 0.26 X 0.20 X 0.35 X 0.15 0.15 X 0.29 0.20 -0.19 X -0.22 -0.16 -0.20 X 0.17 -0.21 0.45 X 0.16 0.32 0.39 X 0.19 0.16 X 0.19 X -0.17 X 0.16 0.23 1979 Lag triglycerides Cholesterol Log GGTP Log total protein Globulin Hemoglobin % lymphocytes Monocytes % Monocytes 0.04 2.88 0.07 0.003 0.02 -0.32 -1.95 0.10* 1.58* * See text for abbreviations. -0.03 -14.7 -0.02 -0.01 -0.13 -0.68 -4.65 0.02 0.71 0.12 20.4 0.16 0.01 0.17 0.03 0.75 0.17 2.46 0.19 0.34 0.16 -0.14 \ * -0.26 -0.54 -0.16 0.40 0.18 0.18 0.21 0.23 0.27 -0.15 0.19 -0.15 0.19 0.21 -0.21 -0.23 X X X X X X X X 0.20 0.27 17G LAWTON ET AL. Table 7A. Regresaion and correlation coefflcienU for 1976 and 1979 clinical variables .howinjr at least one .i^ ific a n t association with a measure of serum HPCIJ (1254). as determined by three regression procedures: log hubs serum HI CBe c Dependent variable" 1976 Log triglycerides Cholesterol Log SGPT Log GGTP Log alk. phos. Log total bilirubin Direct bilirubin Log total protein Lag albumin Log BUN Creatinine Log B/C ratio Iron RBC WBC Lymphocytes % Monocytes r 2.14 251.9 1.46 1.10 1.42 -0.34 0.15 0.86 0.65 1.19 1.30 1.08 110.7 4.94 7.0 2.33 4.37 Loft gross serum HPCBs Simple regression Multiple regression Confid. limits Confid. limits . 5 95 . 5 95 0.13* 38.7* 0.04 0.12* 0.07* -0.08* -0.03 -0.002 -0.01 0.04* -0.01 0.05* -13.6-0.08 1.00* 0.23* -1.07* 0.06 19.8 -0.03 0.03 0.02 -0.15 -0.05 -0.01 -0.02 0.001 -0.09 0.002 -26.1 -0.23 0.40 0.02 -1.91 0.21 57.6 0.11 0.21 0.11 -0.02 0.001 0.01 0.003 0.09 0.06 0.09 -1.16 0.07 1.59 0.63 -0.23 0.11* 31.1* 0.08* 0.15* 0.06 -0.12* -0.04* -0.01 -0.008 0.02 -0.06 0.04 -10.5 -0.06 1.34* 0.72* -1.03* 0.01 7.63 0.002 0.03 -0.004 -0.21 -0.08 -0.02 -0.02 -0.04 -0.15 -0.02 -27.3 -0.21 0.55 0.29 -2.98 0.20 54.6 0.15 0.26 0.13 -0.03 -0.006 0.01 0.007 0.06 0.03 0.09 6.33 0.10 2.14 1.15 -0.63 Step regression Confid. limits . 5 95 0.12* 35.1* 0.06 0.15* 0.07* -0.10* -0.04* -0.006 -0.007 0.02 -0.05 0.04 -10.1 -0.09 0.95* 0.70* -1.67 0.04 13.1 -0.003 0.05 0.01 -0.18 -0.07 -0.02 -0.02 -0.03 -0.13 -0.01 -24.1 -0.22 0.2S 0.27 -2.61 0.21 57.1 0.13 0.25 0.12 -0.03 -0.008 0.01 0.008 0.06 0.04 0.08 4.14 0.04 1.63 1.13 -0.73 1979 Log triglycerides Cholesterol Log GGTP Log alk. phos.. Direct bilirubin Log blood glucose Globulin Log A/G ratio Log BUN Log B/C ratio Monocytes % Monocytes 2.16 238.2 1.20 1.47 0.14 2.00 2.92 0.18 1.23 1.11 0.48 6.96 `See Thble 6 footnotes. 0.22" 47.3* 0.14* 0.06* -0.02 0.04 0.21* -0.04* 0.08" 0.06' 0.08* 1.23* 0.15 31.6 0.05 0.02 -0.05 -0.02 0.07 ` -0.06 0.04 0.02 0.01 0.38 0.29 63.0 0.22 0.10 0.004 0.06 0.36 -0.01 0.12 0.11 0.16 2.07 0.19* 36.1* 0.07 0.04 -0.02 0.02 0.11 -0.01 0.04 0.03 0.10* 1.32* , 0.10 . 15.9 -0.03 -0.01 -0.05 -0.003 -0.07 -0.04 -0.001 -0.02 0.01 0.25 0.27 56.3 0.17 0.10 0.01 0.04 0.29 0.02 ' 0.09 0.09 0.19 2.38 0.19* 39.4* 0.10* 0.05* -0.02 0.02* 0.12 -0.01 0.05" 0.04 0.08* 1.13* 0.12 20.7 0.02 0.01 -0705 0.003 -0.02 -0.04 0.005 -0.01 0.01 0.27 0.26 58.2 0.19 0.10 0.001 0.04 0.27 0.02 0.09 0.09 0.16 1.99 the effects of conversion from gross serum to serum lipid concentration of the xenobiotic. The most consistent finding in Tables 6 -9 was the strong, positive associations of log serum triglycerides and cholesterol with the log gross serum level of PCBs and DDE which has been found by others (13,15,21,22). The association disappeared when the xenobiotic level was expressed as the concentration in serum lipids, or as the PCB exposure estimate. This disappearance was associated with a 70 to 130% decline in the value of (3,. This finding was exhibited in all regressions whether against log LPCB, HPCB, or DDE, and whether for 1976 or 1979. A similar, but less dramatic change was exhibited by regressions against measures of the he\ patic enzymes (log GGTP, log SGPT and log alkaline phos phatase). These parameters frequently showed signifi cant associations with the log gross serum levels of PCB or DDE, but the (3, values dropped 20 to 50% upon conversion to serum lipid levels. The only assov ^ eintions then remaining significant in the final backward p regressions were those of 1976 log GGTP with log CB, log HPCB, and EXE. None of the other clinical parameters showed consistent declines of this magni tude in fV A second consistent finding was strong negative as sociations of PCBs, but not DDE, with the two meas ures of bilirubin, direct bilirubin and log total bilirubin, in 1976. These associations were seen for all measures of log PCB (LPCB, HPCB and EXE) in almost all of the regressions. Only two associations remained in 1979 (HPCB, EXE), when the PCB levels were lower. A third set of findings consisted of associations be tween the hematological variables and the serum lipid PCBs. In 1976 significant negative associations were found between RBCs and both log serum lipid LPCB and HPCB, and between the monocytes and both log serum lipid HPCB and the exposure estimate. There were positive associations between lymphocytes and log serum lipid LPCB, and between WBCs and log serum lipid HPCB. Similar associations with the log gross serum DDE were found for the lymphocytes, mono cytes, and WBCs. In separate step regression studies log total bilirubin was selected in place of log serum lipid PCB values in the cases of RBCs and the lympho- i.M ir a i. a m ) in. ,km . nsnist.s i.\ I . ... ir>>, ii r : i:< ITT Table 7H. Kerreanion and correlalion coefficients for 197h and 197 clinical variable* ahnwing at least one niirnificant association with a measure of serum HI*CB (1254). as determined by three regression procedures: log serum lipid Ill'CHs. Dependent variable* 1976 Log triglycerides Cholesterol Log SGPT Log GGTP Log alk. phos. Log total bilirubin Direct bilirubin Log total protein Log albumin Log BUN Creatinine Log B/C ratio Iron RBC WBC Lymphocytes % Monocytes Simple repression Confia, limits 3, 5 95 Log scrum lipid HPC13* Multiple repression Confid. limits 3, 5 95 -0.02 2.68 -0.01 0.06 0.04 -0.08-0.02 -0.005 -0.01 0.03 -0.05 0.05* -15.2* -0.13 0.97* 0.30 -0.87 -0.10 -17.8 -0.08 -0.04 -0.02 -0.15 -0.05 -0.02 -0.02 -0.01 -0.1 0.002 -28.1 -0.28 0.35 -0.02 -IrTT 0.06 23.2 0.06 0.16 0.09 -0.01 0.005 0.01 0.002 0.07 0.03 0.09 -2.21 0.03 1.59 0.62 0.02 -0.07 -11.1 0.03 0.09 0.02 -0.12-0.04* -0.01 -0.01 0.003 -0.10* 0.04 -13.1 -0.12 1.20* 0.58* -1.45' -0.17 -35.3 -0.04 -0.02 -0.04 -0.21 -0.07 -0.03 -0.03 -0.05 -0.18 -0.02 -29.9 -0.28 0.39 0.14 -2.64 0.03 13.1 0.11 0.21 0.09 -0.03 -0.001 0.01 0.002 0.05 -0.01 0.09 3.77 0.03 101 1.02 -0.60 Step repression Confid. limits 3. 5 95 -0.08 -11.5 0.03 0.11* 0.03 -0.11* -0.03* -0.015* -0.014* 0.003 -0.09* 0.04 -13.2 -0.15* 0.98* 0.36 -1.61* -0.17 -35.6 -0.04 0.002 -0.03 -0.19 -0.07 -0.03 -0.028 -0.05 -0.18 -0.01 -28.2 -0.29 0.26 -0.02 -2.63 0.02 12.6 0.10 0.21 0.09 -0.02 -0.002 -0.003 -0.001 0.05 -0.01 0.09 1.74 -0.02 1.69 0.74 -0.59 1979 Log triglycerides Cholesterol Log GGTP Log alk. phos. Direct bilirubin Log blood glucose Globulin Log A/G ratio Log BUN Log B/C ratio Monocytes % Monocytes See Table 6 footnotes. 0.11* 25.3* 0.09 0.05* -0.02 0.04* 0.14 -0.02 0.08* 0.05* 0.10* 1.48* 0.03 6.72 -0.01 0.003 -0.05 0.02 -0.02 -0.04 0.03 0.01 0.03 0.55 0.20 43.9 0.19 0.10 0.004 0.06 0.31 0.01 0.12 0.10 0.18 2.41 0.04 2.06 0.02 0.03 -0.02 0.02 -0.004 0.001 0.04 0.02 0.12* 1.56* cytes. Similar selections, reflecting similar co lineari ties, occurred in the cases of phosphate (with LPCB) and creatinine (with HPCB). Smoking was a significant confounding variable in the cases of direct bilirubin, WBCs, and phosphate. In 1979 strong positive associations were found be tween all measures of exposure to PCB (but not DDE) and the absolute and relative monocyte levels. The bil irubins did not replace PCBs in the step regressions, but smoking was a significant confounder. The positive associations of cholesterol and serum sodium with the relative exposure estimate in 1976 and the negative as sociation of log serum lipid DDE with serum iron in 1979 appeared to be singular findings. The 1530 regressions used to generate the data shown in Tables 6-9 showed at least one significant association (5% level) with one measure of exposure in one of the regressions for 26 of the 42 clinical variables. The in cidence of significant associations with the clinical var iables was positively influenced by the number of independent variables included in the regressions. The independent variables used in Tables 6 -9 regressions -0.06 -20.1 -0.09 -0.03 -0.06 -0.01 -0.19 -0.03 -0.01 -0.03 0.03 0.44 0.13 24.2 0.13 0.08 0.01 0.04 0.18 0.04 0.08 0.08 0.22 2.69 0.04 119 0.02 0.03 -0.03* 0.02 -0.004 0.001 0.04 0.02 0.11* 1.42* -0.06 -19.6 -0.09 -0.01 -0.06 -0.005 -0.19 -0.03 -0.003 -0.03 0.03 0.48 0.13 24.0 0.13 0.08 -0.002 0.04 0.18 0.04 0.09 0.08 0.19 136 r all showed multiple associations with the clinical labo ratory parameters. This group of independent variables did not, however, include measures of alcohol con sumption or socioeconomic status, both of which have been used in other studies (21,22). The data available indicates that there can be important colinearities among the dependent variables. Because of the uncertainties presented by chance associations, unused or unidenti fied confounders, and colinearities among variables we have focussed our interpretations upon those associa tions that were seen consistently in the step regressions against log serum lipid PCBs, but not in the corres ponding regressions against log DDE. The mean 3, coefficients of Tables 6 -9 express the best estimates of the dependence of the clinical labo ratory variables on log PCB, and their 95% confidence limits indicate the uncertainties of these estimates. For associations observed to be statistically significant we have multiplied the lower and upper {3, confidence limit values by the observed range of log PCBs in order to determine whether the product resulted in values for the laboratory variables outside the normal expected 178 LAWTON ETAL. Table 7C. Regression and correlation coefficient* for 1976 and 1979 clinical variables showing at least one significant association with a measure of serum HPCB (1254), as determined by three regression procedures: independent variables. Independent variables* Partial r Log Dependent variable Age Sex SPG BMI DS SC JS FNF (HPCB)S1. 1976 Log triglycerides 0.20 -0.22 -0.15 0.28 0.26 X Cholesterol 0.35 0.20 X Log SGPT -0.29 0.37 X Log GGTP -0.18 0.19 X 0.28 0.16 Log alk. phoB. 0.14 X Log total bilirubin 0.21 -0.17 X 0.15 -0.20 Direct bilirubin -0.17 -0.17 X 0.19 -0.17 Log total protein -0.16 -0.17 X -0.19 Log albumin -0.27 -0.16 X -0.17 Log BUN 0.16 0.34 0.19 X Creatinine 0.19 -0.40 0.23 X -0.17 Log B/C ratio 0.17 X Iron 0.20 -0.15 X RBC -0.59 X 0.18 WBC 0.29 X 0.21 Lymphocytes X 0.16 % Monocytes X -0.27 -0.27 1979 Log triglycerides Cholesterol Log GGTP . Log alk. phos. Direct bilirubin Log blood glucose Globulin Log A/G ratio Log BUN Log B/C ratio Monocytes % Monocytes 0.21 0.35 0.18 0.29 -0.21 0.20 0.26 -0.28 -0.16 0.40 X 0.19 . X 0.19 0.23 X 0.14 X -0.15 X 0.16 0.38 X 0.27 0.15 X -0.24 -0.16 X 0.34 0.23 X 0.30 X 0.19 -0.21 X -0.24 X -0.14 0.21 0.23 I ranges. For these criteria, the only variables having 6.2% (mean = 3.5%). Inspection of the histogram along statistical significance outside the normal ranges were the monocyte axis indicates a range of percent mono the 1976 percent lymphocytes and the 1979 percent mon cytes from 5.5 to 15.5% (mean = 10.5%) rather than ocytes. For all other parameters the observed associ the normal range of 0 to 8% (Table 1). ations with log PCB levels reflected variations within the normal laboratory ranges. Figures 3 and 4 show partial residual plots for two D iscussion laboratory variables, log total bilirubin in 1976 and the percent monocytes in 1979, as calculated at the end of PCB Levels in the Study Population the stepwise regression, plotted against log serum lipid This investigation has shown that certain former op LPCB. They represent plots of a predictor variable erations in capacitor manufacturing presented high ex against the dependent variable with the "effects" of all posures to PCBs, particularly during the 1954-1977 other predictor variables statistically removed {33). For the log total bilirubin data of Figure 3, the partial r was period when the more volatile lower PCBs (Aroclor 1242 and 1016) were in extensive use. During this period PCB -0.22 (p = 0.01) and about -0.09 log units/decade of air levels in the affected working areas were probably LPCB (Table 6). For 2.2 decades of LPCB, the rang^ at least the 690 p.g/m observed in 1975, accompanied by of the p! product was -0.07 to -0.33 (mean = -10.2).1 extensive dermal contact as well. When controlled for confounders, all study data (Fig. 3) fell between 0.1 and 1.0 mg/dL (mean = 0.3 mg/dL), which is within the normal range (Table 1). \ As shown in Figure 4 the 1979 percent monocyte data In our study population, selected so as to include those working in or near the plant PCB exposure zone, the mean serum lipid LPCB level in 1976 (Table 3) was estimated to be 93 ppm (5-95% range, 15-560 ppm), v were more tightly grouped (partial r = 0.28), and the p, estimate was 1.38%/LPCB decade. For the 2.5 dec ade range of LPCB, the range of pi products corre adjusted for a factor of 10 analytical error. This serum lipid value corresponded to a mean body burden of 2.0 g (0.3-12.3 g) for the population mean body weight of sponded to an increment in percent monocytes of 1.8 to 77 kg (22 kg of fat). These retained LPCBs were com- ttmciiKXh a s h h k v a : / '*uca:. /7 ,\t /.v .- /.v ( HT. iM W tHtKKRS 179 Table 8A. Keffretuion and correlation coefficient for 137 and 1379 clinical variable allowing at least one Biennicant^aociation with a measure o f serum p. p'-l)DE. o determined by three regression procedures: log gros* serum DDE. Dependent variable' 1976 Log triglycerides Cholesterol Log GGTP Log alk. phos. Log total protein WBC Lymphocytes % Lymphocytes % Monocytes * 2.13 251.3 1.09 1.42 0.86 7.02 2.33 33.8 4.38 P. 0.08" 12.7" 0.04" 0.04" 0.002 0.33" 0.23" 1.42 -0.36" Log gross serum DDE Multiple regression Confid. limiti Confid. limits 5 95 i . 5 95 0.05 5.05 0.002 0.02 -0.002 0.08 0.11 -0.11 -0.70 0.11 20.4 0.08 0.06 0.006 0.57 0.34 2.95 -0.02 0.07" 8.37 0.02 0.04" 0.009 0.41* 0.30" 2.04" -0.40 0.03 -0.13 -0.02 0.02 -0,001 0.12 0.16 0.17 -0.80 0.10 16.9 0.06 0.07 0.009 0.70 0.44 3.91 0.004 Step regression Confid. Pi 5 95 0.08* 10.1" 0.03 0.04" 0.005" 0.35* 0.26* 1.37 -.46* 0.05 2.00 -0.01 0.02 0.001 0.03 0.13 -0.38 -0.83 0.11 18.2 0.07 0.07 0.009 0.62 0.40 3.11 -0.08 1979 Log triglycerides Cholesterol Log alk. phos. Log total protein Log albumin Globulin Log A/G ratio Serum Ca Serum Iron "See Table 6 footnotes. 2.16 238.2 1.47 0.86 0.64 2.92 0.18 9.67 108.6 0.31* 48.6* 0.09* 0.01 -0.02* 0.30* -0.06* -0.02 -10.3 0.21 25.8 0.04 -0.004 -0.03 0.10 -0.09 -0.22 -24.3 0.41 71.5 0.15 0.02 -0.01 .50 -0.02 0.18 3.68 0.28* 32.6* 0.09" 0.01* -0.006 0.23* -0.03 -0.10 -10.0 0.18 7.56 0.03 0.001 -0.02 0.02 -0.07 -0.33 -25.5 0.38 57.6 0.15 0.03 0.01 0.44 0.01 0.13 5.50 0.30* 30.4* 0.10* 0.01 -0.006 0.24* -0.03 -0.08 -8.81 0.21 6.07 0.04 -0.001 -0.02 0.04 -0.07 -0.29 -22.9 0.39 54.7 0.16 0.03 0.01 0.43 0.01 0.13 15.30 Table 8B. Regression and correlation coefficients for 1976 and 1979 clinical variables bowing at least o n e ai^ifican t association with a measure of serum p, p '-DDE, as determined by three regression procedures: log serum lipid DDE. Dependent variable" 1976 Log triglycerides Cholesterol Log GGTP Log alk. phos. Log total protein WBC Lymphocytes % Lymphocytes % Monocytes Simple regression Confid. limits 6. 5 95 Multiple regression Confid. limits . 5 95 i t Step regression Confid. limits 3. 5 95 1979 Log triglycerides Cholesterol Log alk. phos. Log total protein Log albumin Globulin Log A/G ratio Serum Ca Serum Iron 0.08 * -3.10 0.07* -0.003 -0.02* 0.15 -0.04 -0.17 -19.7" "See Table 6 footnotes. -0.04 -29.7 0.002 -0.02 -0.04 -0.08 -0.08 -0.39 -35.0 0.20 23.5 0.13 0.01 -0.004 0.38 0.002 0.06 -4.43 0.06 -20.8 0.07 0.001 -0.01 0.07 -0.01 -0.24" -19.6V -0.06 -47.9 -0.002 -0.01 -0.03 -0.16 -0.06 -0.48 -36.0 0.18 6.34 ' 0.13 0.02 0.008 0.30 0.03 -0.001 -3,20 0.07 -21.6 0.08* 0.001 -0.01 0.08 -0.02 -0.22 -17.5* -0.04 -47.9 0.01 -0.01 -0.03 -0.15 -0.06 -0.45 -33.1 0.19 4.76 0.14 0.02 0.007 0.30 0.03 0.008 -2.00 posed almost entirely of PCB isomers with gas chronatographic peaks having retention times (relative to v ^DDE = 100) of 37, 71 and 84 (27M ) , the other LPCB components of Aroclors 1016 and 1242 being much more readily metabolized in the human. Regression studies of serum lipid LPCB values with length of service in dicated no significant statistical association, so! that steady state levels appeared to have been reached within the first few years of employment. Thus, the levels of retained LPCBs observed in 1976 could have been rep- IH) LAWTON ETAL. Table 8C. Regrwiion and correlation coefficient, for 1976 and 1979 clinical variable, showing at leaat one significant aaaociatlon with ) a measure of aenim p. p'-DDE. aa determined by three regression procedure.: independent vanabtea. Dependent variable____ Age____ Sex 1976 Log triglycerides Cholesterol Log GGTP Log alk. phos. Log total protein WBC Lymphocytes % lymphocytes % Monocytes 0.31 -0.23 -0.19 SPG Independent variables Partial r BMI DS SC 0.29 0.27 0.20 0.17 -0.16 -0.17 0.25 0.16 Loff JS_____FNF (DDE) 0.3S ; 0.19 0.26 0.31 0.16 0.20 0.33 -0.25 -0.21 1979 Log triglycerides Cholesterol Log alk. phos. Log total protein Log albumin Globulin Log A/G ratio Serum Ca Serum Iron 0.21 0.35 -0.14 -0.21 -0.21 "See Tibie 6 footnotes. -0.20 0.40 0.19 0.14 0.23 0.27 -0.24 -0.19 -0.15 0.15 -0.16 0.15 Table 9. Regression coefficients for 1976 and 1979 clinical variables showing at least one significant association with a estimate of relative exposure to PCB in 1976 (EXE), as determined by three regression procedures/ Dependent variable* 1976 Cholesterol Log GGTP Log total bilirubin Direct bilirubin Log blood glucose Log A/G ratio Serum Ns Serum iron Lymphocytes Monocytes % Monocytes N 193 192 194 193 193 191 192 192 153 153 153 Simple regression Multiple regression Step regression Confid. limits Confid. limits Confid. limits Partial z P. 5 95 0. 5 95 Pi 5 95 (EXE) 251.3 1.09 -0.34 0.15 2.01 0.21 139.5 112.1 2.33 0.31 4.38 7.86 0.03 -0.04* -0.01 -0.01* 0.02* -0.10 -6.35* 0.14 -0.01 -0.34 -1.33 -0.QI -0.08 -0.02 -0.02 0.001 -0.55 -12.3 -0.01 -0.05 -0.74 17.0 0.10 -0.01 0.002 -0.004 0.03 0.34 -0.36 0.28 0.02 0.06 9.62 \ -0.16 0.06* 0.01 -0.05* -0.09 -0.02* -0.03 -0.01 -0.02 0.01 -0.005 -o.& r -1.02 -3.89 -10.8 0.19- 0.02 -0.01 -0.04 -0.35 -0.82 19.4 0.10 -0.02 -0.001 0.003 0.03 -0.06 3.00 0.36 0.03 0.11 10.50.06-0.05-0.02* -0.01 0.01 -0.53-4.23 0.15 --0.01 -0.46- 1.26 0.01 -0.08 -0.03 -0.02 -0.004 -0.98 -10.8 -0.01 -0.04 -0.90 19.7 0.10 -0.01 -0.001 0.001 0.03 -0.08 2.30 0.31 0.03 -0.02 0.18 0.20 -0.20 -0.17 -0.18 1979 Cholesterol Log GGTP Log total bilirubin Direct bilirubin Log blood glucose Log A/G ratio Na Iron Lymphocytes Monocytes % Monocytes 174 174 174 174 174 174 . 174 172 175 175 175 238.5 1.20 -0.22 0.14 2.00 0.18 139.4 108.5 2.09 0.48 6.9 "See footnotes to Tibie 6. 4.72 0.01 -0.02 -0.01 -0.004 -0.003 -0.05 -0.96 -0.03 0.04* 0.58 -4.09 -0.03 -0.05 -0.02 -0.01 -0.02 -0.46 -6.17 -0.14 0.001 0.14 13.5 0.06 0.01 0.01 0.01 0.01 0.37 4t25 0.08 0.08 1.03 4.83 0.02 -0.03- -0.01 -0.001 -0.006 -0.02 -0.78 -0.03 0.04- 0.61" -3.91 -0.02 -0.06 -0.03 -0.01 -0.02 -0.43 -6.17 -0.16 0.004 0.17 13.6 0.07 -0.001 0.003 0.01 0.07 0.40 4.62 0.09 0.08 1.05 4.94 0.02 -0.03* -0.01 -0.002 -0.005 -0.02 -0.78 -0.03 0.050.64" -3.43 -0.02 -0.06 -0.03 -0.01 -0.02 -0.43 -6.17 -0.14 0.01 0.20 13.3 0.07 -0.004 0.001 0.01. 0.011 0.39 ; 4.62 , 0.09 0.08 1.07 -0.18 0.18 0.22 >;resentative of those present throughout the Aroclor 1242 and 1016 periods of use. The mean serum HPCB level in 1976 was estimated to be 8 ppm (2-34 ppm), corresponding to a mean body burden of 0.2 g (0.04-0.7 g). These HPCBs were com posed primarily of penta- and hexachlorobiphenyl iso mers (27,54,55) resulting either from exposures to Aroclor 1242 (which contained a few percent of the lower IM I0 . Vt iOlKt It 0 (2221 o.x 0.1 -0.0 1 - 0.1 *1 - 0.2 11J - 0 . 3 1130I - 0 . 4 11 - 0 . 3 i1r1 -0.A 10 117 - 0.7 I - 0.1 s X -0.2 X - 1.0 -l.l # - 1.x -1.3 11 11 1 1 11 11 1 - 1-121 212 1 I11111X111212111212111I111212t i131321233111111111XX12112211I11111I12111 1III111111111U111 21 11 11 1 X 11 11 0 . 12 C0 1 . 192 10T4L -1.0 -0.0 1.0 1.0 -0 .3 0.1 1.3 ** Citi <4.. 1 3 11 * T X tom 1 13 13 3 4 4 11 13 11 2 X IMM 3 IS X LOG SERUM UPI0 LPCS Figure 3. Partial residuals for log total bilirubin in 1976 following y final step regression (Table 6) vs. log serum lipid LPCBs. Histo grams are shown along each axis. The number of values in the cells are given by the numerals. | HPCBs), or from earlier exposures to Aroclor 1254. I Their levels were significantly associated with length of ! service, so that the upper 5% of HPCB levels observed j in 1976 (long service, directly exposed workers) may be representative of the exposed population in 1954, at the end of the period when Aroclor 1254 was the major dielectric fluid used. The Aroclor 1242 values reported to us by the analyst I were approximately 4 times the calculated LPCB values | (Table 3). We have shown elsewhere that the minimum j initial concentrations of Aroclors 1016 plus 1242 reI quired to account for the most persistent LPCB peaks I were about 7.5 times the residual LPCB levels (27). i Thus, the minimal LPCB uptake by the study popula- jI tion may be estimated as15 g (2.5-92 g). This estimated uptake is a minimum since even the most persistent I LPCB peaks have a finite clearance rate (unpublished observations). The longer service employees could have v absorbed several times this minimal PCB uptake, but ' until better data on PCB clearance becomes available, actual cumulative LPCB uptakes cannot be estimated more precisely. ,, A The mean LPCB body burden in 1979 (29 months after ^continuance of major exposure) was estimated as 0.4 v s (0.7-2.0 g), indicating a clearance of 80%. The serum lipid HPCB means in 1976 and 1979 were within the (IU I* Viali IM M in v i 21. 14.8 ll.B 12.0 1 1 .0 20.0 11.0 U.Q ;i U1 oaOoz ! *ii r 1i1l m il ii i VUJ) 2 2 (E 1 J < S 2 $Q ir.o u.o 11.0 1 4 .0 11.0 12.0 11.2 12.0 2.0 i.n r.c 4.8 1.2 4J 1.2 2.2 9 1.2 1IIIIIIlVi. i ii11iiti1i1iiiii11111121111 1 1 1212111 11 I l I I 11 1 .1 1 I I 1111 2 1 1 II II 2 1 I I I 2211 1 I 1I t i t i 1111 1 i l in 1 lim 1 ill 1 1 i 1 HI 1 1 111 11 1 1i 1 1t 1 1.1 -1.2 -2.1 --2.2 1.1 I. 1.1 2.0 1.1 1 2 12 14 10 * i i> r i 2 1 I 11 I I 1 2 1 I I I I 12 1 1.0 LOG SERUM LIPIO LPCS F igure 4. Partial residuals for % monocytes in 1979 following final step degression (Table 6) vs. log serum lipid LPCBs. Histograms are shqwn along each axis. The number of values in the cells are given by the numerals. error band of the analytical technique (27) and a con clusion of significant clearance is probably unwarranted. Chen {35) has reported clearance rates for various pentachlorobiphenyl isomers in Taiwanese Yusho patients that appear to be at least 10-fold greater than we ob served in capacitor workers (27); however, such clear ance might be related to P-448 oxidase induction by PCDFs in the Yusho patients. Although these estimates of PCB fat levels and uptakes are only approximate, they serve to define the dose ranges within which the observed responses can be ascribed. PCB Effects on Serum Lipids and Serum Enzymes Previous investigators have reported statistical as sociations between log gross serum levels of PCBs and the serum levels of triglycerides, total cholesterol and certain serum enzymes. Our findings were in general agreement with such reports. We found significant as sociations of log serum triglycerides and total choles- 182 LAWTON ET AL. teroi with every measure of log gross serum PCBs and DDE in the final step regressions. Similar associations were found for log GGTP, and in some cases for log SGPT, with log gross serum PCBs, and for log alkaline phosphatase with log gross serum DDE. It is now evident, however, that such associations arise because the level of PCBs in the serum is itself determined by that of the serum lipids (23,2L). When the PCB concentrations were expressed as levels in serum lipids, which would be expected to correlate with the pharmacological activities (24,25), all the associa tions between serum lipids or enzymes and log gross serum LPCB, HPCB or DDE disappeared except for those between log GGTP and log PCBs, and between log alkaline phosphatase and log DDE. Similarly, Chase et al. {13), who reported associations with gross serum PCBs, found no significant correlation between either serum triglycerides or SGOT and the adipose tissue fat biopsy PCB levels. Although elevated levels of serum triglycerides, total cholesterol and SGPT were found in our study population (Table 3), serum GGTP levels were generally in the normal range, andthe regression stud ies indicated stronger associations with obesity (body mass index). PCB Effects on Microsomal Enzymes Elevated serum GGTP levels have been prominently mentioned in the literature as an index of microsomal enzyme induction in the human {36-10). Recent studies (41), however, have in part disputed this conclusion. In their review of environmental hepatic injury, Popper et al. U2) found only meager evidence of substantiated chronic hepatic effects. They concluded that conven tional liver function tests were of limited value in de tecting hepatic abnormalities and that the evidence so far indicates that too many other processes influence GGTP activity in man to make it a useful index of in duction. The clinical implications of elevated GGTP lev els in man are review-ed by Guzelian elsewhere in this symposium. However, the interpretation of the relationship be tween serum GGTP and serum lipid PCBs as evidence of microsomal enzyme induction is strengthened by the finding of strong inverse statistical associations between both direct and total bilirubin levels and serum lipid PCBs in 1976 during the period of active PCB exposure. A weak residual association between serum lipid HPCB and direct bilirubin (partial r = 0.14) was found in 1979 when the mean total bilirubin (mostly unconjugated) had risen from 0.46 to 0.61 mg/dL without change ih the mean direct bilirubin (conjugated form) (Table 1). Since the original observations of Yaffe et al. (43) on the therapeutic use of phnobarbital to enhance glucuronide-conjugating capacity in infant hyperbilirubi nemia, a substantial literature has developed linking drugs (36,44) and other xenobiotics (45,46) to induction of microsomal enzymes, including enhanced glucuronyl transferase, resulting in increased conjugation and ac celerated hepatic elimination of bilirubin. [>Glucaric arid is the main excretory product of the glucuronic acid conjugation pathway and its urinary level also has been advocated as a measure of enzyme induction (47-50). Drugs and chemicals inducing microsomal enzymes are often metabolized by the induced oxidases (32), and it appears likely that such induction would accelerate clearance of PCBs as well as bilirubin (49,51). Hirayama et ai. (52) reported serum total bilirubin levels of 0.87 * 0.33 mg/dL in 257 normal controls and 0.48 * 0.26 mg/dL in 121 Yusho patients, indicating a 46% decrease in the latter. However, the observations were made at a time when the PCB levels had declined to background values, so that the pharmacologically ac tive agents present were probably the PCDFs. Induc tion of glucuronyl transferase by Aroclor 1242 and 1016 has been reported in rats (16,53), as has also that of demethylases (16,53), aniline hydroxylase, (53) and cy tochrome P^450 (16,53), but not that of the P-448 cy tochrome (16). Burse et al. (51) have shown that rats fed 100 ppm of Aroclor 1242 or 1016 (3.9-6.6 and 3.5-6.9 mg kg/ day) reach equilibrium adipose tissue fat concentrations of 35 to 143 or 69 to 236 ppm, respectively, at 6 months, levels in the range of those observed in this study (Table 3). Similar levels in rats were also reported by Goldstein et al. (53), who found a 3-and 12-fold increase in liver glucuronyl transferase activity with Aroclors 1016 and 1242, respectively. The most compelling evidence for microsomal enzyme induction in man is a response to drug administration (42). In 1977, Alvares et al. (16) reported antipyrene half-times reduced 31% over matched controls in five capacitor workers with PCB exposures of 4 to 16 years drawn from the same plant population as the present study. The effect was statistically significant (p<0.005) and the half-times were significantly less than the av erage values of much larger groups of normal healthy subjects. The statistical associations of the present study, taken together with the findings of Alvares et al., ap pear to indicate induction of the microsomal enzymes GGTP, glucuronyl transferase, and P-450 oxidase in the study population during active exposure in 1976. The enzyme inductions appeared to be smaller than those in rats carrying similar levels of lipid LPCBs and to have become virtually undetectable by 1979. PCB Effects on Hematological Parameters The findings of significant associations between log serum lipid PCBs and some hematolotical variables was unexpected because Maroni et al. (14) reported normal hematology including WBCs and differentials and no abnormal reports in occupationally exposed PCB work ers have been mentioned by others (11-13,15,17). In 1976 the clinical reports (Table 2) suggested a slight decline in PMNs with some increases in lymphocytes, monocytes and eosinophils, and normal values for RBCs, hemoglobins and hematocrits. r* Kim `ifkm: \xn h i :'.:a t o u k ;h `m / . \ > is / . v r . i ; \ n v ` >. l.'KKKS ISi Interpretations of these daLa.are difficult because of the shared associations with DDE and the colinearities ,, y with either total or direct bilirubin or both. For the associations with log total bilirubin, the partial r values were: RBC, +0.19; lymphocytes -0.22; and creatinine, .0.24; for direct bilirubin, they were: phosphate, -0.18; and creatinine, 0.23. We tend to the interpretation that these variables are correlated with the enzyme induc tion process. In 1979, although both the monocytes and eosinophils were only marginally elevated in the study population based on clinical reports (Tables 1 and 2), the statistical association of serum PCB levels and the monocyte counts was strong. Because of the absence of a monocytosis in 1976, its lack of association with either DDE or total or direct bilirubin in 1979, and its absence among retirees in 1979, we hypothesized that the effect might be related to a new exposure in the workplace, perhaps to the PCBsubstitute dielectric. However, we recently studied 54 workers with no prior PCB exposure who were cur rently exposed to the substitute and found no monocyte elevations. Although a number of solvents were in use in the environment in 1979, the monocytosis was mild compared to that observed by Minot and Smith (55) in the case of tetrachlorethane exposure, where the mon ocyte levels were elevated 20 to 40%. The effect of the change'ft'pm the manual (1976) to automated differential analysis (1979) is unknown. We have also hypothesized that the monocytosis observed in 1979 might be related to the increased incidence of chronic obstructive lung disease, heavy smoking or concurrent respiratory infections. We are currently studying the associations be tween the spirometric variables (FVC, FEV, FEW FVC) and the hematological parameters in the study population and are restudying the entire population, which should cast further light on any such relationships. Clinical Consequences of PCB Exposure These findings differ from those of most other recent studies of environmentally or occupationally exposed populations in offering evidence of microsomal enzyme induction, which is consistent with the extensive LPCB clearance also observed in these highly exposed work ers. The evidence is statistical in nature and the asso ciated parameters remained within their normal clinical ranges, except possibly for the lymphocytes in 1976 and the monocytes in 1979. The effects were therefore phys iological rather than pathological and appeared to sub side following the cessation of PCB use. The long-tern^ consequences of such enzyme induction are obscure US). > The present clinical health of the study population appears to meet community medical standards. Despite the prevalence of cardiovascular risk factors (obesity, elevated cholesterol levels, smoking, etc.) the mortality xperience has been normal. The paucity of clinical ab(er_/_!ormalities is consistent with laboratory studies in rats involving chronic administration of Aroclor 1242 and lUIli, producing tissue fat levels comparable to the serum lipid PCB levels observed here. Such studies (54) have shown no overt symptoms of clinical poisoning and no pathological evidence of liver damage. The production of serious liver damage in rats (55) requires the use of a largely nonmetabolizable PCB, such as Aroclor 1260, at a cumulative dose of 2.5 g/kg, which would result in tissue lipid HPCB levels above 10,000 ppm, as compared to those in our study population of 8 ppm in 1976 and 5 to 6 ppm in 1979. It is evident, therefore, why the se rious hepatic effects in the test rat could not be observed in man. The author* are grateful for the technical assistance of Sheila Grady in data collation, Barbara Sack, Ranald Chapman and Barbara Moreen in statistical analysis support, and Helen Walton in manuscript prep aration. We particularly appreciate the advice and counsel of Gerald Hahn, Carolyn Morgan and John Bergeron in statistics. REFERENCES 1. Versar, Inc. PCBs in the United States: Industrial Use and En vironmental D istribution. R eport under EPA Contract No. 68013259. U.S. Dept, of Commerce, NTIS PB 252012. Feb. 25, 1976. 2. Drinker, C. K., Warren, M. F ., and Bennett, G. A. The problems of possible systemic effects from certain chlorinated hydrocar bons. J. Ind. Hyg. Toxicol. 19: 283-299 (1937). 3. Treon, J . F., Cleveland, F. P., Cappel, J. W., and Atchley, R. W, The toxicity of the vapors of Aroclor 1242 and 1254. Am. Ind. Hyg. Assoc. Quart. 17: 204-213 (1956). 4. Elkins, H. E. Chemistry of Industrial Toxicology. Wiley, New York, 1950, p. 149. 5. Clark, F. M. Insulating Materials for Design and Engineering Practice. Wiley, New York, 1962, p. 194. 6. Masuda, Y., Kuroki, H., Yamaryo, T., Haraguchi, K., Kuratsune, M., and Hsu, S. T. Comparison of causal agents in Taiwan and ^ukuoka PCB poisoning. Chemosphere 11: 199-206 (1982). 7. Buser, H. R., Bosshardt, H. P., and Rappe, C. Formation of polychlorinated dibenzofurans (PCDFs) from the pyrolysis of PCBs. Chemosphere 1: 109-119 (1978). 8. Hasegawa, H.t Sato, M., and Tsurata, H. PCB concentration in the blood of workers handling PCB. Occup. Health 10: 50-55 (1972). 9. Kitamura, M., Tsukamoto, T., Sumino, K.. Hayakawa, K., Shib iu . T., and Hirano, I. The PCB levels in the blood of workers employed in a condenser factory. Japan. J. Ind. Health 47: 354355 (1973). 10. Hara, I., Herada, H., Kimura, S., Endo, T., and Kawamo, K. Follow-up health examination in an electrical condenser factory after cessation of PCB usage (1st Report). Japan. J. Ind. Health 16: 365-366(1974). 11. Karppanen, E ., and Kolho, L. The concentration of PCB in human blood and adipose tissue in three different research groups. In: PCB Conference II. National Swedish Environmental Protection Board Publications, Stockholm, 1973, pp. 124-128. 12. Ouw, K. H., Simpson, G. R., and Siyali, D. S. The use and health effects of Aroclor 1242, a polychlorinated biphenyl, in an electrical industry. Arch. Environ. Health 31: 189-194 (1976). 13. Chase, K. H., Wong, O., Thomas, D., Bemey, B. W., and Simon, R. K. Clinical and metabolic abnormalities associated with oc cupational exposure to polychlorinated biphenyls (PCBs). J. Oc cup. Med. 24: 109-114 (1982). 14. Maroni, M., Columbi, A., Arbosti, G., Cantoni, S., and Foa, V. Occupational exposure to polychlorinated biphenyls in electrical worker*. II. Health Effects. Brit. J . Ind. Med- 38: 55-60 (1981). 15. Smith, A. B., Schioemer, J ., Lowry, L. K., Smallwood, A. W.( Ligo, R. N., Tanaka, S., Stringer, W,, Jones, M,, Hervin, R., 184 LAWTON E T A L . and Glueck, C. J. Metabolic and health consequence? of occupational exposure to polychlorinated biphenyls (PCBs). Brit. J. Ind. Med. 39:361-369(1982). 16. Alvares. A. P.. Fischbein, A.. Andereon. K. E., and Kappas, A. Alterations in drug metabolism in workers exposed to poly chlor inated biphenyls. Clin. Pharmacol. Therap. 22: 140-146 (1977). 17. Fischbein, A.. Wolff, M. a., Ltlis, R.. Thornton, J., and SelikofT, 1. J. Clinical findings among PCR-exposed capacitor manufactur ing workers. Ann. N.Y. Acad. Sci. 320: 70C1-715 (1979). 18. Warshaw, R.. Fischbein, A., Thornton, J., Miller, A., and Sellkoff, I, J. Decrease in vital capacity in PC 13-exposed workers in a capacitor manufacturing facility. Ann. N.Y. Acad. Sci. 320:277283 (1979). 19. Brown, D. P., and Jones, M. Mortality and industrial hygiene study of workers exposed to polychlorinated biphenyls. Arch. Environ. Health 36: 120-123 (1981). 20. Humphrey, H. E. B. Evaluation of humans exposed to halogen- ated biphenyls. Am. Chem. Soc. Div. Environ. Chem. Preprints 20 (No. 2): 272 (1980). 21. Baker, E. L.t J r M Landrigan, P. J., Gluck. C. J ., Zack. M. M. J r., Liddle, J . A., Burse, V. W., Houseworth, W. J., and Need ham, L. L. Metabolic consequences of exposure to polychlorinated biphenyls in sewage sludge. Am. J. Epidemiol. 112:553-663 (1980). 22. Kreiss, K., Zack, M. M., Kimbrough, R. D., Needham, L. L., Smrek, A. L., and Jones, B. T. Association of blood pressure and polychlorinated biphenv! levels. J. Am-Med. Assoc. 245: 25052509 (1981). 23. Morgan, D. P., and Roan, C. C. Chlorinated hydrocarbon pes ticide residue in human tissues. Arch. Environ. Health 20: 452457, 1970. 24. Brown, J. F ., Jr. Polychlorinated biphenyl (PCB) partitioning between adipose tissue and serum. Bull. Environ. Contain. Tox icol., 33: 277-2S0. 25. Ferguson, J. The use of chemical potentials as indices of toxicity. Proc. Roy. Soc. (London) BI27:387-403 (1939). 26. Henry, R. J. Clinical Chemistry Principles and Technics. Harper and Row, New York, 1963, pp. 836-844. 27. Lawton, R. W., Brown, J. F., Jr., Ross, M. R., and Feingold, J. Comparability and precision of serum PCB measurements. Arch. Environ. Health, in press. 28. Newson, W. B., Morgan, C. B., and Caporal, P. 1933 STATPAC simplified--a short introduction to how to run STATPAC, a gen eral statistical package for data analysis. General Electric Cor porate RAD Report No. 83CRD146, Schenectady, NY, 1983. 29. Dixon, W. J., and Brown. M. G. Biomedical Computing Pro grams, P. Series. University of California Press, Berkeley, 1977. 30. Metropolitan Life Insurance Company. New weight standards for men and women. Statistical Bull. 40: 1-4 (Nov.-Dec. 1959). 31. Anonymous. Plasma lipid distributions in selected North Ameri can populations: The Lipid Research Clinics Program Prevalence Study. Circulation 60: 427-439 (1979). 32. Conney, A. H. Pharmacological implications of microsomal en zyme induction. PharmacoL Rev. 19: 317-366 (1967). 33. Larsen, W. A., and MeCleary, S. J. The use of partial residual plots in regression analysis. Technometrics 14: 781-790 (1972). 34. WolfT, M. S., Thornton, J ,, Fischbein, A., LUis, R., and SelikofT, I. J. Disposition of polychlorinated biphenyl congeners in occu pationally exposed persons. Toxicol. Appl. Pharmacol. 62: 294306 (1982), 35. Chen, P. H,, Luo, M. L., Wong, C. K., and Chen, C. J . Com parative rates of elimination of some individual polychlo rinated biphenyls from the blood of PCB-poisoned patients in Taiwarh Food Chem. Toxicol. 20: 417-425 (1982). 36. Whitfield, J. B., Moss. D. W., Neale, G., Orme, M.t and Breckenridge, A. Changes in plasma gamma-glutamyl trunspeptidase activity associated with alterations in drug metabolism in man. Brit. Med. J . I: 316-318 (1973). Martin, P. J., Martin, J. V., and Goldberg, D. M. Gamma-glu tamyl transpeplidase, triglycerides, and enzyme induction. Brit. Med. J. 1: 17-18 (1975). 38. Martin, J. V., and Martin, P. J. Enzyme induction as a possible cause of increased serum-triglycerides after oral contraceptives. .Lancet i: 1107-1108(1976). 39. Bartels, H., Hauck, W., and Vogel, I, Aminopyrine--an effective modifier of liver and serum gamma glutamyl transpeptidase. J. Pedialr. 86: 298-301 (1975). 40. Teschke, R., Brand, A., and Slrohmeyer, G. Induction of hepatic microsomal gamma-giutamyltransferase activity following chronic alcohol consumption. Biochem. Biophys. Res. Commun. 75: 718724 (1 9 m 41. Goldberg, D. M. The expanding role of microsomal enzyme in duction, and its implications for clinical chemistry. Clin. Chem. 26:691-699(1980). 42. Popper, H., Gerber, M. A., SchaiTner, F.t and SelikofT, I. J. Environmental hepatic injury in man. In: Progress in Liver Dis ease, Vol. 7 (H. Popper and F. SchaiTner, Eds.), Grue and Strat ton, New York, 1982, pp. 605-638. 43. YafTe, S. J., Levy, G., Matsuzawa, T., and Baliah, T. Enhance ment of giucuronide-coqjugating capacity in a hyperbilirubinmie infant due to apparent enzyme induction by phnobarbital. N.Engi. J. Med. 275: 1461-1466 (1966). 44. Kuntzman, R. Drugs and enzyme induction. In: Annual Review of Pharmacology, Vol. 9 (H. W. Elliott, W. C. Cutting and R. H. Dreisbach, Eds.), Annual Reviews Inc., Palo Alto, CA, 1969, pp. 21-36. 45. Juchu, M. R., Gram, T. E., and Fauts, J. R. Stimulation ofhepatic microsomal drug-metabolizing enzyme systems in pri mates by DDT. Gastroenterology 51: 213-218 (1966). 46. Hunter, J., Maxwell, J. D.t Stewart, D. A., Williams, R., Ro binson, J., and Richardson, A. Increased hepatic microsomal en zyme activity from occupational exposure to certain organo chlorine pesticides. Nature 237: 399-401 (1972). 47. Marsh, C. A. Metabolism of o-glucuronolactone in mammalian systems. Identification of o-glucaric acid as a normal constituent of urine. Biochem. J. 86: 77-86 (1963). 48. Hunter, J., Maxwell, J. D., Stewart, D. A., and Williams, R. Urinary D-glucaric acid excretion and total liver content of cy tochrome P-450 in guinea pigs: relationship during enzyme in duction and following inhibition of protein synthesis. Biochem. Pharmacol. 22: 743-747 (1973). 49. Notten, W. R. F., and Henderson, P. T. The interaction of chem ical compounds with the functional state of the liver. I. Alterations in thd metabolism of xenobiotic compounds and D-glucuronic acid pathway, Int. Arch. Environ. Health 38: 197-207 (1977). 50. Notten, W. R. F., and Henderson, P. T. The interaction of chem ical compounds with the functional state at the liver. II. Esti mation of changes in D-glucaric acid synthesis as a method for diagnosing exposure to xenobiotics. Int. Arch..Environ. Health 38:209-220 (1977). 51. Calabrese, E. J. Insufficient conjugate glucuronidation activity: a possible factor in polychlorinated biphenyl (PCB) toxicity. Med. Hypothesis 3: 162-165 (1977). 52. Hirayama, C., Okumura, M., Nagai, J., and Masuda, Y. Hypobilirubinemia in patients with polychlorinated biphenyls poisoning. Clin. Chim. Acta 55: 97-100 (1974). 53. Goldstein, J. A., Hickman, P., Burse, V. W., and Bergman, H. A comparative study of two polychlorinated biphenyl mixtures (Arodors 1242 and 1016) containing 42% chlorine on the induction of hepatic porphyria and drug metabolizing enzymes. Toxicol. Appl. Pharmacol. 32: 461-473 (1975). 54. Burse, V. W., Kimbrough, R. D,, Villanueva. E. C., Jennings, R. W., Linder, R. E., and Sovocool, G. W. Polychlorinated bi phenyls. Storage, distribution, excretion, and recovery: liver morphology after prolonged dietary ingestion. Arch. Environ. Health 29:301-307 (1974). 55. Minot, G. R., and Smith, L. W. The blood in tetrachlorethane poisoning. Arch. Int. Med. 23: 637-702 (1921). 56. Kimbrough, R. D., Squire, R. A., Linder, R. E., Strandberg, J. D., Montali, R. J., and Burse, V. W. Induction of liver tumors in Sherman strain female rats by polychlorinated biphenyl Aroclor 1260. J. Natl. Cancer Inst. 55: 1453-1459 (1975). Spiiometric Findings in Capacitor Workers Occupationally Exposed to Polychlorinated Biphenyls (PCBs) Richard W. Lawton, MD; M arie R. Ross, RN; a n d Joseph Feingold, MD Spirometric findings (forced vital capacity [FVC}, forced expiratory volume at one second [FEVX], FEVjFVC) in a population ofcapacitor workers with occupational exposure to polychlorinated biphenyls (PCBs) are described during active PCS use (1976) and following the PCB ban (1979 and 1983). The initial finding of restrictive impairment (16%) in 1976 was not supported by chest roentgenogram findings, nor con firmed in 1979 and 1983. and was interpreted as artifactual due to test operator inexperience and inadequate expiratory efforts. Obstructive impairment was consistently found in 15% of the total population in 1976 and 1979, A history of respi ratory illness and/or symptomatology and reduced FEVX/FVC was correlated with PCB exposure and serum PCB levels (lower homologs) in females in 1976, but not in males. Smoking was correlated with reduced FEV, values. No correlation of spirometric variables with past exposure or serum PCB levels was found for either sex in 1979. reported by them as equivalent to that of asbestos workers (12.1%). Of those showing FVC reductions. 80% showed a restrictive pattern (FEV:/FVC > 0.70) without radiologic changes. PCBs were not reported as acute direct pulmonary irritants, although eight-hour time-weighted average air levels in the workplace areas of PCB use varied from 0.6.to 11.0 mg/m3.a The present paper reports a spirometry study of capacitor workers, drawn from the same plant popula tion as that of Warshaw et al. but involving only workers with current occupational PCB exposure in 1976. Spi rometry was repeated in 1979, 1981. and 1983 and provided longitudinal data on a selected population. Methods and Materials In spirometric studies of a group of capacitor workers (N = 243), some of whom had high direct occupa tional exposure to PCBs, Warshaw et all found a sub stantial prevalence of reduced vital capacity (FVC < 79.5% of predicted value). The prevalence (14%) was From tho Biological Sciences Branch. Corpo.nte Research and Development, General Electric Company. Schenectady, NY (Dr Lawton. R iD staff): Capacitor Products Department. General Electric Company, Hudson Palls. NY (Ms Ross. Occupational Health Nurse; Du Feingold. Plant Physician. Dr Feingald is retired and currently in the ' practice of internal medicine at 31 East St, Fort Edward. NY 12923). Presented at the American Occupational Health Conference. Kan sas City, MO. April 17. 138S. Address correspondence to: General Electric Company, Corporate Research and Development Center, PO Box 8, Bldg K-l. Room 3B12. Schenectady NY 12301 (Dr Lawton). 0090-1736/&6/2aOtt-33*2.00/0 C o p y rig h t.$ by W illiam s <fc W ilkins The study population consisted of all capacitor work ers with direct work-related PCB exposure In two plants in upper New York state. This population, when selected in 1976, consisted of 194 workers (152 males, 42 fe males) representing approximately 10% of the work force. Ages extended from 22 to 66 years and service times from 1 to 35 years. The cohort was examined in 1976, 1979, 1981, and 1983. Exposure was defined in three categories: (1) high exposure: jobs requiring der mal contact in high air level zones (handling and sealing wet capacitors, capacitor salvage and repair); (2) me dium exposure: jobs requiring brief high exposure and contact such as maintenance men: and (3) low exposure: jobs not involving dermal contact but in or at the periphery of the high-exposure zone. The initial examination consisted of a medical history, physical examination, chest roentgenograms, ECG, se quential multichannel autoanalyzer (SMA-26) blood analysis, hematology, urinalysis, and spirometry. Body weight was obtained in light indoor clothing and height Journal of Occupational Medicine/Volume 28 No. 6/June 1986 453 was determined without shoes. Subsequent examina tions consisted of blood wdrfcup, changes in body weight and blood pressure, and spirometry. Chest roentgeno grams, ECG, and physical examination were repeated in 1983. Smoking classification (smokers, ex-smokers and nonsmokers) was obtained in 1976. Detailed smok ing histories (cigarettes per day, years smoked, etc) were not obtained until 1979 (N = 165). These were generally confirmed in 1983 and the 1979 data were used throughout th analysis with no account taken of the few reports of change in habit over the period. Chest roentgenograms were classified according to the Inter national Labor Organization classification of radi ographs (form OMB No. 68-5 1322). With two excep tions all roentgenograms with positive findings were classified in section 4. Serum PCB levels were obtained in 1976 and 1979 (analysed by Hazleton Laboratories America, Inc, Mad ison, WI) using methods previously described.3 Pulmo nary function was evaluated in 1976 using a Vanguard spirometer with a disposable filter paper flow sensor (Life Support Engineering Corp, Woburn, MA). In 1979 an updated version (Vanguard DS502) with a prepro grammed computer printout was used. Measurements were made in the standing position and the nose closed with the fingers. The calibrating source was a VS300 31syringe and calibration conducted at various volumes and rates as recommended.4 For each worker the max imum expiratory volume-time curve was recorded and examined for freedom from test artifacts.3At least three trials were performed by each worker and reproducibil ity of the expiratory effort was accepted at the 10% level. Measurements were made of forced vital capacity (FVC), the forced expiratory volume at one second (FEVt) and the ratio (FEVj/FVC) using the highest test of the three trials. The data were evaluated as the percent of predicted values using the tabulated predic tions of Knudson et al.e>7The spirometrie patterns were classified as restrictive (FVC < 80% of predicted, FEVt/ FVC > 70%); obstructive (FVC > 80% of predicted. FEV,/FVC < 70%); or mixed (FVC < 80% of predicted, FEVi/FVC < 70%). Initially (1976) the nurse conducting the test had had little spirometrie training and only subsequently was certified (1978) for pulmonary screening. Retro spective review of the 1976 expiratory spirograms in dicated 15 testa (7:7%) to be unacceptable due to tech nical error. Subsequently, only two workers'test results were unacceptable: one due to thoracic arthritis and another due to poorly fitting dentures. Nine workers' test results were unacceptable in 1976 but were accept able in 1979. Statistical analysis was performed on a Honeywdl} Series 600/6000 computer using available time-sharing applications programs.8 Regression analyses were con ducted using STATPAC, a versatile statistical package9 for distribution plotting and the derivation of coeffi cients and correlations by multiple stepwise regression. An incremental F-ratio of 3.9 was used for entering and removing variables so that variables significant at the 5% level were included. Only one marginal association 454 was found and this was treated as significant. PCB exposure was described in terms of the exposure cate gory, treated as a dummy variable, or the serum PCB level, either as reported, or as the concentration in total serum lipids..3 As a consequence of dropouts due to retirement or employee separation, rejected tests, refusal of the pro cedure, or death, the number and composition of the participating population changed during the three ex amination periods. For these reasons, and because of incompleteness of detailed smoking data, the analysis focused on 136 workers for whom complete data were available for the 1976, 1979, and 1983 time periods. In addition, 93 workers were examined at some point in 1981 so that a total of 80 workers had four sets of measurements. These subgroups provided a longitudinal assessment of the spirometry changes with time and of the reliability of the procedure. The entire available population was used to assess abnormal spirometric findings and for the relations with serum PCB levels. Results Workers in the study population were overweight (50% of the workers had body weights above the normal "desirable" range for large-framed individuals10) and heavy smokers. Those who smoked or had smoked (71.4% of males, 54.2% of females) averaged a pack a day for 18 to 20 years. In 1976 there were seven cases of clinically recognized emphysema and 12 workers were told by their physi cians they had chronic bronchitis, although only two complained of current shortness of breath, cough, or sputum production. Past histories of significant pulmo nary infections were reported in 16 cases. Chronic upper respiratory tract infections and allergies were preva lent. Four workers (heavy smokers) reproted leuko plakia and/or laryngeal polyps. On chest roentgenogram examination emphysema was suspected in six cases. The predominant finding was evidence for past pulmonary infection (granulomata, calcifications, scarring). A coin lesion (malignant) was discovered in one worker. Evi dence for fibrosis was found in only two workers. One worker (with carcinoma of the lung) showed bilateral circumscribed pleural thickening; another worker showed old healed pleural thickening at the left base. Mean 1979 serum PCB levels were elevated 35 to 40 times the normal background community levels in hu mans. The levels for PCBs were higher in the female subgroup. The Figure shows spirometry data converted to per cent of predicted value for 80 workers who had meas urements made at four periods during this study. The plotted points are means the 95% confidence limits for the population average. For the males. (N = 75) the mean FVC value rose between 1976 and 1979 (4.53 L to 4.90 L) and then declined (4.70 L in 1983). With the smaller accompanying change in FEVi, the FEV^FVC ratio fell in 1979 and then returned to its 1976 value. However, as indicated in the Figure, only the FEV! population average appeared to be significantly less PCB Exposure/Lawton et al than 100% of the predicted value in 1976. These data indicated a-continuing improvement in the reliability of the measurements with experience of both the operator - and the workers. 4 j Using the predictions of Knudson et al,6workers were classifed in restrictive (FVC 80%; FEVj/FVC > 70%) and obstructive (FEVt < 80%; FEVi/FVC < 70%) cat- C Figure. FVC, FEV,,and FEVi/FVC as percent of predicted value8 for portion of study population with four spirometry tests (N = 80). Means and 95 % confidence limits. egories (Table). In 1976 for the total population (N -- 179) we found that 16.2% of the population showed restrictive impairment. In 1979 the percent of the pop ulation showing this finding declined to 2.9%. In con trast, the obstructive category was relatively constant (15.6% and 15.3%) with 2.2% to 3.4% of the population showing a mixed result (FVC and FEVi both 80% and FEVi/FVC < 70%). Obstructive defects were more prev alent in smokers and ex-smokers. Of those with FEVi/ FVC 70% about 70% had values between 65% and 70% and the ventilatory impairment was considered mild. Four of the 25 workers not returning for the test in 1979 were known to have clinical chronic obstructive lung disease. Because the study dropouts in 1983 made up such a large percentage of the population (21%), scoring for these data provided a lower estimate of the obstructive category. Using multiple stepwise regression we examined the significant partial correlations of the measured spirometric parameters with the conventionally used varia bles (sex, age, body height), as well as those of interest in this study (smoking, respiratory history, and FCB exposure). Although the number of workers with satis factory spirograms and complete data varied for the three examinations, we found strong and complete regressions for FVC and FEVlf but incomplete and weaker equations for FEVx/FVC. Spirometric variables were inversely associated with smoking in all cases except FEVi/FVC in 1976. Body weight was inversely associated with FVC in 1979 and 1983. This latter relation was characteristic of the male subpopulation. The presence of a history of respiratory infection and/ or symptomatology was significantly associated with a decreased FEVt and FEVt/FVC in 1979 and 1983. High exposure to PCBs in the workplace, as measured either by the exposure category or the serum FCB level, was significantly associated with a decreased FEVi/FVC in 1976, a period of active FCB use. However, this associ ation of exposure and FEVi/FVC was found only for the female subpopulation. There were no significant associ ations for the serum FCB level and the spirometric parameters in 1979 for either sex. The Pearson product-movement correlation coeffi- TABLE Restrictive, obstructive and mixed spirometric findings in total population in 1976 and 1979 using predictions of Knudson et al.a N Restrictive FEV, (FVC 80%; i 70%) Male Smokers nonEM da 103 39 19 6 1976 Female Total Male Smokers Mn,, _Ex-smokers J S L Smokers ,,_ ' ud Ei-tnw k. 22 15 179 100.0 94 38 2 2 29 16.2 2 \ \ 0 1979 Female Total Smokers __ and Ns. % E i-n u k . MK**" 22 16 170 100.0 1 2 5 2.9 Obstructive 14 3 9 2 28 15.6 14 2 9 1 26 15.3 (FVC a 80%; ^ - , 70%) Mixed 4 0 1 1 6 3.4 2 0 2 0 4 2.4 l FEV, FVC 8 0 % ;------ 70%) FVC Journal of Occupational Medicine/Volume 28 No. 6/June 1986 455 cients for the constant population (N = 136) in 1976 population. In the female subgroup we found a signifi and 1979 indicated an association of exposure with cant relation between exposure in the workplace and respiratory history and symptomatology in females, al respiratory history and symptomatology in 1976. The though the relation was marginally significant (P = serum PCB levels in females were 11% higher than in 5.12%), In males we found the respiratory history to be the males in 1976 and 50% higher in 1979. These significantly associated with smoking. Smoking was also elevated levels were related initially to the women's associated with a reduced FEV-!, in males in 1976 and employment as capacitor sealers (solderers) and in 1979 but only for females in 1979. salvage and repair, jobs with relatively high exposure levels, and subseqeuently perhaps to a diminished PCB Discussion metabolism compared with that of males. The associa Medical surveillance of a group of capacitor workers directly exposed to FCBs was undertaken in 1976 to assess the workers' general health. The medical exami nation included spirometry, which was repeated in 1979 and 1983, and on a part of the group in 1981. We found pulmonary-function testing to be one of the most difficult procedures to perform well in the industrial medical setting. The procedure requires a trained and certified nurse or technician, functional and properly calibrated equipment, and motivated and cooperative workers If adequate data are to be obtained. These conditions were not fulfilled at the inception of this study: the nurse had little spirometric training and testing experience and worked in cramped quarters with new equipment. The test was new for the workers whose motivation was uncertain, given the intense media coverage of FCBs at the time. As a consequence only 179 out of 194 workers tested (92%) had apparently satisfactory forced expir atory volume-time curves on retrospective review. Fol tion of serum PCB level and/or exposure and reduced FEVi/FVC in females in 1976 appeared as an isolated finding, and no other correlation with spirometry in males or females was found. The present study provided no compelling evidence that PCB exposure, as measured by exposure category or serum PCB levels, was correlated with the occurrence of persistent spirometric abnormalities. The restrictive impairments observed in 1976 were transient and were interpreted in terms of inadequate expiratory effort rather than pulmonary fibrosis. The past histories of respiratory illness and symptomatology, and the ob served obstructive impairment by spirometry appear most likely to be coupled with heavy smoking, a complex of industrial exposures both prior to and during em ployment,11 and rural upstate New York demorgraphic factors.12-13 Our 1976 data, however, are suspect and this study serves to emphasize the importance of nurse/ technician training in the conduct of spirometry. lowing nurse certification and repeated worker experi ence with the procedure, unsatisfactory data were ob r tained in 1% or less of tested cases in 1979, 1981, and References 1983. Because of the technical deficiencies in the spirometry 1. Warshaw R, Fischbein A. Thornton J. at ai: Decrease in vital capacity in PCB-exposed workers in a capacitor manufacturing facil procedures and the uncertainties associated with the ity. Ann N T Acad Sei 1979:320:277-283. serum PCB meausrements in 1976,3 we were inclined to reject the data. However, these observations were ob tained during the period of active PCB use and were characterized by high serum PCB levels in the workers. In addition, the observations coincided in time with 2. Fischbein A. Wolff MS. Lilis R. at al: Clinical Findings Among PCB-Exposed Capacitor Manufacturing Workers. Ann N Y Acad Sci 1979:320:702-713, 3. Lawton RW, Brown JF Jr. Ross MR. et al: Comparability and precision of serum PCB measurements. Arch Environ Health 1983:40:29-37. those of Warshaw et al.1In their studies (1976), as well as tn ours, reduced FVCs were observed. FVCs < 80% of predicted with normal FEV^FVC (>70%) were found in 14% of the population studied by Warshaw et al and 4. Snowbird Workshop on standardization of spirometry. Am. Rev. ReapirDia 1979:119:831-838. 5. Lewis BM: Pitfalls of spirometry. J Occup Med 1981:23:35-38 6. Knudson RJ. Slatin RC. Lebowitz MD. et al: The maximum expiratory flow volume curve: Normal standards, variability, and in 16./2% of our study population. In both studies there were no radiologic evidences of pulmonary fibrosis to account for the restrictive impairment observed. In our study, based on longitudinal observations, we believe the finding to be artifactual. FVC values as effects of age. Am Rev Reap Dis 1976:113:587-600. 7. Cotton Dust Standard. Federal Register 1978:43 (July 23)^7414-27417. 8. Honeywell Series 600/6000 Time-Sharing Applications Library Guide, vol 2: Statistics. Honeywell Information Systems, Inc. (DA 44). 1973. percent of predicted for the population in 1976 were normally distributed over the range, indicating a sys tematic decrease, which could have resulted from gen erally poor coaching and inadequate expiratory effort. 9. Nelson WB. Morgan CB. Caporal P: 1979 STATPAC Simplified: A short introduction to how to run STATPAC. a general statistical package for data analysis. General Electric Co. CRD TXS Report 78CRD276. December 1978. 10. New weight standards for men and women. S tat Bull Metrop In addition, regression studies indicated substantiahdivergence of the prediction coefficients and a higher degree of variability .of the 1976 data compared with that found on subsequent testing. We know of no re versible restrictive pulmonary pathology that could ac Life Found 1959:40:1-1. 11. Brown DP, Jones M: Mortality and industrial hygiene study of workers exposed to polychlorinated biphenyls. Arch Environ Health 1981:36:120-129. 12. Rubin BB: Mortality from lung cancer, emphysema and bron chitis for counties in New York State, excluding New York City. 1960- count for these findings. The principal and consistent respiratory finding in this work force exposed to PCBs was obstructive im pairment which was found in approximately 15% of the 1973. Albany. New York State Department of Health monograph No. 16. 1980. 13. Nosca PC, Burnett WS, Greenwald P. et al: Population density as in indicator of urban-rural differences in cancer incidence, upstate New York. 1968-1972. Am J Epidemiol 1980:112:362-373. 456 PCB Exposure/Lawton et al 86 THE RELATION OP OCCUPATIONAL POLYCHLORINATED BIPHENYL EXPOSURE TO CANCER AND TOTAL MORTALITY PHILIP R. TAYLOR1 JEANETTE M. STELMA2 IVAN AUGER2 CHARLES E. LAWRENCE2 11 April 1988 87 Footnote Page Abbreviations 1. Cl *= confidence interval 2. In - natural logarithm 3. FCBs - polychlorinated biphenyls 4. ppb a parts per billion 5. SE standard error 6. 0 - Obs - observed 7. E Exp - expected 8. SMR standardized mortality ratio 9. ICD - international classification of disease 1Cancer Prevention Studies Branch, Division of Cancer Prevention and Control, National Cancer Institute, Blair Building, Room 6A01, Sethesda, Maryland 20892 (Reprint requests to Dr. Taylor) Statistical and Computer Science Laboratory, Wadsworth Center for Labs and Research, New York State Department of Health During part of this investigation Dr. Taylor was supported by a National Cancer Institute National Research Service Award (5T32-Ca-0900) from the Harvard School of Public Health. The authors gratefully acknowledge the help of Marie Ross and Dr. Richard Lawton, General Electric Corporation; Dr. Ho-Ling Hwang, New York State Department of Health; Drs. George Hutchison, Marcello Pagano, and Richard Monson, Harvard School of Public Health; Dr. Edward Baker, National Institute for Occupational Safety and Health; Emily Brooks, Information Management Services; and the employees whose . cooperation made tfiia study possible. Running Heads Polychlorinated biphenyls and cancer. W 88 Taylor, P.R, (National Cancer Institute, Bethesda, Maryland 20892), J.M. Stelma, I. Auger, and C.E. Lawrence. The relation of occupational polychlorinated biphenyl exposure to cancer and total mortality. ABSTRACT We studied the relation of polychlorinated biphenyls (PCBs) to total and cancer mortality in a cohort of men and women occupationally exposed to PCBs during the manufacture of capacitors. A total of 6292 employees who worked more than three months between 1946 and 1975 were followed for subsequent mortality through 1980 with identification of 136 cancer and 510 total deaths. The effect of PCB exposure was estimated in two ways: by comparing mortality rates for the entire cohort with national rates, and by comparing exposure in cancer deaths to that in matched reference subjects from within the cohort. For the internal comparison; cumulative PCB exposure, cumulative high-homolog serum PCB exposure, and average high-homolog serum PCB levels wre estimated separately for each case \ and reference subject using an independently derived empirical model Results of thifc study indicate that overall cohort . 89 mortality is .significantly less than expected compared to the general populace (observed = 510, expected = 614.6; standardized mortality ratio [SMR] = 8 3 , 95 per cent confidence interval [Cl] = 76-90). Total cancer deaths do not vary from expectation (observed 136, expected = 144.5; SMR = 9 4 , 95 per cent Cl 79-111). Rone of the estimates of PCB exposure are associated with an increase in risk of total cancer, lung cancer (N-32 cases), or colorectal cancer (N=25 cases). In summary, these data provide no evidence for an effect of high-homolog PCB exposure on cancer mortality, though the numbers of cancer deaths at individual sites are small. V 90 INTRODUCTION x Polychlorinated biphenyls (PCBs) are chlorinated aromatic hydrocarbons consisting of mixtures of 210 different isomers having varying degrees of chlorination. First prepared in 1867, PCBs were commercially manufactured in the USA between 1929 and 1977. The chemical and thermal stability of PCBs resulted in their wide use in capacitors, transformers, hydraulic fluids, heat transfer fluids, lubricants, plasticizers, and as components of surface coatings and inks. Retention of these lipophilic compounds in animals, humans, and the general environment has been well documented and is related to both the degree of chlorination and the position of the chlorine atoms on the PCB molecule (1,2). The PCBs with lower degrees of chlorination (low homologs) tend to be more rapidly excreted while the more highly chlorinated PCBs (high homologs) are retained. Increasing concern about environmental contamination and potential health effects contributed to a ban in 1979 on the further distribution of PCBs in this countiy. [ i There have beh a number of studies in animals which have found administration of various of the PCBs to result in tumor formation (3). The literature relating PCB exposure to cancer in humans is, however, much more I 91 limited. A series of comparisons of fat sample analyses in cancer and noncancer patients conducted by Unger et al in Denmark found increased PCB levels in cancers at several sites, though results were inconsistent (4-6). Seven reports of cancer based on five different groups of occupationally exposed workers are found in the literature. Bahn et al. reported two melanomas from a group of 72 petrochemical workers exposed to PCBs in New Jersey when only 0.04 was expected (7). Roush found lung cancer in excess (7 observed, 2.7 expected) among 89 workers employed in the manufacture of PCBs in St. Louis (8). Excesses of liver (3 observed, 1.07 expected) and rectal (4 observed, 1.19 expected) cancer among capacitor manufacturing workers in two facilities located in upstate New York and western Massachusetts were noted by Brown and Jones (9). Follow-up of the same cohort seven years later continued to find an excess of cancer of the liver (5 observed, 1.9 expected) and rectal cancer (4 observed, 1.9 expected) (10). In Sweden, Gustavsson et al. found 21 total deaths among a cohort of 142 male workers engaged in the manufacture of capacitors when 22.1 were expected (11). Neither total cancer deaths (7 observed, 5.39 expected) nor incident cases (7 observed, 7.58 expected) were elevated. Most recently, Bertazzi et al. studied total and cancer mortality in Italian workers employed f' 92 in a capacitor manufacturing plant (12,13). An excess of total cancer among males was seen (14 observed, 5.5 expected), mainly at digestive, lung, and hematologic sites. Overall mortality in females exceeded expectation (34 observed, 25.8 expected) as did cancer mortality (12 observed, 7.7 expected). We report here our study of cancer and total mortality among a large group of workers occupationally exposed to PCBs in the manufacture of capacitors. i 93 METHODS Description of the cohort Between 1946 and 1977 two facilities of the sane company located in upstate New York manufactured capacitors using PCBs with Aroclors 1254/ 1242, and 1016 as their primary dielectric fluid* Our study population consists of the cohort of all employees who worked at either of these two facilities between the years 1946 and 1975 for a minimum of three months. Employees were first identified by personnel records for these years and later verified by obtaining copies of Social Security premium payment records for the years 1945 to 1965. A total of 6303 persons were verified in this way to have worked at least three months during the defined study period. Considering the Social Security premium reports as the reference, missing personnel records numbered only 41 after verification efforts were finished, indicating that the cohort was 99 per cent complete for 4129 persons who started work before 1965; A total of ii additional exclusions were made for impossible dates (n * 10) or a missing date of birth (n 1), leaving a final analytic cohort of 6292, including 2'69l \ females and 3601 males. Case ascertainment Mortality was identified by searching a number of 94 sources. The primary resource used was the Social Security Administration premium payment report files (through 1980). This was supplemented by additional searches of the New York State Death files (through 1981), New York State Tumor Registry (through 1981), New York State Department of Motor Vehicles files (through 1982), company personnel records (through 1980), National Death Index files (for 1979 and 1980), and Veterans Administration records (for Vietnam deaths, through 1981). Records were considered complete through December 31, 1980 and this was used as the common closing date for all subjects in the study. Subjects with no indication of death after searching the above sources were assumed to be alive as of this date. Deaths known to have occurred after this date were not included in the analysis. Exposure assessment Work history information on the personnel records included a code, a description, and dates for each job at the facilities. Tfcfe job descriptions in combination with manufacturing process information and industrial hygiene data were used to categorize all jobs into two broad exposure groups. Direct-exposure jobs were defined as those in which dirSct contact with FCBs occurred during the manufacturing proiess. These jobs were further characterized into subcategories as follows: 95 Low.... air contact only Medium..air contact plus occasional dermal contact High....air contact plus frequent dermal contact All other jobs within the plants, including office and manufacturing areas where PCBs were not directly used, were termed indirect-exposure jobs* While PCBs were still in use at the facilities, environmental monitoring was performed in both direct- and indirect-exposure job areas during three industrial hygiene surveys. Results (summarized in Table 1) showed that in 1977, concentrations of PCBs in the indirect-exposure job areas were an order of magnitude below those in the direct-exposure job areas, although the indirect-exposure job areas had much higher PCB concentrations than did areas surrounding the plants, where values averaging 6.2 micrograms per cubic meter were recorded prior to discontinuation of PCB use. These concentrations all exceed previously reported urban ambient air averages of 0.1 microgram per cubic meter (14). Serum total PCB determinations performed on workers in these / facilities (summarized in Table 2) demonstrated a geometric \ mean PCB concentration for workers in direct-exposure jobs more than four-fold higher than that in indirect-exposure jobs and almost 20-fold higher than that in non-exposed referenda subjects. Exposure to chemicals at these facilities was liAifced to thosa used in the manufacture of / 96 capacitors including PCBs, trichlorethylene, methyl isobutyl ketone, lead, zinc, tin, aluminum, iron, and epoxides (15,16). Exposure was homogeneous compared with the highly mixed exposure environment found in most chemical work environments. Serum PCB levels and work histories were available for a sample of 194 employees at these plants. Linear regression methods were used on these data to develop a model for the prediction of high-homolog serum PCB levels. The resulting empirically derived model formed the basis for our primary index of high-homolog PCB exposure. This model is described in the Appendix. The model allowed us to estimate high-homolog serum PCB level in each cohort member at the time of each event being studied. Furthermore, through incorporation of different kinetic assumptions, the model can be used to estimate a number of different exposure parameters. For the present study, three different exposure estimates were made. The first exposure estimate, cumulative exposure, is the sum of duration of employment times level of exposure across 5 levels and 4 eras of exposure. Algebraically it is summarized as: CE^i cumulative exposure Et' 97 where T = year of diagnosis, and t = year of exposure. The second exposure parameter is cumulative serum PCB level. Using first-order kinetic assumptions and a half-life determined empirically, this approach first estimates the serum PCB levels for each year beginning with initial employment and ending with the death of the case (for cases) or matched subject (for referent subjects), and then sums them up. This is expressed algebraically as: CST cumulative serum PCB level * ~ o x=-o = o where T and t are as above, k - elimination constant, and Sx - serum level in year x. The third exposure parameter used here is average annual serum PCB level. This measure is the estimated cumulative serum PCB level divided by the number of years since first employed and is expressed as: ST * average annual serum PCB level CST T+l T Z % rro X Z Et exp(-k[x-t]) i - o __________________ T+l 98 Statistical methods Two different analytic approaches were used in this study. The first estimated standardized mortality ratios (SMRs) to compare.the observed mortality experience in the cohort with that which was expected based on national mortality. The second analysis was limited to cohort members only and compared estimated serum PCS levels in cases to those in reference subjects. For the first analysis, the life table computer program of Monson was used (17) Person-years were combined into 5-year calendar time periods and 5-year age groups and multiplied by the corresponding U.S. white male (for male cohort members) and U.S. white female (for female cohort members) cause-specific mortality rates to yield the expected number of deaths. Observed and expected cause-specific deaths were compared using standardized mortality ratios (SMR, observed/expected x 100) with 95 per cent confidence intervals (Cl) calculated using square root transformations of the Poisson variates. By convention, no Cl was calculated when the difference between observed and expected deaths was less than 0.5, and no SMR or Cl was calculated when no cases were observed or the expected . 99 value was less than 1.0. For the second analysis, a matched case-referent approach was used (18). Cancer deaths were identified and ordered by time since entry into the cohort. Hatching by age (+ 5 years of date of birth), sex, and time since initial employment at the capacitor manufacturing facility was used with selection of 10 reference subjects for each case. To simulate the risk set approach, reference subjects had to be alive (and therefore at risk of death) for at least as many years following initial entry into the cohort as the case to which they were matched. For one case only nine matches could be found, for another only eight could be found, and for a third only two were found. Selection of reference subject's was made with replacement so that a subject could serve as a reference for more than one case. In addition, reference subjects for early cases were eligible themselves to become cases later on. Each of the three PCB exposure estimates was calculated separately for each case and reference subject for the time period of initial employment until death of the case. Distributions for these estimates were skewed and ln-transformation was employed so that they more closely approximate normal. Computations for these analyses were conducted in SAS using PROC PHGLH with the blocking option to simulate a conditional logistic regression (19). 100 RESULTS As of 31 December 1980, 5593 subjects in the cohort were known to be living, 510 were known to be dead, and 189 (3 per cent) were not known to be living or deceased. The number of deaths, years of employment, and person-years for the cohort are shown in Table 3. A total of 510 deaths including 136 cancer deaths were identified over 122,782 person-years of observation. Total mortality is significantly less than expected for males and females combined (observed 510, expected - 614.6; SMR - 83, 95 per cent Cl - 76-90) but cancer deaths are not different n from expectation (observed - 136, expected * 144.5; SMR 94, 95 per cent Cl * 79-111). The observed and expected cause-specific mortality for males is shown in Table 4. The overall SMR (355/429.9 x 100 * 83) is significantly less than 100, as is the SMR for all respiratory disease (SMR 10/21.6 x 100 * 46). No other significant deviations from eipbcted are seen. Table 5 shows similar cause-specific mortality for females. The overall SMR (155/184.6 x 100 - 84) is also significantly less than 100 for females. In addition, deaths from diseases of the circulatory system are less than expected (SMR * 38/64.4 x 100 59) while deaths from accidents exceed expectation (SMR * 21/11.0 x 100 - 190). No significant deviations 101 from expectation are seen for site-specific cancers in either males or females. The observed number of cases is slightly less than,expected for males (69 versus 83*5) and slightly greater than expected for females (67 versus 60.9). Table 6 shows SMRs by latency intervals separately for males and females for cancer and total mortality. For longer latency periods, the cancer SMRs show a slight decrease for males and an increase for females. Tables 7 and 8 group cancers by systems and show, separately for males and females, the observed and expected numbers of cancer deaths adjusted for all-cancer mortality. This r \ simple adjustment attempi:s to provide a more appropriate reference group for the occupational group being examined. Compared to unadjusted values, this approach results in SMRs for males that are slightly higher and for females that are slightly lower. Table 9 presents characteristics of the subjects who are part of the matched case-referent study while Tables 10 - 12 show results of the conditional logistic analyses for all cancer mortality sites combined. Lung and colorectal cancer are also analyzed separately because they are the sites with the greatest numbers and are implicated based on a priori hypotheses. None of the three PCB exposure estimates is associated with an increased risk of cancer in the case-referent analyses. In fact, coefficients for these exposure estimates for all-sites, lung, and colorectal cancer analyses are negative, indicating decreased risk with increased PCB exposure. Though inadequate numbers precluded-more detailed analysis of the cancer sites previously mentioned in relation to PC: exposure (i.e., lung, colorectal, liver, melanoma, and lymph- and hematopoietic), calculated average serum PCB exposure is shown by case and reference status for the latter three sites in Figure 1. 103 f DISCUSSION We studied the relation of high-homolog PCB exposure to subsequent cancer mortality in a large group of men and women occupationally exposed during the manufacture of capacitors A total of 510 deaths were identified from among the 6292 cohort members after over 122,000 person-years of follow-up. Several features of this study plan and the findings make it unlikely that bias accounts for the results observed in this study. Persons were considered part of the cohort on the basis of personnel records originally obtained in 1975. Completeness of the records was confirmed using Social Security premium payment records for persons employed before 1965. Exposure was examined in two ways, both objective. First, all members of the cohort were considered exposed in comparison with the general population. And second, an additional analysis using only the cohort itself was conducted using an independently derived high-homolg serum PCB estimation model which relied on job code and date information obtained from the personnel records in conjunction with a separate assignment of exposure levelv (indirect, low, medium, high) to each job code. And finally -, case ascertainment included a wide variety of resourcs and appears complete or nearly complete for mortality. Though bias seems unlikely in t study, confounding could certainly be present. As is typical for occupational cohort studies, information on covariates is extremely limited. There are a number of other limitations of the preset study which bear mention. There is an acknowledged, inherent limitation in the interpretation of results froi occupational cohort studies which compare the mortality c workers to that of the entire U.S. population. The "healthy worker effect" for total mortality.typically observed in such studies is also seen in the present study. Data on the ideal reference group, a large exterr cohort with characteristics identical to those of the cohort under study but lacking PCB exposure, does not exist. The comparisons conducted within the cohort itsel are limited because of the small numbers of site-specific cancers. Liing cancer deaths, for example, are the most common cause of death from cancer for both sexes combined but number only 32. We prefer to examine cancer incidenc rather than mortality. However, we are unable to identif all incident cancers in persons who move out of New York state and are still alive. The consequence of *this is a j reduction in numbers of outcome events which is particularly acute for non-fatal cancers. And finally, o detailed, site-specific exposure evaluation is limited to 105 high-homolog PCB exposure. There are a number of strengths that this study has in comparison with other studies of PCB-related mortality. First, this is the largest cohort of PCB-exposed workers in the U.S., and possibly in the world. As a result, this study has 2.2 times as many person-years of observation and cancer deaths, and approximately 1.7 times as many total deaths as the next largest reported study (10) Second, this is the only cohort study in which an attempt is made to quantify the level of individual exposure to PCBs. Studies based on the five different occupationally-exposed groups reported in the literature all assumed only that cohort members were exposed but do not rank or otherwise discriminate among their exposure levels. The present study includes both methods for classifying exposure. Finally, although the cancer site-specific analyses are limited to persons from within the plant, estimated PCB exposure includes both low and high values and varies widely among subjects. Model-derived estimates of average annual serum high-homolog PCB levels at the time of death, for example, range from 24 to 24,319 parts per billion. In conclusion, we studied the relation of high-homolog PCB exposure to subsequent total' and cancer mortality in an occupationally-exposed cohort. Total mortality was less than expected for both males and females. No effect of PCB 106 r" exposure on overall cancer rates was observed though there was a suggestion of an increase in rates with longer latency periods for females. Caution must be used in interpreting the results in light of the limited number of site-specific cancer deaths available for evaluation at this time. Follow-up of this cohort in future years will be possible and will give a more secure evaluation of the FCB hazard. i REFERENCES I 1. Bush B, Turnasonis CF, Baker FD. Toxicity and I persistence of PCB homologs and isomers in the avian system.. Arch Environ Contam Toxicol 1974; 2 (3): 195-212. 2. Sundstrom G, Hutzinger 0, Safe S. The metabolism of chlorobiphenyls-- a review. Chemosphere 1976;5:267-98. 3. Polychlorinated Biphenyls and Polybrominated Biphenyls. In: IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. Lyon, France: International Agency for Research on Cancer, 1978;18:66-7. 4. Unger M, Olsen J. Organochlorine compounds in the adipose tissue of deceased people with and without cancer. Environ Res 1980;23:257-63. 5. Unger M, Olsen J, Clausen J. Organochlorine compounds in the adipose tissue of deceased persons with and without cancer: a statistical survey of some potential confounders. Environ Res 1982;29:371-6. 6. Unger M, Kiaer H, Blichert-Toft M, et al. Organochlorine compounds in human breast fat from deceased with and without Breast cancer and in a biopsy material from newly diagnosed patients undergoing breast surgery. Environ Res 1984;34:24-8. \ 7. Bahn AK, Rosenvaike I, Herrmann N, et al. Melanoma after exposure to FCBs (letter). N Engl J Med 108 1976;295:450. 8. Roush 6. Polychlorinated Biphenyls. In: NIOSH Criteria Document. Cincinnati, Ohio: National Institute for Occupational Safety and Health, 1976:65. 9. Brown DP, Jones J. Mortality and industrial hygiene study of workers exposed to polychlorinated biphenyls. Arch Environ Health 1981;36:120-9. 10. Brown DP. Mortality of workers exposed to polychlorinated biphenyls - an update. Arch Environ Health (in press) 11. Gustavsson P, Hogstedt C, Rappe C. Short-term mortality and cancer incidence in capacitor manufacturing workers exposed to polychlorinated biphenyls (FCBs) Am J Ind Med 1986;10:341-4. 12. Bertazzi PA, Zochetti C, Guercilena S, et al. Mortality study of male and female workers exposed to PCB's. In: Prevention of occupational cancer-- International Symposium. Geneva: International Labor Office, Occupational Safety and Health Series, 1982;46:242-8. 13. Bertazzi PA, Zliboldi L, Pesatori A, et al. Cancer mortality of capacitor manufacturing workers. Am J Ind Med 1987;11:165-76. \ 14. Kutz FW, Yang HSC. A note on polychlorinated biphenyls in air. In: Proceedings of the National Conference on 109 Polychlorinated Biphenyls, Washington, D.C.: Environmental Protection Agency, 1975:182. 15. Jones H. Industrial hygiene survey of the General Electric Company. Hudson Falls, Hew York; Fort Edward, New York. National Institute for Occupational Safety and Health, Cincinnati, Ohio. January 6, 1978. 16. Lawrence C. pc.:? nd melanoma (letter). N Engl J Med 1977;296:108. 17. Honson HR. Analysis of relative survival and proportional mortality. Comput Biomed Res 1974;7:325-32. 18 Lubin JH, Gail, MH. Biased selection of controls for case-control analyses of cohort studies. Biometrics .1984;40:63-75. 19. SAS Institute, Inc: SAS User's Guide. 1982 ed. Cary, NC, Sas Institute, Inc., 1982. \ 110 TABLE III.l Capacitor manufacturing plant PCB monitoring1 (geometric means/ micrograms per cubic meter) Area Air Samples (October 1975) Area Air Personal Air Samples Samples2 (April 1977) (April 1977) Direct Exposure Areas 679 (n-30) 310 (n-16) 168 (n-31) Indirect Exposure Areas 260 (n-l) 27 (n-16) Written personal communication, R. Lawton, 1980 (Note: caution should be used in making direct comparisons between results from 1975 and 1977 as the sampling techniques and analytic procedures employed in these 2 surveys were different) 2Time-weighted average , 'A Ill J TABLE III.2 Serum PCB concentrations in capacitor workers, 19791 Geometric Mean (parts per billion) N Low High Total homologs homologs PCBs Workers in Direct- Exposure Areas 147 269 33 302 Workers in IndirectExposure Areas 18 50 11 61 Reference Group2 16 7 9 16 Written personal communication, R. Lawton, 1980. 2Employees without occupational PCB exposure working for the same company at another facility. 112 TABLE III.3 Characteristics of cohort and events by gender Hales Females Total Number of persons Number of deaths Person-years Average age at entry Average year of entry Average age at death Average year of death 3601 355 68505 27 1960 54 1970 2691 155 54278 28 1960 55 1972 6292 510 122783 27 1960 55 1971 ICD 0 1 2 9 140-9 150-9 150 151 153 154 155-6 157 160-3 161 162 TABLE III.4 Observed and expected deaths by <cause for males Cause Observed Exoected SMR f95% CI1 All causes of death 355 429.9 83 (74-92) All malignant neoplasms 69 83.5 83 (64-105) All infective & parasitic diseases 3 5.1 59 (12-151) All tuberculosis 1 2.5 39 (1-216) Cancer of buccal cavity & pharynx 2 2.6 75 (8-269) Cancer of digestive organs & peritoneum 25 21.3 117 (76-173) Cancer of esophagus 3 1.8 159 (32-463) Cancer of stomach 1 3.7 26 (0-147) Cancer of large intestine 9 6.9 130 (59-246) Cancer of rectum 4 2.3 173 (47-444) All cancer of liver 2 1.5 133 (15-480) Cancer of pancreas 6 4.3 138 (50-301) Cancer of respiratory system 20 29.0 69 (42-106) Cancer of larynx 0 1.2 -- Cancer of lung 20 27.4 73 (44-112) 170 Cancer of bone 172-3 Cancer of skin 185 Cancer of prostate 186-7 Cancer of testis 188 Cancer of bladder 189 Cancer of kidney 190 Cancer of eye 191-2 193 Cancer of brain & nervous system Cancer of thyroid 200-9 All lymph-- & hematopoietic cancer 200 Lympho- & reticulosarcoma 201 Hodgkin's disease 204-7 Leukemia & aleukemia 202--3,208 Cancer of other lymphatic tissue 210-39 Benign neoplasms 240 Allergic, endocrine, metabolic, & nutritional diseases 2 5 0 Diabetes mellitus r 1 0.4 2 2.0 98 (11-355) 4 3.6 110 (30-281) 1 0.9 -- O 1.9 4 2.1 184 (49-470) 1 0.0 1 3.3 30 (0-165) O 0.1 -- 6 9.6 62 (23-136) O 2.0 O 1.6 4 3.7 106 (29-272) 2 2.1 94 (11-339) O 1.3 -- 5 7.0 71 (23-165) 4 5.8 69 (18-176) i4i----*** 280-9 290-317 320-89 390-458 393-8 410-4 430-8 460-519 480-6 492 493 520-77 531-3 571 580-629 All diseases of blood & blood-forming organs Mental, psychoneurotic, & personality disorders All diseases of nervous system & sense organs All diseases of circulatory system Chronic rheumatic heart disease Arteriosclerotic heart disease All vascular lesions of CHS All respiratory diseases All pneumonia Emphysema Asthma All diseases of digestive system All gastric & duodenal ulcer Cirrhosis of liver A H diseases of genito-urinary system ! -V V .- 1 0.9 3 3.2 92 (19-269) 3 4.4 68 (14-199) 164 190.9 86 (73-100) 4 4.5 89 (24-227) 126 137.1 92 (77-109) 14 21.3 66 (36-110) 10 21.6 46 (22-85) 4 7.5 53 (14-135) 3 5.0 60 (12-174) 0 0.7 16 23.2 68 (39-111) 2 2.6 76 (8-273) 10 13.6 73 (35-135) 3 5.0 59 (12-173) 582 Chronic nephritis 680-709 All diseases of skin 710-38 All diseases of bones 780-99 Symptoms, senility, & ill-defined conditions 800-998 All external causes of death 800-949 All accidents 810-27 Motor vehicle accidents 950-9 Suicide J- 1 2.0 49 (1-275) 0 0.2 -- 1 0.7 -- 5 5.5 91 (29-212) 72 75.2 96 (75-120) 56 50.8 110 (83-143) 22 27.2 81 (51-122) 10 16.1 62 (30-114) ICD 0 1 2 9 140-9 150-9 150 151 153 154 155-6 157 160-3 TABLE III.5 Observed and expected deaths by cause for females . Cause Observed Expected SMR f95% c n All causes of death 155 184.6 84 (71-98) All malignant neoplasms 67 60.9 110 (85-140) All infective and parasitic disease 0 2.6 -- All tuberculosis 0 1.0 -- Cancer of buccal cavity & pharynx 2 0.8 -- Cancer of digestive organs 6 peritoneum 19 12.4 152 (92-238) Cancer of esophagus 0 0.4 -- Cancer of stomach 2 1.5 125 (14-453) Cancer of large intestine 9 5.4 165 (75-313) Cancer of rectum 3 1.2 235 (47-686) All cancer of liver 1 1.1 90 (1-500) Cancer of pancreas 4 2.1 182 (49-466) Cancer of respiratory system 12 7.1 167 (86-292) 161 Cancer of larynx 162 All cancer of lung 170 Cancer of bone 172-3 Cancer of skin 174 Cancer of breast 180-4 Cancer of all genital organs 180-2 Cancer of all uterus 180 Cancer of cervix uteri 181-2 Cancer of corpus uteri 183-4 Cancer of other genital organs 188 Cancer of bladder 189 Cancer of kidney 190 Cancer of eye 191-2 Cancer of brain & nervous system 193 Cancer of thyroid 200-9 All lymph- & hematopoietic cancer 2 0 0 Lymph- and reticulosarcoma 2 0 1 Hodgkin's disease ir o 0.1 -- -- 12 7.4 160 (83-280) O 0.2 -- 1 1.0 -- 13 15.4 84 (45-144) 13 10.5 123 (65-210) 7 5.6 125 (50-257) 5 3.5 140 (45-326) 2 1.8 108 (12-391) 6 4.9 122 (44-265) 1 0.4 -- 1 0.8 -- O 0.0 -- O 1.9 . -- O 0.1 -- 4 5.3 75 (20-193) O 1.1 ------ o 0.7 0ii----0* 204-7 Leukemia & aleukemia 202-3,208 Cancer of other lymphatic tissue 210-39 Benign neoplasms 240-79 Allergic, endocrine, metabolic & nutritional diseases 250 Diabetes 280-9 All diseases of blood blood-forming organs 290-317 Mental, psychoneurotic, & personality disorders 320-389 All diseases of nervous system sense organs 390-458 All diseases of circulatory system 393-8 Chronic rheumatic heart disease 410-3 Arteriosclerotic heart disease 430-8 All vascular lesions of CNS 460-519 All respiratory diseases 480-6 All pneumonia 3 2.0 144 (29-421) 1 1.2 79 (1-440) 0 1.1 -- 4 5.2 76 (20-194) 4 4.2 94 (25-240) 1 0.7 -- 1 0.9 -- 2 2.7 72 (8-259) 38 64.4 59 (42-81) 2 3.9 51 (6-184) 23 34.5 67 (42-100) 9 13.4 67 (30-127) 7 8.0 87 (35-180) 3 3.2 92 (18-180) \L J ' 492 Emphysema 493 Asthma 520-77 All diseases of digestive system 531-3 All gastric 6 duodenal ulcer 571 Cirrhosis of liver 580-629 All diseases of genito-urinary system 582 Chronic nephritis 680-709 710-38 All diseases of skin & cellular tissue All diseases of bones 780-99 Symptoms, senility, & other ill-defined conditions 800-998 All external causes of death 800-949 All accidents 810-27 Motor vehicle accidents 950-9 Suicide c o 1.0 2 0.6 -- 6 10.6 57 (21-123) O 0.7 -- 3 5.9 50 (10-146) 2 3.1 64 (7-232) 2 1.0 190 (21-686) O 0.3 -- 1 0.9 -- O 2.1 -- - 26 18.2 142 (93-209) 21 11.0 190 (117-290) 8 6.3 126 (54-248) 4 5.1 77 (21-198) roo TABLE III.6 Standardized mortality ratios (and number of cases) for cancer and total mortality by years since starting work and sex Cancer Mortality Males Females Total Years Since Starting Work Ori IQ-19 20+ Total 92 (13) 83 (23) 78 (33) 83 (69) 63 (8) 115 (23) 126 (36) 110 (67) 78 (21) 97 (46) 97 (69) 94 (136) Total Mortality Males Females Total 77 (80) 85 (123) 84 (152) 83 (355) 67 (28) 81 (47) 94 (80) 84 (155) 74 (108) 84 (170) 87 (232) 83 (510) ro TABLE III.7 Observed and expected cancer deaths adjusted1 for all-cancer mortality (males) Site ilCDl All cancers (140-209) Observed Exoected SMR2 69 83.5 83 Adjusted Adjusted Expected SMR <95% C l h 69.0 100 (-- ) Gastrointestinal (140-57) 27 24.0 112 19.9 135 (89-191) Respiratory (160-3) 20 29.0 69 24.0 83 (50-123) Genitourinary (180-9) 9 8.7 103 7.2 124 (56-220) Lymph- and Hematopoietic ^- 6 9.6 62 7*9 75 (27-148) -- -- - ro ro r' (200-9) other (170,172-3, 190,191-2,193) 5 6.1 81 5.0 98 (31-206) ^Expected values are adjusted by multiplying them by the O/E for all-cancer mortality (0.83). 2. SMR standardized mortality ratio = observed/expected x 100. Cl = confidence interval. r ro CO r "' TABLE III.8 Observed and expected cancer deaths adjusted1 for all-cancer mortality (females) Adjusted Adjusted Site ICD Observed Exnected a s 2 Exoected SMR f95% All cancers 67 60.9 110 67.0 100 (-- ) (140-209) Gastrointestinal 21 13.3 157 14.6 143 (88-211) (140-57) Respiratory 12 7.4 160 8.1 147 (76-243) (160-2) Genitourinary 15 11.9 126 13.0 114 (64-181) (180-9) Lymph- and Hematopoietic 4 5.3 75 5.8 68 (17-153) (200-9) Other 14 18.8 74 20.6 67 (37-108) (170,172-3,174, 4ro* V*t-r" - "vV r 190,191-2,193) ^-Expected values are adjusted by multiplying them by the O/E for all-cancer mortality (1.10). 2SMR standardized mortality ratio - observed/expected x 100. 2Cl - confidence interval. r. rcon r' TABLE III.9 Characteristics of subjects from matched case-referent study Median frange) All Subjects Lung Cancer Colorectal Cancer Cases Referents Cases Referents Cases Referents (H135)1 (N1339) (N=32) (N=318) (11=25) (11=249) Age at death 56 54 58 58 (years)2 (21-84) (20-84) (38-84) (28-84) Sex (% male) 50% 50 62% 62 Duration exposure 20 20 19 19 (years)3 (1-32) (1-32) (5-30) (5-30) First employed 1951 1950 1951 1951 (year) (1946-72) (1946-71) (1946-70) (1946-70) 56 (36-78) 52 54 (26-81) 52% 22 22 (2-32) (2-32) - 1951 1949 (1946-65) (1946-70) f V. Cumulative PCB 2482 2826 2987 2897 1612 3084 exposure3'4 (24-57656) (24-85559)(43-44469)(24-83421) (115-57656)(24-62782) (ppb x years) Cumulative serum PCB 11389 14048 15794 15063 8310 15953 exposure3'4 (98-301659)(74-436911)(218-232345)(226-436911)(328-301659)(74-326375) (ppb x years) Average serum PCB 613 749 709 799 455 820 (ppb)3 '4 (24-12066)(24-24319)(34-8605)(24-24319) (60-12066)(24-17478) 3One case in the SMR analysis is deleted from the case-referent analysis because of lack of adequate information to calculate exposure level* 2For referents, age at time of death of matched case. JPrior to death (cases) or prior to death of matched subject (referents) 4See methods for detailed description of exposure calculation. ro  TABLE III.10 Relation of In of estimated cumulative PCB exposure level to cancer mortality from conditional logistic regression model Site Number Coefficient1 SE2 95% Cl3 P-value All 135 -0.1116 Lung4 32 -0.1150 Colorectum5 25 -0.0216 0.0620 0.1261 0.1391 -0.2331,0.0099 -0.3621,0.1321 -0.2942,0.2509 0.072 0.361 0.876 *Ln odds ratio per unit increase in cumulative PCB exposure. 2Standard error. 3Confidence interval. 4ICD 160-3 5ICD 153-4 ro CD cO TABLE 1 1 1 ,1 1 Relation of In of estimated cumulative serum PCB exposure level to cancer mortality from conditional logistic regression model Site Number Coefficient1 SE2 95% CI3 P--value All 135 -0.1406 Lung4 32 -0.1261 Colorectum5 25 -0.0509 0.0686 0.1389 0.1521 -0.2750,-0.0061 0.058 -0.3983,0.1461 0.364 -0.3490,0.2472 0.737 1Ln odds ratio per unit increase in cumulative PCB exposure. 2Standard error. ^Confidence interval. 4ICD 160-3 5IC0 153-4 e- i ^ :: iros TABLE I I I . 12 Relation of In of estimated average annual high-homolog PCB level to cancer mortality from conditional logistic regression model Site Humber Coefficient1 SE2 95% Cl3 P-value All Lung4 s' Colorectal5 135 32 25 -0.1290 -0.1238 -0.0334 0.0695 0.1415 0.1541 -0.2660,0.0064 -0.4011,0.1534 -0.3354,0.2686 0.061 0.381 0.828 *Ln odds ratio per unit increase in cumulative PCB exposure. 2Standard error. 3Confidence interval. I .1*t A' 131 APPENDIX A regression model approach vas used to develop a high-homolog serum PCB estimation model based on data from a company survey. The survey collected blood from 194 persons including 152 males and 42 females in 1976 selected to include all current employees whose job required direct PCB contact in high air level zones, employees in the immediate periphery of the high-exposure zone, and employees having high but intermittent exposure. A follow-up survey was conducted in 1979 on all of the 194 directly exposed workers who were still available, numbering 174. Because of our interest in PCBs that are retained in the body longer, we focused our efforts on a model for high-homolog PCBs (Aroclor 1254) Because of concern over a systematic laboratory error in the 1976 data, only values from 1979 were used in this analysis. Complete job history records through 1976 and serum Aroclor 1254 measurements from 1979 were available on 157 employees. While additional information was gathered on these 157 persons, information used in development of the prediction model was limited ib items found in the company personnel \ records and the jcffc exposure categorization (described earlier under Methods). This was done to allow us to generalize use of &e model to the entire cohort of 6292 . i \ t* 132 persons. Work history variables included the number of months at jobs from each of four exposure levels (described in the text under Methods) for each of four time intervals. In 1954 the use of highly-chlorinated Aroclor 1254 was phased out and replaced by Aroclor.1242; in 1965 major engineering changes occurred, including closed process filling of capacitors and improvements in ventilation, which presumably resulted in substantial reduction in exposure to PCBs; and in 1971 Aroclor 1242 use was stopped in favor of Aroclor 1016. To account for the potential influence of these changes, separate variables corresponding to the periods 1946-54, 1955-65, 1966-71, and 1972-76 were incorporated into the model to indicate the era of exposure: Era 1......... 1946-54 (25-33 years before sampling) Era 2 .... .1955-65 (14-24 years before sampling Era 3 ......... 1966-71 (8-13 years before sampling) Era 4........ ;1972-76 (3-7 years before sampling) Weighting of tiift independent variables was employed .to satisfy the regression assumptions of homoscedasticity, normal error distribution, and linearity. The model assufi&s first order kinetics. With an * t estimated half-lifS of 3^32 years determined empirically by application of simple linear regression to the difference in serum PCB estii&tifes from 1979 and 1983 on a subset of 133 150 persons in this cohort, serum PCB level can be estimated at any point in time. When the half-life is set to infinity, cumulative PCB exposure up to the time of interest is estimated. Indicator variables for exposure levels by time intervals were created using data from all 157 persons. Running the full least squares regression model with 16 exposure-time categories, ten influential outlier values (standardized residual < -2.5 or > 2.5) were identified, reviewed, and removed. Removal of the ten outliers resulted in an improvement in the R2 from 0.56 to 0.71. The model was then simplified by removal of all four of the statistically insignificant variables from era 4 and a single insignificant variable for low exposure'during era. The final, reduced model was: Serum PCB - 24.7 + 75.3 X (No. months at + 652.1 X (No. months at + 882.7 X (No. months at + 77.5 X (No. months at + 6:6 X (No. months at + 39.4 X (No. months at + 95.8 X (Np. months at + 1.7 X (No. months at + 0.7 X (No* months' at 4" ;& % 134 + 7.3 x (No. months at medium in era 3) + 16.0 x (No. months at high in era 3) Regression coefficients are equal to serum PCB concentration per era-specific exposure category month. The R2 for this reduced model vas 0.69 based on 147 observations. /v \ LN OF AVERAGE ESTIMATED SERU M PCB V S. CAN CER SITE Figure I I I . 1 3 u^ jg*** cfS*o** g* RtfS' CAN C ER SITE Legend A PCB X median uin ( FIGURE III.l - LEGEND 136 Estimated serum PCB level is shown for only two cases of melanoma. The one additional skin cancer found in the SMR analysis lacked adequate information to calculate exposure. V