Document yp7Bm9OGEvwVZ05G2gVOZnzGr

3M Company Page 1 Date: May 16.2006 FINAL REPORT Epidemiology', 220-3W-05 Medical Department 3M Company St. Paul. MN 55144 Title: An Analysis of the 2000 Fluorochemical (Perfluorooctanoate, PFOA) Medical Surveillance Program at 3M Company's Antwerp (Belgium), Cottage Grove (Minnesota), and Decatur (Alabama) Facilities Geary W. Olsen, DVM, PhD*1 Larry R. Zobel, MD, MPH' 1. 3M Company, Medical Department, Mail Stop 220-6W-08, St. Paul, MN 55144-1000 3M Company Page 2 Abstract The 3M Company has published several periodic medical surveillance studies of its fluorochemical production workers at its Antwerp (Belgium), Cottage Grove (Minnesota) and Decatur (Alabama) manufacturing facilities. These programs have compared clinical chemistry results in relation to serum measurements of periluorooctanesulfonatc (PFOS) and/or perfluorooctanoate (PFOA). Toxicologically, PFOA is a peroxisome proliferator alpha receptor (PPARa) agonist and exerts morphological and biochemical effects characteristic of PPARa agonists including beta-oxidation of fatty acids, increased CPY450-mediated reactions, and inhibition of the secretion of very' low-density lipoproteins and cholesterol from the liver in rats and mice. In worker studies there have been no consistent associations with PFOA and various clinical chemistry measurements. A weak positive association between PFOA and serum total cholesterol was reported in an analysis of a the 2000 fluorochemical medical surveillance data for Antwerp and Decatur employees combined but it was considered a spurious finding based on the understanding that a hypolipidemic effect would be the hypothesis to be tested from the toxicological data. A subsequent unpublished brief letter from DuPont reported that total cholesterol, LDL and triglycerides, but not HDL, were positively associated with their employees' serum PFOA concentrations but the authors urged caution as minimum variance was explained in their analyses. In light of these findings, the purpose of this study was to reanalyze the 2000 fluorochemical medical surveillance male employee database (n = 552) which included three facilities: Antwerp, Cottage Grove and Decatur and to stratify the analyses by employees* self-reported cholesterol lowering medication 3M Company Page 3 status concentrating on those individuals (n = 506. 92 percent) who reported they were not taking cholesterol lowering medications. Using several types of statistical analyses (analysis of variance, analysis of covariance, logistic regression and multiple regression using multiplicative models), there was no evidence that PFOA [adjusted for age, body mass index (BMI) and alcohol] was statistically significantly (p < .05) positively or negatively associated with serum total cholesterol or LDL (calculated by the Friedwald formula when serum triglycerides were less than 400 mg/dL) regardless of cholesterol lowering medication status. Although HDL was negatively associated with PFOA for the combined three locations, only one percent of its variance was explained with PFOA and this association was not observed when the data were stratified by location. Thus, the association observed with HDL and PFOA was likely due to residual confounding between the locations themselves rather than to any possible causal association with PFOA. For the combined locations, serum triglycerides were positively associated with PFOA but not consistently by location. Antwerp, the location with the lowest mean triglyceride and PFOA concentrations, showed a positive association whereas Cottage Grove, with the highest mean PFOA and triglyceride levels, showed no association between triglycerides and PFOA. A hypothesis is offered that the inconsistent associations observed for serum triglycerides and PFOA might be due to non-adherence to fasting requirements by night shift production workers since all sample collections occurred in the morning for production and non-production employee participants. The former would have higher serum PFOA concentrations. Also, shift workers, in general, have been reported to have more elevated 3M Company Page 4 serum triglycerides. To clarify an association, if any, between PFOA concentrations and serum triglyceride levels, the methodological questions raised herein merit further inquiry. Adjusted for the potential confounding factors of age, BMI or triglycerides, and alcohol, the three locations combined did not result in consistent associations with hepatic enzyme tests. Whereas the individual Antwerp and Cottage Grove locations showed no statistically significant associations, weak positive associations were observed in the multiplicative models between alkaline phosphatase, ALT, and GGT with PFOA (adjusted for age, BMI or triglycerides, and alcohol) among the Decatur male employee population which could be due to residual confounding as minimum variance was explained by PFOA in these models. Several epidemiologic research studies of the Decatur workforce have not found statistically significantly increased risks for hepatic disease (malignant or nonmalignant conditions) using a variety of data sources including episodes of care, self-reports, and death certificates. Other worker populations and a community-based PFOA-exposed population have also not reported positive associations between liver enzyme test results and serum PFOA concentrations. Few individual thyroid-related hormone (TSH, T4, free T4, and T3) results were out-of reference range. Analyses of the multiplicative models predicted these thyroid-related hormone measurements would be well within their normal reference ranges when PFOA concentrations varied between 0.005 pg/mL (average general population concentration reported) and 100 pg/mL (high range of occupational measurements historically 3M CompanyPage 5 reported). The lack of an association with clinically relevant thyroid test results has also been reported in other worker and community-based studies of PFOA whose average serum concentrations were at or below those reported in the present study. Introduction 3M Company Page 6 The 3M Company has published several periodic medical surveillance studies of its fluorochemical production workers at its Antwerp (Belgium), Cottage Grove (Minnesota) and Decatur (Alabama) manufacturing facilities (Ubel 1980; Gilliland and Mandcl 1996; Olsen et al. 1998; 1999; 2000; 2003a; 2003b). These programs have compared clinical chemistry results in relation to serum measurements of perfluorooctanesulfonate (PFOS) and/or perfluorooctanoate (PFOA). Serum PFOS concentrations measured in these 3M employees have been routinely compared with serum total cholesterol levels as a decline in the latter has been a consistent early reliable measure of clinical response reported in laboratory animal studies (3M Company 2003; Haughom and Spydevold 1992; Seacat et al. 2002; 2003). No association between PFOS and cholesterol has been observed in these fluorochemical medical surveillance programs, and is likely due to the fact that the serum PFOS concentrations measured (average approximately 1 to 2 pg/mL, maximum < 14 pg/mL) were below those measured in laboratory animals where effects have been reported (Olsen et al. 1999; 2003a; 2003c). Among cynomolgus monkeys fed PFOS in the diet for 6 months, statistically significant decreases in cholesterol from predose values occurred in the high dose group (0.75 mg/kg/day) that had serum PFOS concentrations above 100 ppm (Seacat et al. 2002). This serum PFOS concentration in the high dose group corresponded to 50 mg/kg and 34 mg/kg cumulative doses in males and females, respectively. Seacat et al. (2002) considered this decrease in serum total cholesterol the 3M Company Page 7 earliest reliable measure of clinical response to PFOS in these cynomolgus monkeys. Lowered HDL values were also reported in the 0.75 mg/kg/day dose group. However, a lack of prestudy and interim HDL values in a lower dose group (0.15 mg/kg/day) made this interpretation more problematic. In a sub-chronic dietary toxicity of PFOS in rats, Seacat et al. (2003) did not report strong evidence for hepatocellular peroxisomal or cellular proliferation at the doses tested (0, 0.5 2.0 5. 0 and 20 parts per million in diet) Lowered serum cholesterol was observed although decreased glucose among male animals at the high dose at 4 weeks (but not at 14 weeks or with females at 4 or 14 weeks) and elevated alanine transaminase levels (ALT) at 14 weeks in males (but not females) were also reported. The mode of action for serum cholesterol reduction is uncertain but may be due, in part, to PFOS acting as a peroxisome proliferator alpha receptor (PPARa) agonist. PFOA is a PPARa agonist and exerts morphological and biochemical effects characteristic of PPARa agonists including beta-oxidation of fatty acids, increased CPY450-mediated reactions, and inhibition of the secretion of very low-density lipoproteins and cholesterol from the liver in rats and mice (Haughom and Spydevold 1992; DePierre 2002; Kennedy et al. 2003; Xie et al. 2003). These effects resulted in a reduction of serum cholesterol and triglycerides in rats and mice. Xie et al. (2003) have shown that severe adipose tissue atrophy occurs upon dietary treatment of mice with PFOA but is rapidly reversed after termination of treatment. Hepatomegaly, however, was much more persistent after dose termination. Unlike rats and mice, there was no reduction in serum cholesterol in cynomolgus monkeys dosed with PFOA (ammonium 3M Company Page 8 salt) for 6 months (Butenhoff ct al. 2002). PFOA was statistically significantly positively associated with triglycerides in the high-dose (30/20 mg/kg) group whose steady state serum PFOA concentration was 158 pg/mL 10 pg/mL (range 20 to 467 pg/mL). This association between PFOA and serum triglycerides was observ ed in measurements taken after one month of dosing at which time the group mean triglyceride level was significantly higher than control values as well as within group pretreatment values. At the end of the study the mean triglyceride was elevated compared to time related controls but not to the animals' pretreatment values. However, only two primates were evaluated in the high-dose group at end of study. Inspection of individual values for PFOA serum concentration and serum triglyceride values did not reveal a meaningful association between these two parameters (John Butenhoff, personal communication; see Olsen et al. 2003a). In worker studies there have been no consistent associations with PFOS or PFOA and various clinical chemistry measurements (Gilliland and Mandel 1996; Olsen et al. 1998; 1999; 2000; 2003a; 2003b). Gilliland and Mandel (1996) did report that PFOA (measured as the surrogate serum total organic fluorine) may negatively modulate the effect alcohol has on high-density lipoprotein (HDL) levels and exacerbate the effect that obesity has on hepatic enzyme tests. However, three subsequent analyses of this employee population that measured specifically for PFOA did not find that it modulated hepatic responses to either obesity or alcohol consumption (Olsen et al. 2000). 3M Company Page 9 Previous analyses of the most recently published fluorochemical medical surv eillance data conducted at the 3M Antwerp and Decatur facilities (conducted in 2000) primarily concentrated on PFOS analyses although data were also presented for PFOA. In these PFOA-specific analyses, Olsen et al. (2003a) reported a positive association with serum cholesterol and triglycerides with PFOA. They attributed this to be a spurious finding given the toxicological evidence would suggest a hypolipidemic (not hyperlipidemic) effect, along with the fact that minimum variation was explained in the statistical models (partial R2 for PFOA in the models was <0.01 (cholesterol) and 0.03 (triglycerides), respectively) used in the analyses. In the same published paper, Olsen et al. (2003a) also provided a longitudinal analysis of 174 Antwerp and Decatur male employees who participated in at least two of three fluorochemical medical surveillance programs between 1994 and 2000 and found PFOA to also be positively associated with cholesterol as well as triglycerides. This association was primarily attributed to 21 Antwerp employees whose mean serum PFOA levels increased over a six-year period of time from 1.32 ppm to 2.06 ppm. During the same time period, their mean cholesterol values increased from 208 mg/dL to 229 mg/dL and their triglyceride levels increased from 85 mg/dL to 123 mg/dL. Their BMIs, however, also increased from 23.4 to 24.3 during this same time period. No statistically significant associations were reported between PFOA and either serum cholesterol or triglycerides for the Cottage Grove male employees who participated in the 2000 fluorochemical medical surveillance program (Olsen et al. 2003a). Analyses of prior fluorochemical medical surveillance programs at Cottage Grove (1993, 1995 and 1997) showed mean serum triglyceride levels were highest among the Cottage Grove PFOA production workers with the highest (>= 10 ppm) PFOA 3M Company Page 10 serum concentrations although controlling for potential confounders did not provide consistent results as well as the fact that several of the observations in the high exposure category were repeated measurements from the same individuals. Based on the possibility of a positive association between PFOA and serum cholesterol and/or triglycerides that was not supported by laboratory (toxicological) data and was therefore considered a spurious association (Olsen et al. 2003a). others have since provided nonpublished data (Costa 2004; Leonard 2005). Costa (2004) provided cursory analyses of approximately 35 Italian fluorochemical production workers and reported a "slight increase of total cholesterol in workers exposed to PFOA." There was no increase of other lipids, such as triglycerides, but the fraction of non-HDL cholesterol appeared elevated. Costa suggested his findings might be consistent with the hypothesis that PFOA might influence cholesteryl ester transfer protein (CETP). CETP is a plasma glycoprotein that facilitates the transfer of cholesteryl esters from HDL (apolipoprotein A-containing lipoprotein) to apolipoprotein B-containing lipoproteins (e.g., LDL) (Brousseau et al. 2004). Inhibition of CETP has been shown to markedly elevate plasma levels of HDL and apolipoprotein A l. However, there did not appear to be a decrease in HDL in the Costa analysis. Several additional caveats also existed in the Costa data analysis as outlined by Kaplan (2005) including: 1) the Costa data set was a small and arbitrary collection of subjects; 2) there were no pre-employment/baseline lipid levels for historical reference; 3) concomitant exposure to other chemicals was unknown; and 4) inadequate adjustment for important confounding factors. In addition, many of these 3M Company Page 11 individuals were the same subjects over time and no repeated measures analysis was conducted. Leonard et al. (2005) have provided to the U.S. EPA an unpublished one page report that briefly summarized a cross-sectional analysis of 782 male and 243 female (combined eligible population 1,863) DuPont workers with potential exposure to PFOA at the Parkersburg, West Virginia facility. Of the 62 clinical chemistry and hematology endpoints measured, Leonard et al. found most were well within normal ranges and not associated with serum PFOA levels. Measured PFOA concentrations were reported as high as 10 pg/mL (parts per million, ppm) but the average was less than 1 pg/mL (Leonard, personal communication). No statistically significant associations were observed for PFOA and serum liver enzymes or any hematology measures and PFOA. EKGs and C-reactive protein were not associated with PFOA. Statistically significant positive associations were identified for total cholesterol, LDL and triglycerides with serum PFOA levels adjusted for BMI, alcohol and age in both males and females although the percent variation explained was generally low'. The potential change in total cholesterol at the highest serum PFOA level was approximately 10 percent. No effect, how'ever, wras seen in the HDL fraction. Small, but statistically significant increases in uric acid and iron were also reported with the highest PFOA blood levels. A suggested role for PFOA to enhance CETP activity, as discussed by Costa, however, was not supported by the much larger Leonard et al. analyses since the latter investigators showed no association between PFOA and HDL. Leonard et al. offered no hypotheses or suggested causal models (Hernn et al. 2002) to explain their observations. Because 3M Company Page 12 PFOA has been shown to bind to blood albumin in the rat. monkey and human (Han et al. 2003; Kerstner-Wood et al., 2003), a positive correlation, and not a causal association, between lipoproteins (binding) and PFOA is a possibility. Re-examination of the Olsen et al. analyses conducted of the 2000 fluorochemical medical surveillance program at the Antwerp and Decatur facilities (Olsen et al. 2001a; 2001b; 2001c; 200Id; 2003a) and the Cottage Grove facility (Olsen et al. 2003b) suggested some limitations of these data analyses as they may relate to testing a hypothesis that PFOA is positively associated with total cholesterol and its fraction, LDL. First, all male subjects were included in the original analysis regardless of their cholesterol lowering medication status. A positive association may be masked by inclusion of subjects whose serum cholesterol levels have been reduced by medication if such an increase was, in part, associated with higher PFOA concentrations. Second, LDL analyses were not restricted to those instances where serum triglycerides were 400 mg/dL or less. The potential for bias steadily increases with higher triglyceride levels (Nakanishi et al. 2000). Third, the Antwerp and Decatur facility study (Olsen et al. 2003a) concentrated primarily on PFOS and not PFOA. Both PFOS and PFOA have been shown to result in hypolipidemia in rats at high concentrations and thus the causal model hypothesized in these earlier analyses concentrated on testing for this effect, not hyperlipidemia. Olsen et al. (2003a) did not show a causal association between PFOS and lowered cholesterol at the serum concentrations measured in these workers. Direct comparison with PFOA from the Olsen et al. (2003a) report is not possible since the many of the tabular analyses (e.g., Table 2 of Olsen et al. 2003a) were PFOS-specific. 3M Company Page 13 Although the employees' PFOS and PFOA concentrations were correlated, the categorical analyses of PFOS did result in overlapping of employees' PFOA concentrations. Fourth, there was no combined analysis of the three manufacturing facilities, Antwerp, Cottage Grove and Decatur, as the Cottage Grove analysis has historically been reported separately (Ubel et al. 1980; Gilliland and Mandel 1996; Olsen et al. 2000; 2003a). Those sites that were combined (Antwerp and Decatur) may have resulted in unmeasured and/or residual confounding since the two locations were distinctly different with some potential confounding variables, most especially alcohol consumption and BMI (Olsen ct al. 1999; 2003a). With the above limitations in mind, the purpose of this study was to reanalyze the 2000 fluorochemical medical surveillance program data in order to examine the hypothesis that PFOA may be positively associated with increased cholesterol, LDL levels and triglyceride levels. To address some of the above limitations, the three manufacturing facilities were analyzed separately and jointly. Analyses concentrated on those male employees who self-reported that they were not taking cholesterol lowering medications in order to minimize any unexplained bias. Because PFOS concentrations were not previously associated with lowered serum cholesterol (Olsen et al. 2003), it (PFOS) was not considered an important potential confounding variable in these reanalyses for PFOA. Three covariates were: age, BMI and alcohol. Age is known to be positively associated with serum cholesterol. BMI is positively associated with serum triglycerides (and to some degree cholesterol) and alcohol consumption (e.g., red wine) has been positively associated with increased HDL. In addition to the lipid variables analyzed, liver enzyme 3M Company Page 14 and thyroid tests were also reanalyzed. Although liver enzymes, in particular gamma glutamyl transferase (GGT), aspartate aminotransferase (AST) and alanine aminotransferase (ALT) can be positively associated with heavy alcohol consumption, liver enzymes can also be due to obesity and dyslipidemia (Collantcs et al. 2004; Mofrad and Sanyal 2003; Ruhl and Everhart 2003) as nonalcohol fatty liver disease has substantially increased in prevalence in the United States population, as indicated by the third National Health and Nutrition Examination Survey (NHANES) (Clark et al. 2003). Whereas the majority of elevated aminotransferase activity in NHANES could not be explained by alcohol consumption, viral hepatitis or hemochromatosis, unexplained aminotransferase enzyme elevation was significantly associated with higher BMI, waist circumference, triglycerides, fasting insulin, and lower HDL (Clark et al. 2003). Clark et al. (2004) concluded that unexplained aminotransferase elevation was strongly associated with adiposity and other features of the `metabolic syndrome', and thus may represent nonalcoholic fatty liver disease. [Note: The metabolic syndrome includes abdominal obesity, atherogenic dyslipidemia (elevated triglyceride, small LDL particles, low HDL cholesterol), raised blood pressure, insulin resistance (with or without glucose intolerance), and prothrombotic and proinflammatory states (Expert Panel 2001). Clinical identification of the metabolic syndrome in men includes any three of the following: waist circumference > 40 inches (surrogate BMI >= 30); triglycerides 150 mg/dL; HDL < 40 mg/dL; blood pressure 130/85 mmHg; and fasting glucose 110 mg/dL.] As they relate to the present study, the NHANES findings indicate BMI and serum triglycerides are important predictors of liver enzyme values and therefore should be controlled in any analyses that examine the relationship between PFOA and liver 3M Company Page 15 enzyme tests (e.g., ALT. GGT), as was done previously with analyses with serum PFOS concentrations in this employee population (Olsen et al. 2003a). Methods The following clinical chemistry' variables were considered dependent variables for reanalysis of the 2000 fluorochemical medical surveillance program data: alkaline phosphatase (IU/L), gamma glutamyl transferase (GGT, IU/L), aspartate aminotransferase (AST, IU/L), alanine aminotransferase (ALT, IU/L), total and direct bilirubin (mg/dL), blood glucose (mg/dL), cholesterol (mg/dL), low density cholesterol (LDL, mg/dL), high density cholesterol (HDL. mg/dL), and triglycerides (mg/dL). These measurements were performed at Allina Laboratories (St. Paul, MN). LDL was an indirect calculation using the Friedwald formula [LDL = total cholesterol (triglycerides/5)]. Presented in this report are LDL values where serum triglycerides did not exceed 400 mg/dL. LDL was also calculated when serum triglycerides did not exceed 250 mg/dl as well as for all triglyceride levels. Results were not substantially different for LDL regardless of formula used. Thyroid stimulating hormone (TSH; pIU/mL); serum thyroxine (T4; pg/dL); free thyroxine (free T4; ng/dL) and serum triiodothyronine (T3; pg/mL) were also re-examined. These thyroid related hormones were measured by LabCorp (Kansas City, MO). Thyroid hormone analyses excluded the five individuals who were already diagnosed with thyroid-related conditions and likely taking medication. Analyses did not substantially differ with these individuals included. 3M Company Page 16 Details of the method validation including matrix extraction and analytical measurement procedures used to determine serum PFOA and PFOS can be found elsewhere (see U.S. EPA docket AR226-1208, AR226-1209, AR226-1210). The analytical method consisted of a liquiddiquid extraction procedure followed by evaporation and reconstitution of the extract residue with 30:70 20 mM ammonium acetate in water:20 mM ammonium acetate in methanol (v/v). The samples were analyzed by liquid chromatography/tandem mass spectrometry using a PE Sciex API 3000. The instrument was operated in the multiple reaction monitoring (MRM) mode under optimized conditions for detection of PFOS and PFOA for detection of negative ions formed by turbo ionspray ionization. Laboratory analyses were conducted at Tandem Laboratories, Salt Lake City, UT (formerly Northwest Bioanalytical Laboratory Inc.). All employee serum values (pg/mL) for PFOA (and PFOS) were above the lowrer limit of quantitation (LOQ). Statistical analyses included both univariate and multivariable methods (analysis of variance, analysis of covariance, logistic regression, multiple regression) using JMP (Cary, NC) and Stata (College Station, TX) software. Age, BMI and alcohol (drinks per day) were considered as covariates in multivariable models. For analysis of hepatic enzyme tests in these models, triglycerides were also considered in place of BMI. Distributions were examined to assess normality using nontransfonned and transformed (log, square root, inverse) variables. In general, log transformations improved normality assumptions of response and explanatory variables and thus multiplicative models were considered applicable (T. Church, personal communication). For the log transformation of alcohol, 0.1 was added to drinks per day to prevent the log of 0. Residual plots w^ere 3M Company Page 17 examined to detect model inadequacies including homoscedasticity. Goodness of fit statistics were examined including R~ and adjusted R:. Multicollinearity in the models was assessed by the inverse of tolerance (Variation Inflation Factor). Crude and adjusted odds ratios for PFOA categorized by deciles were determined via logistic regression analyses for reference range values of the response variable. A review of the 2000 Cottage Grove database used in the Olsen et al. (2003b) report found eight male employees had missing BMI and/or alcohol information. For the present study, a review of the subjects' Cottage Grove medical records by the plant nurse provided these data and was generally within two years of the 2000 Cottage Grove medical surveillance program. Graphical analyses were performed to examine patterns of association (see Appendix A) for PFOA. Graphs in Appendix A include nontransformed as well as transformed (natural log) variables. Density ellipses are also provided in Appendix A for PFOA. A density ellipsoid was considered a good graphical indicator of the correlation between two variables. The ellipsoid collapses diagonally as the correlation between two variables approaches either 1 or -1. The ellipsoid is more circular (less diagonally oriented) if the two variables were uncorrelated. Results A total of 196 (95%) of the 206 Antwerp male employee participants, 122 (93%) of the 131 Cottage Grove male employee participants and 188 (87%) of the 215 Decatur male 3M Company Page 18 employee participants were included in the primary reanalyses of the 2000 fluorochemical medical surveillance program (Table 1). These 506 (92%) participants represent those who did not self-report taking choleseterol lowering medications. The mean PFOA concentration was statistically significantly (p < .05) higher among Cottage Grove than either Antwerp or Decatur male employee participants but the median PFOA concentration for Decatur was somewhat higher than either Antwerp or Cottage Grove participants (Table 2). [Note: Decatur also had a slightly higher mean concentration for PFOS than Antwerp or Cottage Grove employee participants, with the median value approximately twice that of Antwerp or Cottage Grove.] As has been reported before (Olsen et al. 1999; 2003a), Antwerp employees were statistically significantly younger, had lower BMIs, and consumed more alcohol than their counterparts at Decatur, and in the present analysis, also Cottage Grove. Whereas mean cholesterol and LDL were not statistically significantly different between the three locations, mean HDL levels were statistically significantly higher among Antwerp employees whereas their triglyceride and blood glucose levels were significantly lower. Mean (arithmetic) liver enzyme tests were statistically significantly lower for Antwerp compared to Cottage Grove and Decatur employees except for total and direct bilirubin. Antwerp employees had a statistically significantly lower mean TSH than Decatur employees and y higher mean T3 than the Cottage Grove or Decatur employees. These mean thyroid values, however, were well within the reference range. 3M Company Page 19 The number and percent of employees by reference points for demographic factors and clinical chemistry results are presented in Table 3. As expected based on the data presented in Table 2, there were substantially fewer Antwerp employees (5%) who would be categorized as obese (BMI > 30) compared to Cottage Grove (44%) or Decatur (32%) employees. More than 40% of Antwerp and Cottage Grove employees reported drinking 1 alcohol beverage per day compared to only 1% of Decatur employees. Statistically significant differences between the three locations were seen for blood glucose. HDL, triglycerides. ALT, and total bilirubin. There was only 1 percent of the Antwerp employees who were categorized as having the `metabolic syndrome' compared to 20% and 13% among the Cottage Grove and Decatur employees, respectively. The present study definition of the metabolic syndrome included any three of the following four: BMI > 30; triglycerides > 150 mg/dL; HDL < 40 mg/dL; and fasting glucose >110 mg/dL. Presented by PFOA quartile ranges are the data for Antwerp (Table 4), Cottage Grove (Table 5) and Decatur (Table 6). Quartiles were divided to allow for a defined break in PFOA concentrations. Among Antwerp employees, the mean value in the first PFOA quartile w'as statistically significantly different than the fourth PFOA quartile for PFOA, PFOS, age, blood glucose, triglycerides and T3. Among Cottage Grove employees, the mean value in the first PFOA quartile was statistically significantly different (higher) than the fourth PFOA quartile for PFOA and AST. Among Decatur employees, the mean value in the first quartile wfas statistically significantly different (lower) than the fourth PFOA quartile for PFOA, PFOS, ALT, AST/ALT ratio, total bilirubin and T3. The 3M Company Page 20 AST/ALT ratio is a useful index for distinguishing nonalcoholic fatty liver (steatohepatitis) disease (AST/ALT less than one) from alcoholic liver disease (AST/ALT greater than one). It should be noted that those locations (Cottage Grove and Decatur) with the highest obesity (as measured by BMI which is not necessarily as good of a measurement as waist/hip circumference ratio) had the lowest AST/ALT ratio. The 506 employees from the three locations were combined and categorized into PFOA deciles in Table 7. Mean PFOA decile concentrations ranged from decile 1 of 0.06 pg/mL (range 0.007 - 0.13 pg/mL) to decile 10 of 12.15 pg/mL (range 3.71 - 92.03 pg/mL) with corresponding median values of 0.06 pg/mL and 4.94 pg/mL, respectively. The distributions of demographic factors by PFOA decile are presented in Table 8. There was an uneven distribution of employees by location in these deciles. If there was a proportional distribution across deciles by location, then there would be 39% Antwerp, 24% Cottage Grove, and 37% Decatur employees in each decile. Instead, as seen in Table 8, there are higher percentages of Antwerp employees in the First 3 deciles and lower percentages in the highest two deciles (9 and 10). Likewise, there are lower percentages of Decatur employees in the first three deciles and higher percentages in the upper deciles (six through 10), in particular decile 9. Cottage Grove employees have higher than expected percentages in decile 1 and decile 10. Because of these uneven distributions, mean BMI values are greater in the upper deciles and alcohol consumption is lower, reflecting the demographic differences seen across the three facilities. 3M Company Page 21 The mean and median PFOS concentrations by PFOA decile are presented in Table 9. In general PFOS concentrations increased by decile but not consistently. Deciles 7 through 10 are all statistically significantly higher than deciles 1 through 5. Presented in tables 10 through 13 are the analyses for cholesterol, LDL, HDL and triglycerides. Table 10 is the univariate analysis by PFOA decile. Table 11 presents the mean decile values adjusted for age, BMI and alcohol using analysis of covariance. Table 12A presents adjusted odds ratios and 95% confidence intervals of the deciles by reference range cutoff points listed in the table using decile 1 as the reference. Odds ratios were adjusted for age, BMI and alcohol. Table 12B presents crude and adjusted (for location) odds ratios and 95% confidence intervals of the deciles by reference points for HDL and triglyceride. Table 13 presents non-adjusted and adjusted PFOA coefficients with cholesterol, LDL, HDL and triglycerides adjusted for age. BMI and alcohol in the multiplicative models. Analyses included all locations as well as presented individually for each location by itself. Based on the analyses presented in Tables 10 through 13, cholesterol and LDL were not statistically significantly associated with PFOA whether across the combined three locations or each location analyzed separately. Adjusted mean HDL levels were lower in the highest (10th) PFOA decile compared to deciles 1 through 4 (Table 11). The crude odds ratio was 3.0 (95% Cl 1.2-7.5) for HDL < 40 mg/dL in decile 10 compared to decile 1 (Table 12B). However, this was reduced to an adjusted odds ratio of 1.5 (95% Cl 0.64.0) when location was considered a covariate. Similar results were observed for 3M Company Page 22 triglycerides (Table 12B). The adjusted PFOA coefficient in the multiplicative model for HDL was statistically significant at p = .01 (Table 13) but only explained 1 percent of the HDL variance in the full model (R2= 0.26), which was primarily attributable to BMI and alcohol. This finding with PFOA was also attributed to the decile disproportionate distribution of subjects as shown previously with the much greater proportion of Cottage Grove and Decatur employees in the highest two PFOA deciles. Multiplicative models stratified for each of the three locations showed no statistically significant findings between PFOA and HDL (Table 13). Mean serum triglyceride levels were highest in decile 10 whether unadjusted or adjusted (Tables 12A and 12B). Odds ratios for triglyceride levels > 150 mg/dL were highest for deciles 8 through 10. The multiplicative model for the combined three locations indicated a statistically significant coefficient (p < .0001) for PFOA (Table 13) which explained approximately 4 percent of the variance of the response variable. Stratified by location, however, showed only Antwerp to produce a statistically significant coefficient although Decatur had a marginal nonstatistically significant coefficient for PFOA (p = .07). No association ( p = .38) was found between PFOA and triglycerides for Cottage Grove in the multiplicative models (Table 18). Cottage Grove had the highest PFOA concentrations of the three locations. As seen in Tables 2 and 3, serum triglyceride levels were considerably lower among Antwerp than Cottage Grove or Decatur employees. Presented in tables 14 through 18 are the analyses for alkaline phosphatase, AST, ALT, AST/ALT ratio, GGT, and total and direct bilirubin. Table 14 is the univariate analysis 3M Company Page 23 by PFOA decile. Table 15A presents the mean decile values adjusted for age. BMI and alcohol using analysis of covariance. Table 15B presents the mean decile values adjusted for age. triglycerides and alcohol using analysis of covariance. Table 15C presents the adjusted AST/ALT ratios. Table 16 presents the percent of employees by PFOA decile that were above reference range for these liver clinical chemistry tests. Tables 17A and 17B present adjusted odds ratios and 95% confidence intervals of the deciles by different reference range cutoff points for ALT and GGT provided in table 16 using decile 1 as the reference. These odds ratios were non-adjusted (crude), adjusted for age, BMI and alcohol, and adjusted for age, triglycerides and alcohol. Table 18 presents non-adjusted and adjusted PFOA coefficients of these liver enzymes using either age, BMI and alcohol or age, triglycerides and alcohol (natural logs) as covariates in the multiplicative models. Analyses in Table 18 included all locations as well as presented individually for each location. Overall, the mean non-adjusted ALT is statistically significantly higher in deciles 9 and 10 compared to deciles 1 through 6 (Table 14) but this association is not evident upon adjustment of the mean values in Tables 15A or 15B. Odds ratios for values above the reference range are only calculated for ALT and GGT (Tables 17A and 17B) because these were the only liver clinical chemistry tests that had more than a few employees outside the reference range (Table 16). There were no statistically significant odds ratios (non-adjusted or adjusted) for ALT or GGT w'hen compared to decile 1 (Table 17A). Analysis of a lower cutoff reference point (> 40 IU) did not substantially alter the findings (Table 17B). Analyses did not result in any statistically significant associations 3M Company Page 24 when all locations were included in the multiplicative models, except for total bilirubin (negatively associated) (Table 18). Stratified by location, Decatur had marginally statistically significant positive coefficients for alkaline phosphatase, ALT, GGT and a negative log coefficient for total bilirubin in the models. The amount of variance of the hepatic response variables explained by PFOA in these models was minimal ranging from <1 to 3 percent. These hepatic enzyme associations with PFOA were not observed for the individual Antwerp or Cottage Grove analyses. Presented in tables 19 through 22 are the PFOA analyses with TSH, T4, free T4 and T3. Table 19 is the univariate thyroid analyses by PFOA decile. The mean TSH value in the 4th decile (3.41 pIU/mL) is influenced by one subject whose TSH value was 65.28 filU/mL. If removed from analysis, the mean of the 4th decile became 2.15 pIU/mL and not statistically significant from any other decile. Table 20 presents the mean values adjusted for age, BMI and alcohol using analysis of covariance which includes the outlier in decile 4. The adjusted decile 10 of mean free T4 was statistically significantly lower than that of the first decile (Table 20). Table 21 presents the percent of employees by PFOA decile that were above reference range for these thyroid tests. Adjusted odds ratios and 95% confidence intervals are not presented in a separate table because of the few subjects whose thyroid tests were above or below the reference range as shown in Table 21. Only two of the 18 subjects in Table 21, that had TSH values above the upper reference range, had another thyroid related hormone result out of reference range (Table 22). One individual had a T4 value below reference range (with the highest TSH value in the analysis) and the other was above the reference range for T4 (Table 22). The former 3M Company Page 25 was likely a clinically diagnosed hypothyroid individual (subject 45 in Table 22) who also had a low T4 just above the reference range. Table 23 presents non-adjusted and adjusted coefficients (natural log) of PFOA in the multiplicative models. For all locations combined, there were no statistically significant adjusted coefficients in the models except for free T4 (negative coefficient for PFOA) and T3 (positive coefficient for PFOA). However, the full multiplicative models (4 independent variables) explained only 5 and 2 percent of the variance of free T4 and T3, respectively. There were also positive PFOA coefficients for T3 in the individual Antwerp and Decatur locations (Table 23). Percent variances explained in these models were 9 and 7 percent, respectively. Because so few values were out-of-reference range (Table 21). the results presented in Table 23 are not suggestive of clinically relevant findings as the predicted results from these models (Table 24) were well within normal reference ranges for these thyroid parameters. Predicted serum PFOA concentrations, up to 100 jig/mL in Table 24, were 5 orders of magnitude greater than the average PFOA concentration reported in the general population (approximately 0.005 pg/mL). As shown in Table 24, these predicted PFOA concentrations resulted in TSH, T4, free T4, and T3 all within their normal reference ranges. There was no association observed between metabolic syndrome, as categorized in this study, and PFOA, as shown in Table 24 which presents the number of people defined with metabolic syndrome and non-adjusted and adjusted (age) odds ratios and 95% confidence intervals by PFOA decile. 3M Company Page 26 In addilion to the analyses that focused solely on those male subjects who self-reported they were not taking cholesterol lowering medications as shown in Tables 1 through 24, other analyses examined those subjects (n = 46) who self-reported cholesterol lowering medication usage (See Appendix B, Tables B1 through B4). These 46 subjects were significantly older, and had tended to have higher mean BMI, serum glucose, triglycerides and liver enzyme tests than the 506 subjects who did not report cholesterol lowering medications. As seen previously, the Cottage Grove and Decatur employees were significantly more likely to have higher BMI values than Antwerp employees. Multiplicative models did not indicate, however, PFOA to be associated with the independent variables presented in Table B4, including cholesterol, LDL, HDL and triglycerides. Appendix C (Tables Cl through C9) contains the combined (N = 552) analyses of the 506 male employees who did not report cholesterol lowering medications (Table 1 through Table 24) and the 46 who did (Tables B1 through B4 in Appendix B). Analyses were not substantially different whether they excluded (Tables 1- 24) or included those employees who self-reported cholesterol lowering medications (Tables Cl through C8). For each analysis presented in Appendix C for the 552 male employees, there is a comparable analysis for those employees who did not take cholesterol lowering medications. For example, Table Cl is the PFOA decile distribution analysis for all 552 employees, regardless of cholesterol lowering medication status. The 46 employees who took cholesterol lowering medication are included in the deciles used for those employees who did not take such medication. See Table 7 for the comparable analysis for only 3M Company Page 27 those employees (n = 506) who did not self report cholesterol lowering medications. Findings did not differ. Presented in Tables C2 and C3 are the odds ratios by decile for the lipid out-of-reference range values for all male employee participants (N = 552) regardless of cholesterol lowering medication status. See Tables 12A and 12B for the 506 employees who did not self-report cholesterol lowering medications. Results were comparable. Table C4 presents the odds ratios by PFOA decile for ALT and GGT values out-of-reference range for all male employee participants (N = 552) Comparable analyses for the more restricted subset of 506 employees who did not take cholesterol lowering medications are found in Table Cl. Findings did not differ. Tables C4 through C6 contain both the non-prescribed (n = 506) and all male participants (n = 552) regression coefficients for the lipid (Table C4), liver enzyme (Table C6), and thyroid (Table C7) analyses. Analyses from the multiplicative models were not substantially different whether all male employee participants, or only those who did not self report cholesterol lowering medication usage, are included. The few employees among all participants (n = 552) who had out-of-reference range thyroid tests are presented in Table C8 (see Table 21 for comparison purposes for the 506 employees who did not take cholesterol lowering medications). Table C9 (as was previously done in Table 24 for fable 22) are the predicted thyroid test values for the regression model coefficients that wrerc shown in Table C7. These results in Table C9 again showed that an increase in serum concentrations of PFOA, as high as 100 pg/mL. would not result in out-of reference range thyroid clinical chemistry tests. 3M Company Page 28 PFOA and PFOS serum concentrations were highly correlated among these employee participants of the 2000 fluorochemical medical surveillance program (Appendix A). For purposes of brevity. PFOS-related graphs are presented in Appendix D for the 506 employees who self-reported they did not take cholesterol lowering medications. Also, found in Appendix E are the coefficients from the multiplicative models examining the association for PFOA, as was done previously for PFOA for lipids (Table El), liver enzymes (Table E2) and thyroid tests (Table E3) for the combined three facilities (Antwerp, Cottage Grove, and Decatur) as well as each facility separately. As seen with PFOA, there was a positive association with PFOS and triglycerides which is inconsistent with the known toxicological effects of PFOS. This association was likely due to the fact that higher PFOS concentrations were observed among the Cottage Grove and Decatur employees than Antwerp employees, the latter had the lower serum triglyceride values. No association with PFOS and triglycerides was seen when Cottage Grove and Decatur employees were analyzed separately as shown in Table El. A positive association was observed with triglycerides and PFOS among Antwerp workers only, similar to what has already been reported for PFOA in this subgroup of workers (Olsen et al. 2003a). Inconsistent findings were observed for ALT and PFOS as there was a statistically significant negative association with Antwerp employees and a positive association with Decatur employees (Table E2). Other liver enzyme tests provided no consistent patterns of association. Thyroid related tests also provided no consistent measures of association in the multiplicative models presented in Table E3. 3M Company Page 29 Discussion Based on an analysis of 506 male participants of the 2000 fluorochcmical medical surveillance program offered at the Antwerp. Cottage Grove, and Decatur facilities who self-reported that they did not take cholesterol lowering medications, there was no indication that these employees' serum PFOA concentrations were positively or negatively associated with serum total cholesterol or LDL. A weak negative association was observed with HDL that was likely due to uncontrolled (i.e., residual) confounding, based on lower HDL values observed among the Cottage Grove and Decatur workers than the Antwerp workers and their markedly different demographic factors (e.g., BMI). When the analyses were stratified by location, no statistically significant associations were observed between HDL and PFOA. In another occupational study, Leonard (2005) did not report an association between HDL and serum PFOA concentrations. Neither did Emmett (2004) in a community-based study. To further clarify any possible association with HDL, the A apolipoproteins. which form the major proteins found in HDL, could be measured although this was not pan of the present study. Apolipoprotein A-I wras not associated with serum PFOA of comparable concentrations in a small analysis of Italian production workers (Giovanni Costa, personal communication). Serum triglyceride levels were positively associated with PFOA but it is likely that this association was also due, at least panially, to residual confounding as a consequence of the a disproportionate number of Cottage Grove and Decatur employees with slightly 3M Company Page 30 higher PFOA concentrations than the Antwerp employees. The same positive association was also observed between PFOS and scrum triglycerides which are entirely inconsistent given the well-established hypolipidemia reported in PFOS-related toxicological studies whose concentrations were much higher than the present study (Seacat et al. 2002; 2003). Of the four lipid measurements used in this study, serum triglycerides have approximately 3 times more intraindividual biological variation than either total cholesterol, LDL, or HDL (Stein and Myers 1994). Serum triglycerides may also be influenced by obesity, alcohol intake, and inattention to fasting requirements for blood collection. We hypothesize that the associations observed for serum triglycerides and PFOA in these workers might be the consequence of the non-adherence to fasting requirements by some shift production workers and/or the effect of postprandial metabolic responses in shift workers. Our explanation follows. First, production workers would be more likely to have higher PFOA concentrations than non-production workers (e.g., administrative, laboratory, engineers). This has been shown to be the situation at all three facilities: Antwerp (Olsen et al. 2001c); Cottage Grove (Olsen et al. 2003b) and Decatur (Olsen et al. 200Id; 2003). Second, only production workers engage in shift work. Third, required fasting blood collection occurred only during the morning for all participants. Fourth, it is conceivable that the night shift production workers, compared to day shift production workers as well as non-production workers, could have been less compliant with fasting requirements as well as consumed caffeinated beverages prior to their morning blood collection. Fifth, several studies have indicated postprandial serum triglyceride levels are higher among night shift workers (Al-Naimi et al. 2004; Morgan et 3M Company Page 31 al. 2003; 1998; Karlsson et al. 2001; Lund et al. 2001; Sopowski et al. 2001). Therefore if a subset of subjects (production workers with higher PFOA serum concentrations) who worked night shift were less likely to adhere to the fasting requirements and/or have postprandial metabolic profiles of night shift workers, then a non-causal positive association between PFOA concentrations and serum triglycerides could be observ ed when all subjects are included in the analysis. At the time of data blood collection we did not obtain shift status information and therefore within this database we are unable to further address this methodological question. Countering this possible hypothesis is the fact that blood glucose, also requiring a fasting sample, was not associated with PFOA at any site. However, hyperglycemia has not been consistently associated with night shift workers (Al-Naimi et al. 2004; Karlsson et al. 2003). Also, the association between serum triglycerides and PFOA was not observed at the Cottage Grove site which had both production and non-production workers. Nevertheless, a positive association between serum triglycerides and PFOA was reported by Leonard et al. (2004) whose study population also consisted of production (shift workers) and nonproduction (nonshift workers) participants. To further clarify possible association, if any, between workers' PFOA concentrations and serum triglyceride levels, adjustment for shift work becomes a methodological necessity in the data analyses. Adjusted for BMI and/or serum triglycerides, which are important potential confounders in the analysis of liver enzymes as they are associated with the increasingly prevalent nonalcohol fatty liver disease observed in the United States (Clark et al. 2003), there was no indication that the measured liver enzymes (alkaline phosphatase, AST, ALT, GGT 3M Company Page 32 and total bilirubin) were causally associated with employees' serum PFOA concentrations. A weak positive association between alkaline phosphatase. ALT and GGT with PFOA was observed among the Decatur male employees, however, several epidemiologic research studies of this workforce, past and present, have not reported associations with hepatic disease (malignant or nonmalignant conditions) using a variety of data sources including death certificates (Alexander et al. 2003), episodes of care (Olsen et al. 2004) and self-reports (Alexander and Grice 2006). Neither have other worker populations (Leonard et al. 2004) or a community-based PFOA-exposed population (Emmett 2004) reported positive associations between liver enzyme test results, liver disease and serum PFOA concentrations measured at or below those reported in the present study. For example, in the Emmett study (a presentation is available at http://www.lhwc8studv.org/index.htm). the median serum PFOA concentration measured in this community-based study was 0.34 pg/mL. Individuals older than 60 years had significantly higher levels of PFOA compared to all groups except those less than six years of age. There was no relationship between the blood levels of PFOA and the results for cholesterol, liver function tests (serum protein, albumin, bilirubin, serum alkaline phosphatase, AST, ALT and GGT), kidney function tests (BUN, creatinine) and TSH. Furthermore, there was no relationship between PFOA serum concentrations and being treated for or informed by a physician that a communityparticipant had liver disease (cirrhosis, hepatitis, and any other liver condition). In the present study, there were no consistent associations with TSH, T4, free T4 or T3 with PFOA (or PFOS) across the individual facility locations. For the combined location 3M Company Page 33 analyses, adjusted PFOA coefficients that were statistically significant were the result of models whose range of predictions for serum PFOA concentrations (up to 100 pg/mL) were well-within the normal reference ranges for these thyroid related tests. This was not unexpected as few thyroid-related tests were out-of-reference range. The lack of thyroid related hormone associations is also consistent with the worker data reported by Leonard et al. (2004) and a population-based study by Emmett (2004) that included thyroidrelated conditions (hyperthyroidism, hypothyroidism, goiter). In rats the presence of PFOS or fatty acids such as oleic acid in serum appeared to compete with free T4 for protein bindings, and their presence in serum resulted in a negative bias in analog free T4 measurements (Tanaka et al. 2005). PFOS, however, did not reduce either free T4 by equilibrium dialysis radioimmunoassay (ED-RIA) or the liver response to thyroid hormones [e.g., malic enzyme and uridinediphosphate-glucuronsyltransferase (UGT1 A)]. These observations therefore suggested that prior reports of reduced free T4 in the presence of PFOS may have been artifacts of the analog methods. It is conceivable this negative bias of the assay could also effect measurement of PFOA. 3M Company Page 34 References 3M Company 2003. Health and Environmental Assessment of Perfluorooctane Sulfonic Acid and Its Salts. US EPA docket AR-226-1486. Alexander BH. 2001. Mortality study of workers employed at the 3M Cottage Grove facility. Minneapolis (MN):University of Minnesota. US EPA docket AR-2261030a018. Alexander BH. Olsen GW, Burris JM. Mandel JH. Mandel JS. 2002. Mortality of employees of a perfluorooctanesulphonvi fluoride manufacturing facility. Occup Environ Med 60:722-729. Alexander BH, Grice M. 2006. Self-reported medical conditions in perfluorooctanesulfonyl fluoride manufacturing workers. Minneapolis (MN):University of Minnesota. US EPA docket AR-226. Al-Naimi S, Hampton SM, Richard P, Tzung C, Morgan LM. Postprandial metabolic profiles following meals and snacks eaten during simulated night and day shift work. Chronobiol Int 21:937-947. Brousseau ME, Schaefer EJ, Wolfe ML, Bloedon LT, Digenio AG, Clark RW, Mancuso JP, Rader DJ. 2004. Effects of an inhibitor of cholesteryl ester transfer protein on HDL cholesterol. N Engl J Med 350:1505-1515. Butenhoff J, Costa G, Elcombe C, Farrar D, Hansen K. Iwai H, Jung R, Kennedy G, Lieder P, Olsen G, Thomford P. 2002. Toxicity of ammonium perfluorooctanoate in male cynomolgus monkeys after oral dosing for 6 months. Toxicol Sci 69:244-257. Clark JM, Brancati FL, Diehl AM. 2003. The prevalence and etiology of elevated aminotransferase levels in the United States. Am J Gastroenterology 98:960-967. Collantes R, Ong JP. Younossi ZM. 2004. Nonalcoholic fatty liver disease and the epidemic of obesity. Cleveland Clinic J Med 71:657-664. Costa G. 2004. Letter. US EPA docket AR 226-1868. DePierre JW. 2002. Effects on Rodents of Perfluoro fatty Acids. (In) The Handbook of Environmental Chemistry. Part N. Organofluorines. Berlin:Springer-Verlag (ed Neilson AH). Volume 3, pages 203-248. Emmett E. 2004. Community C8 Study. Little Hocking Water Service Area. Available at http:/Avww.lhwc8 study.org/index.htm. 3M CompanyPage 35 Expert Panel. 2001. Executive Summary of the Third Report of the National Cholesterol Education Program (CNCEP) Expert Panel on Detection. Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 285:2486-2497. Ford ES, Giles WH, Dietz WH. 2002. Prevalence of the metabolic syndrome among US adults. JAMA 287:356-359. Gilliland FD, Mandel JS. 1993. Mortality among employees of a perfluorooctanoic acid production plant. J Occup Med 35:950-954. Gilliland FD, Mandel JS. 1996. Serum perfluorooctanoic acid and hepatic enzymes, lipoproteins and cholesterol: a study of occupationally exposed men. Am J Ind Med 129:560-568. Grundy SM, Brewer HB, Cleeman JI, Smith SC, Lenfant C. 2003. Definition of metabolic syndrome. Circulation 109:433-438. Han X, Snow TA, Kemper RA, Jepson GW. 2003. Binding of perfluorooctanoic acid to rat and human plasma proteins. Chem Res Toxicol 16:775-781. Haughom B, Spydevold O. 1992. The mechanism underlying the hypolipmie effect of perfluorooctanoic acid (PFOA), perfluorooctane sulphonic acid (PFOSA) and cliflbric acid. Biochm Biophys Acta 1128:65-72. Hernn MA, Hemndez-Diaz S, Werler MM, Mitchell AA. 2002. Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology Am J Epidemiol 155:176-184. Kaplan AM. 2004. Ammonium Perfluorooctanoate. US EPA 8EHQ-0904-00373. See also US EPA docket AR226-1867 and AR226-1868. Karlsson B, Knutsson A, Lindahl B. 2001. Is there an association between shift work and having a metabolic syndrome? Results from a population based study of 27,485 people. Occup Environ Med 58;747-752. Karlsson BH, Knutsson AK, Lindahl BO, Alfredsson LS. 2003. Metabolic disturbances in male workers with rotating three-shift work. Results of the WOLF study. Int Arch Occup Environ Health 76:424-430. Kennedy GL, Butenhoff JL, Olsen GW, O'Connor JC, Seacat AM, Perkins RG, Biegel LB, Murphy SR. Farrar DG. The toxicology of perfluorooctanoate. Critical Rev Toxicol 34:351-384. Kerstner-Wood, C, Coward L, Gorman G. 2003. Protein binding of perflurobutane sulfonate, periluorohexane sulfonate, perfluorooctane sulfonate and perfluorooctanoate to 3M Company Page 36 plasma (human, rat, and monkey), and various human-derived plasma protein fractions. Southern Research Institute. Study 9921.7. US EPA docket AR-226-1354. Lakka HM, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo E, Tuomilehto J, Salonen JT. 2002. The metabolic syndrome and total and cardiovascular disease mortality in middle aged men. Leonard RC, Green JW, Krecman KH, Reynolds J, Gilies PS. 2005. Ammonium perfluorooctanoate: cross-sectional surveillance of clinical measures of general health status related to a serum biomarker of exposure. Newark (DE):DuPont Haskell Laboratory for Health and Environmental Sciences. May 31,2005. Lund J, Arendt J, Hampton SM, English J, Morgan LM. 2001. Postprandial hormone and metabolic responses amongst shift workers in Antarctica. J Endocrinol 171:557-564. Mofrad PS, Sanyal AJ. 2003. Nonalcoholic fatty liver disease. Medscape Gen Med 5:113. Morgan L, Arendt J, Owens D, Folkard S, Hampton S, Deacon S, English J, Ribeiro D, Taylor K. 1998. Effects of the endogenous clock and sleep time on melatonin, insulin, glucose, and lipid metabolism. J Endocrinol 157:443-451. Mukai M, Ozasa K, Hayashi K, Kawai K. Various S-GOT/S-GPT ratios in nonviral liver disorders and related physical conditions and life-style. Dig Dis Sci 47:549-555. Nakanishi N, Matsuo Y, Yoneda H, Nakamura K, Suzuki K, Tatara K. 2000. Validity of the conventional indirect methods including Friedewald method for determining serum low-density lipoprotein cholesterol level: comparison with the direct homogeneous enzymatic analysis. J Occ Health 42:130-137. Olsen GW, Gilliland FD, Burlew MM, Burris JM. Mandel JS, Mandel JH. 1998. An epidemiologic investigation of reproductive hormones in men with occupational exposure to periluorooctanoic acid. JOEM 40:614-622. Olsen GW, Burris JM, Mandel JH, Zobel LR. 1999. Serum perfluorooctane sulfonate and hepatic and lipid clinical chemistry tests in fluorochemical production emplovees. JOEM 42:799-806. Olsen GW, Burris JM, Burlew MM, Mandel JH. 2000. Plasma cholecystokinin and hepatic enzymes, cholesterol and lipoproteins in ammonium perfluorooctanoate production workers. Drug Chem Toxicol 23:603-620. Olsen GW, Burlew MM, Burris JM, Mandel JH. 2001a. A cross-sectional analysis of serum periluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) in relation to clinical chemistry, thyroid hormone, hematology and urinalysis results from male and 3M Company Page 37 female employee participants of the 2000 Antwerp and Decatur fluorochemical medical surveillance program. St. Paul (MN):3M Company. U.S .EPA docket AR-226-1047. Olsen GW. Burlew MM, Burris JM, Mandel JH. 2001b. A longitudinal analysis of serum periluorooctanesulfonate (PFOS) and periluorooctanoate (PFOA) levels in relation to lipid and hepatic clinical chemistry test results from male employee participants of the 1994/95. 1997 and 2000 fluorochemical medical surveillance program. St. Paul (MN):3M Company. U.S. EPA docket AR-226-1048. Olsen GW, Schmickler MN, Tierens JM, Logan PW, Burris JM, Burlew MM, Lundberg JK, Mandel JH. 2001c. Descriptive summary of serum fluorochemical levels among employee participants of the Year 2000 Antwerp fluorochemical medical surveillance program. St. Paul (MN):3M Company. U.S. EPA docket AR-226-1030a020b. Olsen GW, Logan PW, Simpson CA, Burris JM, Burlew MM, Lundberg JK. Mandel JH. 200Id. Descriptive summary of serum fluorochemical levels among employee participants of the Year 2000 Decatur fluorochemical medical surveillance program. St. Paul (MN):3M Company. U.S. EPA docket AR-226-1030a020a. Olsen GW, Burris JM, Burlew MM, Mandel JH. 2003a. Epidemiologic assessment of worker serum perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) concentrations and medical surveillance examinations. JOEM 45:260-270. Olsen GW, Butenhoff JL, Mandel JH. 2003b. Assessment of lipid, hepatic and thyroid function in relation to an occupational biologic limit value for perfluorooctanoate. St. Paul (MN):3M Company. US EPA docket AR-226-1351. Olsen GW, Logan PW, Hansen KJ, Simpson CA, Burris JM, Burlew MM, Vorarath PP, Venkateswarlu P, Schumpert JC, Mandel JH. 2003c. An occupational exposure assessment of a perfluorooctanesulfonyl fluoride production site: biomonitoring. AIHA Journal 64:651-659. Olsen GW, Burlew MM, Marshall JC, Burris JM, Mandel JH. 2004. Analysis of episodes of care in a perfluorooctanesulfonyl fluoride production facility. JOEM 46:837846. Rickard RW. Re: PFOA Risk Assessment Science Advisory' Board. (Letter to Dr. Suhair Shallal of US EPA.) New'ark (DE):DuPont Haskell Laboratory for Health and Environmental Sciences. June 29, 2005. Ruhl CE, Everhart JE. 2003. Determinants of the association of overweight with elevated serum alanine aminotransferase activity in the United States. Gastroenterology 124:71-79. 3M Company Page 38 Seacat AM, Thomford PJ, Hansen KJ, Olsen GW, Case MT, Butenhoff JL. 2002. Subchronic toxicity studies on perfluorooctanesulfonate potassium salt in cynomolgus monkeys. Toxicol Sei 68:249-264. Seacat AM, Thomford PJ, Hansen KJ. Clemen LA, Eldridge SR. Elcombe CR. Butenhoff JL. 2003 Sub-chronic dietary toxicity of potassium perfluorooctanesulfonate in rats. Toxicology 183:117-131. Sopowski MJ, Hampton SM, Ribeiro DC, Morgan L, Arendt J. 2001. Postprandial triacylglycerol responses in simulated night and day shift: gender differences. J Biol Rhythms 16:272-276. Stein EA, Myers GL. 1994. Lipids, lipoproteins, and apolipoproteins. (In) Tietz Textbook of Clinical Chemistry, (eds Burtis CA, Ashwood ER). Philadelphia (PA): W.B. Saunders Co. pp 1002-1093. Tanaka S, Thibodeaux J, Eastvold M, Bjork J, Ehresman D, Butenhoff J. 2005. Measurement of thyroid hormones in rat sera containing perfluorooctanesulfonate (PFOS). Toxicologist 840S-1) 423 (abstract 2066). Ubel FA, Sorenson SD, Roach DE. 1980. Health status of plant workers exposed to fluorochemicals - a preliminary report. Am Ind Hyg Assoc J 41:584-589. Xie Y, Yang Q, Nelson BD, DePierre JW. 2003. The relationship between liver peroxisome proliferation and adipose tissue atrophy induced by peroxisome proliferators exposure and withdrawal in mice. Biochem Pharmacol 66:749-756. Table 1. Number (percent) of Male Employees Who Self-Reported Cholesterol Lowering Medications. By Location, 2000 Fluorochemical Medical Surveillance Location Antwerp Cottage Grove Decatur Total Choleseterol Lowering Medication Yes (%) No (%) 10(5) 196 (95) 9(7) 122 (93) 27 03) 188 (87) 46 (8 ) 506 (92) Total 206 131 215 552 Table 2. Mean, Standard Deviation, Median and Range of PFOA, PFOS, Demographic Factors and Clinical Chemistry Results, By Location, 2000 Fluorochcmical Medical Surveillance Program Antwerp (N-196') Mean SD Median Range Cottage Grove fN=122) Mean SD Median Range Decatur fN=188) Mean SD Median Range PFOA 1.0 2 2 1.06 PFOS 0.953 0.97 Age 372,3 8 BMI 24.72,3 3.0 Alcohol 1.1" 1.1 Glucose 842 3 17 Cholesterol 217 41 LDL 1392 37 HDL 552'3 15 Triglycerides 1202,3 83 Aik Phos 602-3 14 AST 232,3 6 ALT 232,3 10 AST/ALT 1.042,3 0 .1 2 0.65 0.01-7.04 0.55 0.04-6.24 36 21-58 24.5 17.5-34.7 0.9 0.0-6.4 87 31-131 216 105-331 134 43-235 53 26-121 100 34-731 61 21-113 22 13-58 21 8-71 1.02 0.81-1.47 4.631' 3 12.53 0.95 0.01-92.03 1.892 1.61 1.51 0.04-12.70 0 .8 6 3 0.98 0.45 0.03-4.79 1.291,2 0.92 1.00 0.06-4.17 411 9 40 24-67 421 9 43 26-63 29.91' 3 4.8 29.5 19.7-52.1 28.61' 2 4.4 27.5 17.2-50.1 0.71' 3 0.7 1 0 0 1-3 23 0.5 0.0-3.0 95 59-259 o . i 1-2 0.3 0.0 0 .0 -2 .0 93 U 14 91 74-184 2 1 0 39 206 130-311 214 41 211 121-319 130' 32 125 37-208 136 36 133 47-225 461 11 45 18-77 441 10 42 24-82 187' 139 142 24-725 182' 110 164 32-796 651' 3 15 63 28-107 731,2 2 0 69 26-160 251 8 24 10-54 26' 8 25 13-69 341 17 29 13-95 341 16 30 6-103 0.941 0.09 0.92 0.70-1.25 0.95' 0.11 0.93 0.75-1.82 Table 2 (Continued) Antwerp O M 96) Mean (SD) Med Range Cottage Grove (N=122) Mean (SD) Med Range Decatur (N=188) Mean (SD) Med Range GGT Total Bil Direct Bil TSH T4 Free T4 T3 232,3 17 1.0 2,3 0.3 0 .13 0.05 2 .0 3 1.6 8.2 1.4 i.i3 0.2 1312-3 19 18 6-111 1.0 0.4-2.3 0.1 0.0-0.4 1.7 0.03-19.4 8.2 4.2-12.0 1.1 0 .6 - 1.6 130 87-85 31' 32 0.91'3 0.3 0.1 0 .0 2 2.4 14 7.93 1.1 1.1 0.1 125* 30 22 7-314 0.8 0.4-2.3 0.1 0 .0 -0 .2 2.0 0.03-9.4 7.9 4.9-11.6 1.1 0 .8 - 1.8 121 78-300 30' 17 27 7-144 0.71' 12 0 .2 0.7 0.3-1.5 0 .11 0.07 0.1 0.0-0.7 2 .8 1 5.2 1.9 0.03-65.3 8.42 1.4 8.5 4.6-11.4 1.1' 0 .2 1.1 0 .6 - 1.5 125' 22 120 86-196 1. Statisticallysignificantly (p < .05) different than Antwerp 2. Statistically significantly (p < .05) different than Cottage Grove 3. Statistically significantly (p < .05) different than Decatur Tabic 3. Number and Percent of Employees by Location for Reference Points of Demographic Factors and Clinical Chemistry Results, 2000 Fluorochemical Medical Surveillance Program Antwerp 196} N (%) Cottage Grove (N==122) N (%) Decatur =188) N (%) p value* BMI > 30 Alcohol > 1drink/day 9 (5) 92 (47) 54 (44) 49 (40) 61 (32) 2 (1) .0001 .0001 Glucose > 110 mg/dL 10 (5) 22 (18) 16 (9) .001 Cholesterol > 200 mg/dL 119 (61) 69 (57) 114 (61) .72 Cholesterol > 240 mg/dL 60 (31) 32 (26) 50 (27) .60 HDL < 40 mg/dL LDL> 130 mg/dL 16 (8) 106 (55) 36 (30) 53 (48) 72 (38) 100 (55) .0001 .40 Triglycerides > 150 mg/dL 46 (23) 58 (48) 103 (55) .0001 Aik Phos> 120IU/L AST > 50 IU/L ALT > 50 1U/L GGT > 50 IU/L Total Bil > 1.5 mg/dl Direct Bil > 0.4 mg/dl TSH <0.35 pIU/mL >5.5 pIU/mL T4 <4.5 pg'dL >12.0 pg'dL 0 (0) 1 (1) 2 0) 13 (7) 13 (7) 0 (0) 1 (1) 5 (3) 1 (1) 0 (0) 0 (0) 4 (3) 20 (16) 15 (12) 1 (1) 0 (0) 1 0) 5 (4) 0 (0) 0 (0) 5 (3) 2 (1) 22 (12) 16 (9) 0 (0) 2 (1) .01 .11 .0001 .22 .0001 .18 1 (l) .92 8 (4) .60 0 (0) .46 0 (0) 1.0 Free T4 <0.70 ng/dL >1.53 ng/dL T3 < 60 ng/dL >181 ng/dL "Metabolic Syndrome"1 1 (0 3 (2) 0 (0) 2 0) 2 0) 0 (0) 1 (1) 0 (0) 3 (3) 24 (20) 2 (1) .49 1 (1) .60 0 (0) 1.0 2 (1) .46 25 (13) .0001 chi square test of significance 1. See text for description Tabic 4. Mean, Standard Deviation, Median and Range o f PFOA, PFOS, Demographic Factors and Clinical Chemistry Results, By PFOA Quartile, 2000 Medical Fluorochemical Surveillance Program for Antwerp (N = 196) PFQA PFOS Quartile I (N=49) Mean SD Median Range o . n 2-3-4 0.07 0.10 0.01-0.25 Quartile 2 (N=49) Mean SD Median Range 0.4 l a 4 0.10 0.40 0.26-0.64 Quartile 3 (N-49) Mean SD Median Range 1.07' 24 0.24 1.10 0.67-1.53 0.34M 0.37 0.25 0.04-2.19 0.803,4 0.88 0.52 0.25-5.19 1.071,4 0.57 1.03 0.06-2.35 Quartile 4 (N-49) Mean SD Median Range 2 491.2.3 1.06 2.23 1.56-7.04 1 59'. 2. 3 1.32 1.25 0.18-6.24 Age 40* * 4 10 40 23-58 36' 7 36 25-53 36' 8 35 21-53 35' 7 35 22-50 BMI Alcohol Glucose 25.0 3.1 1.0 0.9 9,2. m 15 24.6 19.2-34.7 0.7 0.0-4.3 92 60-131 25.23 1.0 82` 3.0 24.9 20.2-33.9 24.12 1.2 0.7 0.0-6.4 1.2 18 85 48-115 811 3.1 24.2 17.5-32.3 1.1 1.0 0.0-5.0 20 86 31-120 24.5 1.2 82' 2.6 24.2 19.5-30.9 1.1 1.1 0.0-4.3 16 84 49-117 Choi 220 42 215 140-331 211 40 216 122-286 213 43 213 105-300 225 40 221 145-316 LDL 145 36 142 68-224 131 34 129 59-205 136 40 128 43-235 144 37 141 54-226 HDL 54 13 55 26-96 57 20 53 29-121 56 12 53 37-88 53 12 51 36-85 Trig 105'* 62 87 37-348 123 104 99 36-731 108 53 198 41-287 145' 98 128 34-546 Aik Phos 61 AST 24 15 63 30-96 6 22 13-36 59 13 58 36-94 243 7 24 13-49 56' 2,2.4 13 58 21-84 4 21 15-31 643 16 64 34-113 253 8 24 15-58 ALT 24 AST/A I T 1.02 10 21 10-61 0.10 1.01 0.81-1.23 23 1.03 8 21 11-45 0.1 1 1.02 0.83-1.37 204 1.06 8 19 8-46 0.14 1.02 0.87-1.42 243 1.04 12 22 9-71 0.14 1.01 0.85-1.47 GGT 22 14 18 8-80 25 20 20 6-111 20 12 17 7-53 Total Bil 1.0 0.3 1.0 0.5-2.0 1.0 0.4 1.0 0.5-2.3 1.0 0.3 1.0 0.5-2.2 Direct Bil 0.1 0.04 0.1 0.0-0.2 0.1 0.06 0.1 0.0-0.3 0.1 0.06 0.1 0.0-0.4 26 20 22 8-89 1.0 0.3 1.0 0.4-2.0 0.1 0.03 0.1 0.0-0.2 TSH 1.8 1.1 1.7 0.4-5.4 1.8 0.9 1.6 0.03-4.3 1.9 1.1 1.8 0.5-6.1 2.4 2.7 1.8 0.6-19.4 Quartile I (N-49) Mean SD Median Range Tabic 4. (Continued) ______ Quartile 2 (N=49) Mean SD Median Range Mean T4 Free T4 T3 8.4 1.5 i.l 0.2 1251 18 8.5 5.0-11.5 1.1 0.9-1.5 124 91-166 8.1 1.2s 130 1.4 7.9 4.2-12.0 0.2 1.2 0.9-1.5 19 131 87-171 8.0 I.l2 132 1. Statistically significantly (p < .05) different than 1*' quartile 2. Statistically significantly (p < .05) different than 2ml quartile 3. Statistically significantly (p < .05) different than 3"' quartile 4. Statistically significantly (p < .05) different than 4,h quartile Quartile 3 0 ^ 4 9 ) SD Median Range 1.4 8.0 5.0-11.1 0.2 l.l 0.6-1.6 20 132 97-185 ______ Quartile 4 (N=49) Mean SD Median Range 8.3 1.5 8.5 4.7-10.7 1.1 0.1 1.1 0.8-1.6 135' 18 132 102-184 Table 5. Mean, Standard Deviation, Median and Range o f PFOA, PFOS, Demographic Factors and Clinical Chemistry Results, by PFOA Quartile, 2000 Fluorochcmical Medical Surveillance Program for Cottage Grove (N = 122) PFQA Mean 0.I24 Quartile I (N =3h SD Median Range 0.08 0.11 0.01-0.30 Quartile 2 (N"28) Mean SD Median Range 0 .5 14 0.17 0.46 0.30-0.87 Mean 1.60' Quartile 3 (N=33) SD Median Range 0.47 1.67 0.91-2.49 Quartile 4 (N=30) Mean SD Median Range 16.50*'2,3 21.48 6.64 2.51-92.03 PFOS Age 0.622 0.83 0.15 0.02-2.64 373 8 37 24-61 1.18* 1.22 0.83 0.12-4.79 393 9 38 26-61 0.90 0.92 0.55 0.02-4.22 45'.2.4 9 44 27-67 0.78 413 0.88 0.48 0.12-4.31 9 41 28-59 BMI 29.4 5 28.8 19.7-40.1 30.1 4.7 29.2 23.4-39.9 29.9 5.6 29.1 23.6-52.1 30.0 3.6 30.6 22.0-37.3 Alcohol 0.5 0.5 0.5 0.0-2.0 0.8 0.8 0.5 0.0-3.0 0.7 0.8 0.5 0.0-3.0 0.7 0.6 0.5 0.0-2.0 Glucose 101 31 95 73-259 99 28 92 77-222 102 13 103 79-143 97 14 95 59-126 Choi 207 38 196 146-278 214 40 210 142-289 206 34 205 151-293 214 44 205 130-311 LDL 128 32 125 58-189 133 31 124 80-201 129 32 118 82-208 129 35 135 37-180 HDL 47 11 46 29-77 46 II 44 23-69 46 11 46 25-68 43 10 44 18-62 Trig 168 115 132 24-503 180 124 155 55-542 174 129 136 50-715 227 179 159 38-725 Aik Phos 65 11 65 44-90 65 15 63 28-101 66 17 63 32-107 62 15 63 37-100 AST 274 10 25 14-52 26 7 25 15-47 25 10 23 10-54 23' 5 22 14-38 ALT 35 17 30 14-84 37 20 30 16-95 33 19 27 13-87 32 11 29 15-56 AST/ALT 0.96 0.09 0.95 0.84-1.25 0.93 0.09 0.93 0.75-1.10 0.94 0.09 0.91 0.80-1.24 0.92 0.09 0.92 0.70-1.13 GGT 37 56 22 7-314 32 22 22 10-97 27 16 24 7-83 28 16 24 10-84 Total Bi 1 0.9 0.2 0.8 0.4-1.5 0.9 0.3 0.9 0.6-2.3 0.8 0.2 0.8 0.4-1.4 0.9 0.3 0.8 0.4-1.5 Direct Bil 0.1 TSH 2.4 0.03 0.1 0.0-0.2 1.1 2.2 0.8-5.0 0.1 0.03 0.1 0.1-0.2 2.3 1.2 2.0 0.9-6.0 0.1 0.0 0.1 0.1-0.1 2.2 1.3 1.9 0.6-7.5 0.1 2.5 0.02 0.1 1.8 2.0 0.0-0.1 0.03-9.4 Quartilc 1__________ Mean SD Median Range Table 5 (Continued) ______ Quartilc 2 Mean SD Median Range Mean T4 8.1 1.4 8.2 5.1-11.6 8.0 1.2 7.9 4.9-10.0 Free T4 1.1 0.2 1.1 0.8-1.8 l.l 0.1 1.1 0.9-1.4 T3 123 34 119 78-277 123 19 123 87-158 1. Statistically significantly (p < .05) different than 1Mquartilc 2. Statistically significantly (p < .05) different than 2ml quartilc 3. Statistically significantly (p < .05) different than 3rd quartilc 4. Statistically significantly (p < .05) different than 4lh quartilc 7.9 1.1 127 Quartile 3__________ SD Median Range 0.8 7.8 6.7-9.7 0.1 l.l 0.9-1.4 30 124 85-270 ______ Quartile 4__________ Mcan SD Median Range 7.7 1.0 7.8 5.6-9.8 l.l 0.1 1.1 0.8-1.3 123 36 118 81-300 Table 6. Mean, Standard Deviation, Median and Range of PFOA, PFOS, Demographic Factors and Clinical Chemistry Results, by PFOA Quartile, 2000 Fluorochcmical Medical Surveillance Program for Decatur (N = 188) PFOA Quartile I (N=48) Mean SD Median Range 0.362' *' 0.18 0.36 0.04-0.69 ______ Quartile 2 (N~51) Mean SD Median Range 1.2 f* 3'4 0.24 1.24 0.73-1.59 ______ Quartile 3 (N=42) Mean SD Median Range 2.151,2,4 0.32 2.15 1.61-2.70 ______ Quartile 4 (N=47) Mean SD Median Range 3.97'-2,1 1.69 3.67 2.72-12.70 PFOS 0 .5 12-3' -110.32 0.45 0.06-1.64 1.033,4 0.51 0.98 0.18-2.53 I.32,A4 0.75 1.18 0.29-3.55 2.351' 2,3 0.85 2.42 0.67-4.17 Age 4 4 * 11 46 27-63 434 8 45 26-61 41 8 43 26-57 391' 2 9 38 27-60 BMI 27.9 4.1 27.0 21.7-40.8 29.4 4.9 27.6 24.0-50.1 28.8 3.5 28.2 22.7-35.1 28.5 4.9 28.3 17.2-45.5 Alcohol 0.2 0.4 0.0 0.0-2.0 0.1 0.3 0.0 0.0-0.8 0.1 0.2 0.0 0.0-0.8 0.1 0.2 0.0 0.0-0.8 Glucose 92 14 90 75-166 95 18 91 74-184 96 12 94 75-135 91 12 87 75-129 Cholesterol 204 ' 41 197 137-305 224' 42 223 146-308 214 42 211 121-297 214 38 208 151-319 LDL I272 35 122 76-205 147' 37 148 80-222 133 39 140 47-199 135 30 133 69-225 IIDL 46 12 43 29-82 43 9 42 24-70 44 10 42 29-68 42 8 41 28-75 Trig I522 82 139 38-399 209' 146 155 32-633 183 67 175 85-379 183 115 166 39-796 Alk Phos 683 18 69 26-117 74 20 68 52-160 77' 24 73 39-142 74 19 68 44-122 AST 26 8 24 15-48 25 7 24 13-51 26 6 26 16-42 27 10 25 15-69 ALT 3O4 15 26 12-91 30* 14 29 17-103 35 13 34 10-63 401,2 19 33 6-99 AST/ALT 0.984 0.10 0.97 0.82-1.30 0.95 0.09 0.95 0.75-1.11 0.94 0.09 0.93 0.76-1.23 0.92' 0.15 0.89 0.78-1.82 GGT 283 18 25 7-119 263 13 21 11-87 361'2 22 31 12-144 32 14 29 10-80 Total Bi 1 0.8' 0.2 0.8 0.3-1.5 0.8 0.2 0.7 0.4-1.2 0.7 0.2 0.7 0.4-1.2 0.71 0.2 0.7 0.4-1.3 Direct Bi 1 0.1 0.06 0.1 0.0-0.2 0.1 0.06 0.1 0.0-0.2 0.1 0.10 0.1 0.0-0.7 0.1 0.09 0.1 0.0-0.6 TSH 3.5 9.3 1.9 0.03-65.3 2.6 2.6 1.9 0.7-18.8 2.3 1.6 1.8 0.2-8.6 2.7 3.3 2.2 1.0-23.2 Table 6 (Continued) Quartile 1__________ Mean SD Median Range ______ Quartile 2___________ Mean SD Median Range T4 8.4 1.5 8.7 4.6-11.0 Free T4 1.1 0.1 1.1 0.6-1.3 T3 1194 19 116 86-180 8.4 l.i 124 1.4 8.3 5.9-11.4 0.2 1.1 0.8-1.4 20 119 93-186 ______ Quartile 3___________ Mean SD Median Range 8.2 1.6 8.6 3.3-11.1 1.1 0.2 1.1 0.4-1.5 125 23 120 93-196 ______ Quartile 4 Mean SD Median Range 8.6 1.4 8.5 5.1-11.4 1.1 0.2 1.1 0.8-1.3 131' 23 128 87-177 1. Statistically significantly (p < .05) different than 1Mquartile 2. Statistically significantly (p < .05) different than 2'*1quartile 3. Statistically significantly (p < .05) different than B"1quartile 4. Statistically significantly (p < .05) different than 4,h quartile Table 7. Number of Employees. Mean, 95% Confidence Interval, Median and Range by PFOA Decile Distribution , 2000 Fluorochemical Medical Surveillance Program for Antwerp, Cottage Grove and Decatur (N = 506) PFOA Decile 1 2 4 5 6 7 8 9 10 N 51 51 51 50 51 49 50 54 49 50 _______________________ PFOA Mean 95% C.I. Median 0.068' 10 0.05-0.07 0.06 0 .2 0 9' 10 0.19-0.21 0.19 0.369-10 0.35-0.37 0.36 0.559,10 0.53-0.57 0.54 0.9110 0.87-0.94 0.91 1.2610 1.23-1.28 1.25 1.6410 2 i 7 U 1.60-1.67 2.12-2.23 1.63 2.18 3.0m,i 2.90-3.10 2.96 12.151'9 7.21-17.10 4.94 Range 0.007-0.13 0.13-0.29 0.30-0.44 0.44-0.71 0.72-1.10 1.11-1.40 1.42-1.85 1.86-2.50 2.51-3.69 3.71-92.03 M0 Statistically significantly (p < .05) different than PFOA decile(s) 1,2, 3 . . . and/orlO Tabic 8. Distribution of Demographic Factors by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program (N = 506) PFOA Location Decile Antwerp Cottage Grove Decatur Age BMI BMI >=30 Mean (SD) Mean (SD) N (%) Drinks/dav Mean (SD) >= 1 Drink/dav N (%> 1 28 (55) 17(33) 6 (12) 39 (11) 27.1 (4.8) 13(25) 0.7 (0.7)9 21 (41) 2 25 (49) 13(26) 13(25) 41(9) 26.7 (4.1)9 7 (14) 0.7 (0.9)9 17 (33) 3 28 (55) 11 (2 2 ) 4 20 (40) 13(26) 5 22 (43) 11 (2 2 ) 6 21 (43) 7 (14) 7 20 (40) 9 (18) 8 18(33) 11 (2 0 ) 9 9 (18) 7 (14) 12(23) 17(34) 18(35) 21 (43) 21 (42) 25 (46) 33 (67) 38 (8 ) 38(10) 40(10) 40 (9) 39 (9) 41 (10) 39 (9) 26.4(3,9)9'10 8 (16) 27.8(4.3) 13 (26) 27.2(5.0) 8 (17) 27.4(4.8) 12(24) 26.9(3.9)' 1 0 (2 0 ) 27.5(5.4) 17(31) 28.9(4.9)2,3,2 21 (43) 0.7 (1.0)9 15 (29) 0.7 (1.0)* 13 (26) 0.7 (1.0)9 16 (32) 0.7 (1.0)9 14 (29) 0.7 (0.9)9 14 (28) 0.6 (0.9)9 16 (30) 0.3 (0.5) 1' 8 5 ( 10) 10 5 (10) 23 (46) 22 (44) 39(9) 28.3(4.0)3 15(30) 0.5 (0.7) 12 (24) D<.000l" I X .0312 1 10 Statistically significantly (p < .05) different than PFOA decile 1,2, 3 . . . 10. _ P < ,I413 n `l2, n p value associated with chi square test for location, BM1 distribution and > 1 drink/day, respectively. Table 9. Mean. 95% Confidence Interval, Median and Range of PFOS Serum Concentrations by PFOA Decile Distribution, 2000 Fluorochemical Medical Surveillance Program (N = 506) PFOA Decile N _________________________ PFOS Mean 95% C.I. Median Range 1 51 0.303' 10 0.15-0.46 0.13 0.02-2.64 2 51 0.616-10 0.38-0.84 0.35 0.02-5.19 3 51 0 .8 6 7' 10 0.59-1.13 0.52 0.13-4.79 4 50 0.787' 10 0.62-0.94 0.58 0.12-3.41 5 51 0.891,7*10 0.74-1.04 0.82 0.06-2.21 6 49 | 121.2,9, 10 0.90-1.33 0.98 0.02-4.22 7 50 1.271'5,9 1.07-1.48 1.21 0.18-3.25 8 53 I.411'5,9 1.10-1.73 1.03 0.20-6.24 9 49 1.741' 8 1.43-2.06 1.83 0.12-4.86 10 50 1.58 1 1.23-1.94 1.06 0.14-4.31 110 Statistically significantly (p < .05) different than PFOA decile(s) 1, 2, 3 . . . and/or 10 Table 10. Mean and Standard Deviation of Serum Lipid Clinical Chemistry Results by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program PFOA Decile 1 2 3 4 5 6 7 8 9 10 Cholesterol Mean (SD) 215(43) 213(42) 208 (37) 210(41) 218(39) 219(42) 214(40) 216(43) 218(40) 214(42) LDL Mean (SD) 137(39) 137 (33) 128 (31) 132 (34) 141 (35) 141 (41) 133 (37) 136(40) 138(31) 132 (35) HDL Mean (SD) 50 (13)10 52 (13)9' 10 52 (14)9-10 50 (19)10 49 (12) 10 49 (12) 10 50(H )910 47(13) 45 ( l l )2' 3-7 43 (9) 1-7 Triclvcerides Mean (SD) 142 (101) 10 1 2 0 (6 6 ) 8' 10 144 (121) 10 147 (101) 10 161 (117)10 161 (104)10 154 (106)' 175 (109)2 174 (93)2 214 (168)1*7 M Statistically significantly (p < .05) different than PFOA decile(s) 1, 2, 3 . . . and/or 10 Tabic 11. Adjusted* Mean and 95% Confidence Intervals for Lipid Clinical Chemistry Results by PFOA Decile PFOA Decile Cholesterol Mean 95% Cl Mean LDL 95% Cl Mean HDD 95% Cl 1 214 203-225 137 127-147 5010 46-53 2 211 199-222 135 125-145 5110 48-54 3 209 198-220 128 118-138 5110 48-54 o O 4 210 199-222 133 123-143 46-53 5 217 206-228 140 130-150 48 45-51 6 218 206-229 141 130-151 48 45-51 7 214 203-225 133 123-143 5010 47-53 8 215 204-226 136 126-146 47 44-50 9 221 210-233 140 130-150 48 44-52 10 216 204-227 133 123-144 44 M '7 41-47 Adjusted for age, BMI and alcohol using analysis of covariance M0 Statistically significantly (p < .05) different than PFOA decilc(s) 1,2, 3 . . . and/or 10 Triglvcerides Mean 95% Cl 14510 124s'10 15310 145' 162 16010 15810 1722 16510 208uw 116-173 95-153 124-182 116-175 133-191 131-190 128-187 144-201 135-194 179-238 Table 12A. Adjusted1Odds Ratios (O.R.) and 95% Confidence Interval (95% C.I.) for Lipid Clinical Chemistry Reference Points, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program PFOA Choi >= 200 me/dL Choi >= 240 me/dL LDL >== 130 me/dL Decile O.R. 1 95% C.I. O.R. 1 95% C.I. O.R.' 95% C.I. 1 1.0 - 2 0.4 0 .2 - 1.0 1.0 1.3 0.5-3.1 1.0 - 0.7 0.3-1.6 3 0.9 0.4-2.0 4 0.9 0.4-2.1 0.9 0 .3-2.2 1.1 0.4-2.7 0.7 0.3-1.6 0 .8 0.4-1.8 5 1.7 0.7-4.0 1.6 0.7-3.9 1.4 0.6-3.1 6 1.0 0.4-2.2 7 0.8 0.4-1.9 1.1 0.5-2.8 1.2 0.5-2.9 0.9 0.4-2.0 0.7 0.3-1.5 8 1.0 0.4-2.2 9 1.4 0.6-3.3 10 1.1 0.5-2.6 1.2 0.5-2.9 1.5 0.6-3.7 1.0 0.4-2.5 1.1 0.5-2.4 1.2 0.5-2.7 1.2 0.5-2.8 1. Adjusted for age, BMI and alcohol HDL<= 40 me/dL Triglycerides >= 150 mg/dl O.R. 1 95% C.I. O.R. 1 95% C.I. 1.0 - 1.0 - 1.1 0.4-3.2 0.7 0.3-1.8 0.5 0.1-1.5 1.0 0.4-2.4 2.0 0.8-5.4 1.0 0.4-2.9 1.7 0.7-4.7 1.3 0.5-3.1 1.2 0.5-3.0 1.7 0.7-4.0 0 .6 0 .2 - 1.8 0.9 0.4-2.2 1.4 0.5-3.7 2.7 1.2-6.5 0.9 0.3-2.4 2.4 1.0-5.9 2 .6 1.0 -6.8 2.4 1.0-5.8 Tabic 12B. Nonadjustcd and Adjusted Odds Ratios (O.R.) and 95% Confidence Intervals (95% C.l.) for IIDL (< 40 mg/dL) and Triglycerides (>= 150 mg/dL) by PFOA Decile, 2000 Fluorochcmical Medical Surveillance Program HDL < 40 me/dL_____________ ____________Triclvcerides >= 150 mg/dL Decile O R .1 95% C.l. O R .2 95% C.l O R .1 95% C.l . O R .1 95% C.l. 1 1.0 - 1.0 - 1.0 - 1.0 - 2 l.l 0.4-3.0 0.9 0.3-2.7 0.7 0.3-1.6 0.6 0.2-1.5 3 0.4 0.1-1.4 0.4 0.1-1.2 0.8 0.4-1.9 0.8 0.3-1.9 4 2.1 0.9-5.4 1.6 0.6-4.4 1.3 0.6-3.0 l.l 0.5-2.6 5 l.l 04-3.0 0.8 0.3-2.3 1.2 0.5-2.7 1.0 0.4-2.4 6 1.8 0.7-4.7 1.3 0.5-3.7 1.6 0.7-3.7 1.4 0.6-3.3 7 0.7 0.2-1.9 0.4 0.1-1.3 0.9 0.4-2.0 0.7 0.3-1.6 8 1.6 0.6-4.0 1.0 0.4-2.7 2.5 1.1-5.6 2.0 0.9-4.6 9 1.3 0.5-3.5 0.6 0.2-1.8 2.7 1.2-6.1 1.7 0.7-4.I 10 3.0 1.2-7.5 1.5 0.6-4.0 2.5 l .l -5.8 1.6 0.7-3.7 'Not adjusted `Adjusted for location Table 13. Non-adjusted and Adjusted Ln PFOA Coefficients for Ln Lipid Clinical Chemistry Result, 2000 Fluorochemical Medical Surveillance Program Ln PFOA Non-adjusted Coefficient SE p value Ln PFOA Adjusted1 Coefficient SE p value Ln Cholesterol All Locations 0.0059 Antwerp 0.0051 Cottage Grove 0.0034 Decatur 0.0221 0.0060 0.0106 0.0089 0.0139 .32 .63 .70 .11 Ln LDL 0.0076 0.0130 0.0021 0.0266 0.0059 0.0096 0.0100 0.0141 .20 .18 .83 .06 All Locations 0 .0 0 1 2 0.0089 .89 0.0021 Antwerp -0.0037 0.0157 .81 0.0106 Cottage Grove -0 .0 0 2 2 0.0139 .87 0.0049 Decatur 0.0258 0.0199 .20 0.0302 Ln HDL All Locations -0.0307 0.0079 .0001 -0.0183 Antwerp -0.0057 0.0136 .68 -0.0095 Cottage Grove -0.0153 0 .0 1 2 2 .21 -0.0192 Decatur -0.0256 0.0149 .09 -0.0207 Ln Trielvcerides All Locations 0.0892 0.0185 .0001 0.0711 Antwerp 0.0840 0.0288 .004 0.0980 Cottage Grove 0.0343 0.0316 .28 0.0280 Decatur 0.0715 0.0400 .08 1. See study methods. Adjusted for Ln Age. Ln BMI. Ln Alcohol 0.0689 0.0090 0.0147 0.0145 0 .0 2 0 0 0.0069 0.0131 0.0120 0.0141 0.0169 0.0270 0.0314 0.0376 .81 .47 .73 .13 .01 .47 .11 .14 .0001 .0004 .38 .07 Table 14. Mean and Standard Deviation of Hepatic Clinical Chemistry Results, by PFOA Decile, 2000 Medical Fluorochemical Surveillance Program PFOA Decile 1 2 3 4 5 6 7 8 9 10 Alk Pos Mean (SD) AST Mean (SD) 65(13) 26 (8)s' 59 (15)7' 9' 10 25 (7) 64 ( 16)7,9 26 (7)5' 6 65(15) 24 (8) 64 (21)7,9 22 (6)1-3,7,8,9 66(19) 22 (5)u -7-8-9 72 ( 17)2,3' 5 26 (7)5' 6 65(17) 26 (9)5,6 72 (21)2-3 S 26 (9)s-6 68 (20)2 24(7) ALT Mean (SD) 29 (15)9' 10 28 (15)9,10 29 (14)9-10 28 (15)9' 10 27 (15)9-10 2 4 (9)79, m 31 (15)6 29 (16)s 36 (14)'-6 8 35 (17)1'6 AST/ALT Mean (SD) GGT Mean (SD) Total Bilirubin Mean (SD) 0.99 (O.l)9'10 32 (45)6 1.00 (O.l)9'10 25(18) 1.00 (O.l)9'10 26 (20) 0.98 (0.1)' 28(20) 0.98 (0.1)' 24 (13)9 1.01 (O.l)9'10 22 (10)1-9 0.98 (0.1)' 28(19) 1.00 (O.l)910 29 (22) 0.93 (0.1)1'3,6,8 33 (18)s' 6 0.93 (0.2)1'8 30(16) 0.99 (0.3) 0.99 (0.3) l.O6' 10 (0.3) 0.99 (0.3) 0.99 (0.3) 0.93'9 (0.3) 0.93,9 (0.3) 0.83 (0.2) 0.71-7 (0.3) 0.813 (0.3) Direct Bilirubin Mean (SD) O.l9 (0.04) 0.1 (0.04) 0.1 (0.06) 0.1 (0.04) O.l9 (0.06) 0.1 (0.06) 0.1 (0.03) O.l9 (0.09) O.l1-5-8 (0.04) 0.1 (0.08) M0 Statistically significantly (p < .05) different than PFOA decile(s) 1, 2, 3 . . . and/or 10 Table 15A. Adjusted* Mean and 95% Confidence Intervals for Hepatic Clinical Chemistry Results by PFOA Decile Adjusted PFOA Decile Aik Phos AST ALT GOT Mean 95% Cl Mean 95% Cl Mean 95% Cl Mean 95% Cl Total Direct Bilirubin Bilirubin Mean 95% Cl Mean 95% Cl 1 66 61-71 26s6 24-28 29 25-33 32 27-38 2 5 9 7 .9 .I0 54-64 25 23-27 29 25-33 25 19-31 3 647 59-69 26s'6 24-28 30 26-34 27 21-33 4 65 61-70 24 22-26 289,10 24-32 28 22-34 5 647 60-69 22'Am 20-25 2 7 9 .1 23-31 249 18-30 6 66 61-71 22l'3,7'9 20-24 257,9,i 21-28 2 l ` 15-27 7 7 2 2-3'5'8 67-77 26s'6 24-28 3 ,6 27-35 28 22-34 8 657 60-70 265,6 24-28 29 26-33 29 23-35 9 70 66-75 265,6 24-28 344-'1 30-38 33 5,6 27-39 10 683 63-72 24 22-26 344* 30-38 30 24-36 *Adjusted for Age, BMI and Alcohol using analysis of covariance M0 Statistically significantly (p < .05) different than PFOA decile(s) 1, 2, 3 . . . and/or 10 0 .9 9 0.86-1.02 0.1 0.09-0.12 0.9 0.84-1.00 0.1 0.09-0.12 l .o610 0.89-1.06 0.1 0.09-0.12 0.99 0.82-0.99 0.1 0.08-0.11 0.9 0.79-0.95 0. 19 0.10-0.13 0 .9 3 0.78-0.95 0.1 0.08-0.11 0.9J 0.77-0.94 0.1 0.08-0.11 0.93 0.77-0.93 0. 19 0.10-0.13 0.81-3,4 0.69-0.85 O.l 5'8 0.07-0.10 0.83 0.75-0.92 0.1 0.08-0.12 Table 15B. Adjusted* Mean and 95% Confidence Intervals (95% Cl) for Hepatic Clinical Chemistry Results by PFOA Decile PFOA Aik Plus___________ AST ALT GOT Decile Mean 95% Cl Mean 95% Cl Mean 95% Cl Mean 95% Cl Total Direct Bilirubin Bilirubin Mean 95% Cl Mean 95% Cl 1 66 61-71 26s-6 24-28 30 26-34 335,6 27-39 0.99 0.85-1.01 0.1 0.09-0.12 2 607,9,10 55-65 256 23-27 30 26-34 26 20-32 0.99 0.83-0.99 0.1 0.09-0.12 3 647 59-69 2656 24-28 29 25-33 27 21-33 l.O610 0.90-1.06 0.1 0.09-0.12 4 66 61-71 24 22-26 299 25-33 289 23-34 0.99 0.82-0.98 0.1 0.08-0.11 5 647 60-69 221.3.7.8 20-24 2 7 9 .,0 23-31 24'.' 18-30 0.9 0.80-0.96 0.19 0.10-0.13 6 66 61-71 22<-3.7-9 20-24 257,9,10 21-28 211,9 15-27 0.93 0.78-0.95 0.1 0.08-0.11 7 722-1-5-8 67-77 26s6 24-28 3 16 27-35 28 23-34 0.93 0.78-0.94 0.1 0.08-0.11 8 658 61-69 265'6 24-28 299 25-33 28 23-34 0.93 0.76-0.93 0.19 0.10-0.13 9 702 66-75 266 24-28 364,5,6'8 31-39 33 28-39 0.8m 0.68-0.85 0.085* 0.07-0.10 10 672 62-72 24 22-26 335'6 29-37 27 22-33 0.93 0.0.77-0.94 0.1 0.09-0.12 Adjusted for Age, Triglycerides, and Alcohol using analysis of covariance M0 Statistically significantly (p < .05) different than PFOA decile(s) 1, 2, 3 . . . and/or 10 Table 15C. Adjusted Means and 95% Confidence Intervals (Cl) for AST/ALT Ratios by PFOA Decile PFOA Decile AST/ALT Mean3 95% Cl AST/ALT Meanb 95% Cl 1 0.9910 0.96-1.02 0.99 0.95-1.01 2 0.98' 0.96-1.02 0.98 0.95-1.02 3 l.OO10 0.96-1.03 l.OO9 0.97-1.03 4 0.98 0.95-1.01 0.98 0.94-1.00 5 0.97 0.94-1.01 0.95 0.94-1.00 6 l.OO9,10 0.97-1.04 1.019'10 0.97-1.04 7 0.98 0.95-1.01 0.98 0.95-1.01 8 l.OO910 0.97-1.03 l.OO910 0.97-1.04 9 0.9568 0.92-0.98 0.953,6'8 0.91-0.98 10 0.941' 3,6,8 0.91-0.98 O.9 5 M 0.0.92-0.98 a. Adjusted for Age, BMI and Alcohol using analysis of covariance b.Adjusted for Age, Triglycerides and Alcohol using analysis of covariance M0 Statistically significantly (p < .05) different than PFOA decile(s) 1, 2, 3 . . . and/or 10 Tabic 16. Number and Percent of Hepatic Clinical Chemistry Results by Reference Points, by PFOA Decile, 2000 Medical Fluorochemical Surveillance Program PFOA Decile Aik Phos >= 120 AST >= 50 N (%) N (%) 1 0(0) 2 0(0) 3 0(0) 4 0(0) 5 1(2) 6 1(2) 7 1(2) 8 0(0) 9 1(2) 10 1(2) p value* .81 2(4) 0(0) 0(0) 0(0) 1(2) 0(0) 1(2) 2(4) 1(2) 0(0) .49 *chi square test o f significance ALT >= 50 N (%) 6(12) 4(8) 4(8) 4(8) 3(6) 0(0) 4(8) 5(9) 7(14) 7(14) .38 GOT >= 50 N (%) 6(12) 2(4) 5(10) 5(10) 2(4) 2(4) 5(10) 5(9) 8(16) 4(8) .46 Total Direct Bilirubin >11.5 Bilirubin > 0.4 2 (4) 1 (2) 4 (8) 2 (4) 1 (2) 1 (2) 1 (2) 0 (0) 1 (2) 1 (2) .58 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (2) 0 (0) 1 (2) .56 Table 17A. Odds Ratios (O.R.) and 95% Confidence Intervals (95% C.I.) for Hepatic Clinical Chemistry Results, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program PFOA Decile 1 2 3 4 5 6 7 8 9 10 ALT > 50 IU/L ----------------------------------------------------------------------- O.R1. 95% C.I. O.R21. 95% C.I. O.R3. 95% C.I. 1.0 - 1.0 - 1.0 - 0.6 0.2-2.4 0.8 0.2-3.0 0.8 0.2-3.2 0.6 0.2-2.4 0.7 0.2-2.9 0.6 0.1-2.2 0.7 0.2-2.4 0.6 0.1-2.3 0.6 0.1-2.4 0.5 0.1-1.9 0.4 0.1-1.9 0.4 0.1-1.7 0.2 0.0-1.0 0.1 0.0-0.9 0.1 0.0-0.9 0.7 0.2-2.4 0.8 0.2-3.0 0.6 0.1-2.3 0.8 0 2 - 2.1 0.7 0.2-2.8 0.7 0.2-2.5 1.3 0.4-4.2 1.0 0.3-3.4 1.0 0.3-3.6 1.2 0.4-4.1 1.1 0.3-3.8 0.7 0.2-2.6 1. N ot adjusted 2. Adjusted for age, BMI and alcohol 3. Adjusted for age, triglycerides and alcohol GGT > 50 1U/L -----------------------------------------------------O.R1. 95% C.I. O.R2. 95% C.I. O.R3 95% C.I. 1.0 0.3 0.0-1.4 0.8 0.2-2.9 0.8 0.2-3.0 0.3 0.0-1.4 0.3 0.0-1.5 0.8 0.2-3.0 0.8 0 2 - 2.1 1.5 0.5-4.8 0.7 0.2-2.4 1.0 0.3 0.0-1.4 0.9 0.2-3.1 0.8 0 2 - 2.1 0.3 0.0-1.3 0.3 0.0-1.3 0.9 0.2-3.1 0.7 0.2-2.6 1.7 0.5-5.6 0.7 0.2-2.5 1.0 0.3 0.0-1.6 0.7 0.2-2.8 0.8 0.2-2.9 0.3 0.0-1.3 0.3 0.0-1.3 0.8 0.2-2.9 0.6 0.2-2.4 1.7 0.5-5.8 0.4 0.1-1.7 Table 17B. Odds Ratios (O.R.) and 95% Confidence Intervals (95% C.I.) for Hepatic Clinical Chemistry Results, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program PFOA Decile 1 2 3 4 5 6 7 8 9 10 ALT > 40 IU/L ----------------------------------------------------------------------- O.R1. 95% C.I. O.R21. 95% C.I. O.R3. 95% C.I. 1.0 - 1.0 - 1.0 - 0.9 0.3-2.4 1.1 0.4-3.2 l.l 0.4-3.3 0.8 0.3-2.1 0.8 0.3-2.5 0.7 0.2-2.0 0.9 0.3-2.5 0.8 0.3-2.4 0.9 0.3-2.4 0.3 0.1-1.1 0.3 0.1-1.1 0.3 0.1-1.0 0.3 0.1-0.9 0.2 0.1-0.9 0.2 0.1-0.9 1.3 0.5-3.4 1.5 0.5-4.1 1.2 0.5-3.4 1.4 0.6-3.7 1.5 0.6-4.3 1.3 0.5-3.6 2.2 0.9-5.6 1.7 0.7-4.8 1.7 0.7-4.6 1.4 0.6-3.8 1.2 0.5-3.4 0.9 0.3-2.5 1. Not adjusted 2. Adjusted for age, BMI and alcohol 3. Adjusted for age, triglycerides and alcohol GOT > 40 IU/L ------------------------------------------------------ O.R1. 95% C.I. O.R2. 95% C.I. O.R3 95% C.I. 1.0 0.6 0.2-1.9 0.7 0.2-2.2 1.0 0.4-2.9 0.9 0.3-2.5 0.5 0.2-1.7 1.2 0.4-3.2 0.7 0.2-2.0 1.5 0.6-4.1 1.6 0.6-4.4 1.0 0.6 0.2-1.9 0.8 0.3-2.4 1.0 0.4-2.8 0.8 0.3-2.4 0.5 0.1-1.6 1.2 0.4-3.5 0.7 0.2-2.0 1.6 0.6-4.5 1.7 0.6-4.5 1.0 0.7 0.2-2.3 0.7 0.2-2.2 1.0 0.3-3.0 0.8 0.1-1.5 1.1 0.4-3.3 0.6 0.2-1.7 1.6 0.6-4.7 1.2 0.4-3.4 1.3 0.3-4.6 Table 18. Non-adjusted and Adjusted Ln PFOA Coefficients for Ln Hepatic Clinical Chemistry, 2000 Fluorochemical Medical Surveillance Program Non-adjusted Ln PFOA Coefficient SE p value Adjusted12 Ln PFOA Coefficient SE p value Ln Alkaline Phosphatase All Locations 0.0155 Antwerp -0.0025 Cottage Grove -0.0141 Decatur 0.0394 0.0082 0.0137 0.0113 0.0191 .06 0.0093' 0.0081 0.00372 0.0081 .85 -0.0060 0.0139 -0.0170 0.0140 .21 -0.0127 0.0117 -0.0140 0.0117 .04 0.0460 0.0192 0.0394 0.0192 Ln AST .25 .65 .67 .22 .28 .24 .02 .04 All Locations -0.0018 Antwerp -0.0048 Cottage Grove -0.0281 Decatur 0.0205 0.0086 0.0137 0.0141 0.0200 .83 -0.0051 -0.0089 .73 -0.0029 -0.0066 .05 -0.0258 -0.0271 .31 0.0114 0.0062 Ln ALT 0.0086 0.0087 0.0138 0.0142 0.0146 0.0145 0.0203 0.0203 .55 .31 .83 .64 .08 .07 .57 .76 All Locations 0.0402 Antwerp -0.0122 0.0143 .005 0.0249 0.0115 0.0220 .58 -0.0085 -0.0293 Table 18. (Continued) 0.0132 0.0136 0.0222 0.0222 .06 .40 .70 .19 Non-adjusted Ln PFA Coefficient SE p value Adjusted12 Ln PFOA Coefficient SE Ln ALT (Confd) p value Cottage Grove -0.0131 Decatur 0.0954 0.0215 0.0300 .54 .002 Ln GGT -0.0096 -0.0008 0.0704 0.0581 0.0209 0.0208 0.0287 0.0287 .65 .69 .02 .04 All Locations 0.0409 Antwerp 0.0170 Cottage Grove -0.0088 Decatur 0.0754 0.0174 0.0307 0.0292 0.0344 .02 0.0326 0.0097 .58 0.0269 -0.0047 .76 -0.0198 -0.0233 .03 0.0800 0.0599 Ln Total Bilirubin 0.0166 0.0163 0.0294 0.0295 0.0286 0.0270 0.0344 0.0329 .05 .55 .36 .87 .49 .39 .02 .07 All Locations -0.0406 0.0101 .0001 Antwerp -0.0117 0.0178 .51 Cottage Grove -0.0060 0.0138 .66 Decatur -0.0528 0.0203 .01 -0.0325 -0.0267 -0.0122 -0.0093 -0.0098 -0.0067 -0.0537 -0.0462 1. See study methods. Adjusted for Ln Age. Ln BMI, Ln Alcohol 2. See study methods. Adjusted for Ln Age. Ln Triglycerides, Ln Alcohol 0.0099 0.0101 0.0182 0.0188 0.0142 0.0141 0.0209 0.0206 .001 .01 .50 .62 .49 .64 .01 .03 Table 19. Mean and Standard Deviation (SD) of Thyroid-Related Clinical Chemistry Results, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program PFOA Decile TSH T4 Free T4 ___T3 Mean (SD)______Mean (SD)______Mean (SD)________Mean SD 1 2.06 (1.17)4 8.28 (1.51) 1.15 (0.15)5,7'10 124 (28) 2 2.03 (1.09)4 8.41 (1.40) 1.11 (0.15) 124 (21) o VI o 3 1.84 (0.73)4 8.12 (1.31) 1.15 125 (19) 4 3.41 (9.07)12-3 8.04 (1-51) 1.12 (0.19) 127 (19) 5 2.60 (2.63) 7.96 (1.37) 1.07 (0.15)1' 3 126 (17) 6 2.10 (1.09) 8.49 (1.33) 1.10 (0.13) 128 (23) 7 2.19 (1.23) 8.08 (1.29) 1.09 (0.13)1 129 (21) 8 2.43 (2.92) 8.37 (1.29) 1.09 (0.17) 130 (27) 9 2.84 (3.40) 8.49 (1.30) 1.10 (0.14) 130 (20) 10 2.41 (1.27) 7.97 (1.33) 1.07 (0.14)1,3 130 (31) M0 Statistically significantly (p < .05) different than PFOA decile(s) 1,2, 3 . . . and/or 10 Table 20. Adjusted* Mean and 95% Confidence Intervals for Thyroid Clinical Chemistry Results by PFOA Decile PFOA Decile _______TSH________ Mean 95% Cl ________T4_________ Mean 95% Cl Free T4 Mean 95% Cl ______ 32 Mean 95% Cl 1 2.074 1.13-3.01 8.27 7.91-8.66 1.157'10 1.11-1.19 124 118-131 2 2.004 1.04-2.96 8.45 8.07-8.83 1.11 1.07-1.15 125 118-131 3 1.894 0.94-2.85 8.08 7.70-8.46 1.I45'10 1.10-1.18 124 118-131 4 3.43w 2.48-4.38 8.04 7.67-8.41 1.12 1.08-1.16 126 120-133 5 2.60 1.65-3.55 7.98 7.60-8.35 1.071'3 1.03-1.11 126 120-133 6 2.074 1.11-3.03 8.5310 8.15-8.91 1.10 1.05-1.14 128 121-134 7 2.21 1.26-3.16 8.0710 7.70-8.44 1.09' 1.05-1.13 129 123-135 8 2.39 1.46-3.32 8.40 8.03-8.76 1.10 1.06-1.14 130 124-136 9 2.84 1.87-3.81 8.42 8.03-8.80 1.11 1.07-1.15 1329 123-136 10 2.41 1.44-3.37 7.946 7.56-8.32 1.07u 1.03-1.11 129 123-136 *Adjusted for Age, BMI and Alcohol using analysis of covariance M0 Statistically significantly (p < .05) different than PFOA decile(s) 1,2, 3 . . . and/or 10 Table 21. Number and Percent Subjects Above or Below Reference Points for Thyroid-Related Clinical Chemistry Results, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program PFOA Decile TSH (uIU/mL) <0.35 >5.5 N (%) N (%) T4 (ug/dL) <4.5 >12.0 N (%) N (%) 1 0(0) 0(0) 2 1(2) 0(0) 3 0(0) 0(0) 4 1(2) 2(4) 5 0(0) 4(8) 6 0(0) 1(2) 7 0(0) 2(4) 8 0(0) 4(8) 9 0(0) 4(8) 10 1(2) 1(2) p value* .63 .10 0(0) 0(0) 0(0) 0(0) 1(2) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) .44 - Free T4 fne/dU <0.70 >1.53 N (%) N (%) 0(0) 1(2) 0(0) 1(2) 0(0) 0(0) 1(2) 1(2) 1(2) 0(0) 0(0) 0(0) 0(0) 0(0) 1(2) 2(4) 0(0) 0(0) 0(0) 0(0) .65 .46 T3 (ng/dL) <60 >181 N (%) N (%) 0(0) 1(2) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 2(4) 0(0) 1(2) 0(0) 2(4) 0(0) 0(0) 0(0) 1(2) - .46 *chi square test of significance Table 22. Thyroid related hormone tests results by subject when TSH > 4.0 pIU/mL (ascending order) and corresponding PFOA (pg/mL) concentration. Results out-of-reference range (see Table 21) are shaded. Subject 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 TSH 4.03 4.06 4.12 4.12 4.15 4.21 4.21 4.23 4.24 4.25 4.26 4.32 4.48 4.60 4.64 4.71 4.90 4.91 4.94 4.96 5.03 5.12 5.20 5.36 5.38 5.46 5.47 5.60 5.68 5.73 5.75 5.87 6.01 6-05 6.09 6.32 7.51 7.99 8.55 S.42 11.83 18.78 19.39 23.24 65.28 14 Free T4 6.1 1.13 6.2 1.13 7.9 0.98 8.2 1.17 5.1 0.76 8.9 1.0 6.0 0.87 10.9 1.23 8.8 1.03 6.4 1.25 6.2 1.22 6.7 0.99 7.2 0.91 7.8 1.01 6.6 1.13 6.1 0.94 8.2 1.14 6.3 0.76 7.9 1.16 7.8 1.35 9.0 1.22 6.2 0.88 6.4 0.97 7.6 0.94 9.4 1.17 8.0 1.09 8.8 1.19 8.8 0.83 7.0 1.00 9.4 1.01 9.1 1.09 7.7 0.94 6.7 0.92 8.9 1.10 8.8 1.25 5.6 0.88 6.7 0.94 8.8 1.04 9.7 1.54 9.0 1.23 9.4 1.07 8.4 0.84 4.7 0.95 6.9 0.91 4.6 0.55 J3 PFOA 103 1.615 106 60.225 117 3.115 156 3.554 90 0.070 127 0.071 109 1.606 116 0.042 164 0.081 153 0.604 154 0.517 125 0.202 112 1.070 143 2.143 114 2.252 116 0.126 109 1.415 118 1.252 100 1.727 119 0.185 128 2.152 119 1.388 102 0.160 157 12.696 117 0.126 106 4.157 132 5.154 144 2.579 135 1.625 124 1.888 118 0.731 116 1.106 140 0.718 136 1.525 146 0.879 114 4.715 134 2.060 102 3.670 109 2.343 133 3.637 164 0.576 123 0.725 102 2.434 120 2.715 121 0.457 Table 23. Non-adjusted and Adjusted' Ln PFOA Coefficients for Ln Thyroid-Related Hormone, 2000 Fluorochemical Medical Surveillance Program Non-adjusted Ln PFOA Coefficient SE p value Ln TSH Adjusted1 Ln PFOA Coefficient SE p value All Locations 0.0395 0.0204 .05 0.0360 0.0207 .08 Antwerp 0.0509 0.0329 .12 0.0391 0.0333 .24 Cottage Grove -0.0016 0.0310 .96 -0.0111 0.0322 .73 Decatur 0.0343 0.0497 .49 0.0365 0.0513 .48 Ln T4 All Locations -0.0037 0.0054 .50 -0.0057 0.0054 .29 Antwerp -0.0022 0.0099 .83 -0.0041 0.0099 .68 Cottage Grove -0.0124 0.0072 .09 -0.0093 0.0072 .20 Decatur -0.0012 0.0126 .92 -0.0083 0.0127 .51 Ln Free T4 All Locations -0.0138 0.0044 .002 -0.0117 0.0043 .01 Antwerp -0.0108 0.0078 .17 -0.0140 0.0078 .07 Cottage Grove -0.0093 0.0058 .11 -0.0071 0.0059 .23 Decatur -0.0138 0.0103 .18 -0.0184 0.0105 .08 Ln T3 All Locations 0.0107 0.0052 .04 0.0105 0.0053 .05 Antwerp 0.0222 0.0077 .005 0.0216 0.0077 .006 Cottage Grove 0.0026 0.0096 .79 0.0006 0.0099 .95 Decatur 0.0317 0.0117 .008 1. See study methods. Adjusted for Ln Agc?Ln BMI, Ln Alcohol 0.0271 0.0119 .02 Table 24. Predicted Thyroid-Related Clinical Chemistry Results Based on Multiple Regression Models for 40 Year Old Male with BMI = 28 and Drinks 0.5 Alcohol Beverages per Day. Predicted Predicted Predicted Predicted TSH (nIU/mL) T4 (pg/dL) Free T4 (ng/dL) T3 (ng/dL) Reference Range (0.25 - 5.51 (4.5-12.01 (0.70-1.531 (60-1811 PFOA Serum Concentration (ue/mLI 0.005 1.58 8.22 1.15 119 0.01 1.62 8.19 1.15 119 0.10 1.76 8.08 1.11 121 0.50 1.87 8.01 1.09 124 1.00 1.91 7.98 1.09 125 5.00 2.03 7.90 1.06 128 10.00 2.07 7.87 1.06 128 50.00 2.20 7.80 1.04 131 100.00 2.26 7.77 1.03 132 Table 25. Number (percent), Odds Ratios (O.R.) and 95% Confidence Intervals (95% C.I.) with Metabolic Syndrome, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program PFOA Decile 1 2 3 4 5 6 7 8 9 10 N (%) 6(12) 4 (8) 1(2) 8(16) 5(10) 6(12) 3(6) 6(12) 6(12) 6(12) Non-adj usted O.R. 95% C.I. 1.0 0.6 0.2-2.4 0.2 0.0-0.9 1.4 0.5-4.7 0.8 0.2-2.9 1.0 0.3-3.6 0.5 0.1-1.9 0.9 0.3-3.2 1.0 0.3-3.6 1.0 0.3-3.5 Adjusted1 O.R. 95% C.I. 1.0 0.6 0.1-2.2 0.2 0.0-1.0 1.5 0.5-4.9 0.8 0.2-2.9 1.0 0.3-3.5 0.5 0.1-2.0 0.9 0.3-3.0 1.1 0.3-3.6 1.0 0.3-3.6 1. Adjusted for age Appendix A Tabic Al. Correlation Coefficients (same as Table Dl) Ln (Variable) Ln (PFOA)_________Ln(PFOS) PFOA PFOS Age BMI Alcohol Cholesterol LDL IIDL Triglycerides Glucose Aik Phos AST ALT AST/ALT GGT Total Bilirubin Direct Bilirubin TSH T4 Free T4 T3__________ 1.0 0.55**** 0.03 0.11* -0.12** 0.04 0.00 -0.17**** 0.210**** -0.01 0.08 -0.01 0.12** -0.15*** 0.10* -0.18**** -0.06 0.09 -0.03 -0.15** 0.09 1.0 0.06 0.00 -0.10* 0.08 0.06 -0.07 0.13** -0.13** 0.08 0.00 0.08 -0.09* 0.10* -0.13** -0.03 0.04 0.002 -0.06 0.03 *p < .05 ** p < .01 *** p < . 001 ****p< oooi LPAgcl 1.0 q 20**** -0.15*** 0.15*** 0.13** -0.13** 0.18**** 0.26**** 0.08 0.00 0.03 -0.05 0.15*** -0.08 -0.03 -0.01 -0.09 -0.18*** -0. 10* Ln(BMI) Ln(Alcohol) 1.0 -0.26**** 0.00 -0.02 -0 43**** q 44**** 0 28**** 0. 12** 0.15*** 0 41**** -0 39**** 0 3i**** -0.18**** -0.03 0.08 -0.05 -0.17*** 0.02 1.0 0. 10* 0.01 037**** -0. 12** -0.19**** -0.25**** -0.04 -0.16*** 0.17*** 0.20**** -0.02 -0.02 -0.02 -0.17*** 0.08 -0.01 Age by PFOA (jig/mL) LN Age by LN PFOA (p.g/mL) In AGE LN Age by LN PFOA (jig/mL) In PFOA BMI by PFOA (ug/mL) In BMI I I BMI 40- 35- 30- 25- 20- 15 , r --| i I ' -r - r - - f - 0 10 20 30 40 50 60 70 80 90 PFOA LN BMI by LN PFOA (ug/mL) 4-- 3 .9 - 3 .8 - 3 .7 - 3 .6 - 3 .5 - 3 .4 - 3 .3 - 3 .2 - 3 .1 - 3- 2 9- 2.8- -5 -4 -3 - 2 - 1 0 1 In PFOA LN BMI by LN PFOA (ng/mL) ln PFOA Alcohol (Drinks/Day) by PFOA( ng/'mL) LN Alcohol (Drinks/Day) by Ln PFOA( ug/mL) In PFOA LN Drinks/day by PFOA (ng/mL) PFOS pig/mL) by PFOA (ng/mL) 0 10 20 30 40 50 60 70 80 90 PFOA Glucose (mg/dL) by PFOA (|ig/mL) LN Glucose (mg/dL) by LN PFOA (|ig/mL) LN Glucose (mg/dL) by LN PFOA (jig/mL) ln PFOA Cholesterol (mg/dL) by PFOA (jig/mL) LN Cholesterol (mg/dL) by LN PFOA (jig/mL) 5 .8 - 5 .6 - _j 5 .4 O X 0 5 .2 - 5- 4 .8 - v * * , A* ml \ k . a. : . -*. o: oe `- r . _ ' . r . -v y - l i r ' .: m ' * i <0 i * i 1i 1i 1 i 1i 1i 1i 1i i 5 -4 -3 -2 -1 0 1 2 3 4 5 In P F O A I LN Cholesterol (mg/'dL) by LN PFOA (jig/mL) LDL (mg/dL) by PFOA (|ag/mL) 0 10 20 30 40 50 60 70 80 90 PFOA LN LDL (mg/dL) by LN PFOA (ng/mL) In PFOA LN LDL (mg/dL) by LN PFOA (ng/mL) i * i 1 i 1 i 1 i * i ! i 1 i 1 ! 1 i------ r~ -5 -4 -3 -2 -1 0 1 2 3 4 5 In PFOA \ HDL (mg/dL) by PFOA (pg/mL) --r--'--i-- ~ !-- '--1-- 10 20 30 40 50 60 70 80 90 PFOA LN HDL (mg/dL) by LN PFOA (pg/mL) -5 -4 -3 -2 -1 0 1 2 3 4 5 In PFOA LN HDL (mg/dL) by LN PFOA (ug/mL) Triglycerides(mg/dL) by PFOA (ug/mL) LN Triglycerides (mg/dL) by LN PFOA (jxg/mL) In P F O A LN Triglycerides (mg/dL) by LN PFOA (ug/mL) i. Alkaline Phosphatase (IU/L) by PFOA (pg/mL) 160- 140- 120-1 * -r . -i 1 0 0 I ** Al I_o0L0 fr * * 60- f e *' ' K- ' . 40- -- 20- r "1 r r D 10 20 30 9 ----i T " 1 i r ** i i r ' 40 50 60 70 80 90 PFOA LN Alkaline Phophatase (IU/L) by LN PFOA (jig/mL) 5- ----------------------------- * *. ''* A-V . * 1O0 JQZ. 4~ *. ^ * ** m . ' ' ^ 1'1 1 I T 1 1 I 5 -4 -3 -2 -1 0 1 2 In PFOA 'I I 1 1 345 LN Alkaline Phosphatase (IU/L) by LN PFOA (pg/mL) AST (IU/L) by PFOA (jig/mL) ir 10- ' I " *" 'i 1 i ' i 1 r .-- i-- -- t-- 8,,,........r n " 0 10 20 30 40 50 60 70 80 90 PFOA LN AST (IU/L) by LN PFOA (ng/mL) LN AST (IU/L) by LN PFOA (jig/mL) 4- ! 00 < c 3-1 --i~*--i-- i---- r -5 - 4 - 3 - 2 - 1 In PFOA ALT (IU/L) by PFOA (jig/mL) i i i r i ' i r~'--r 10 20 30 40 50 60 70 80 90 PFOA LN ALT(IU/L) by LN PFOA (ug/mL) 4.5- 4- . S 3.5< !" * - I 2-5- 2- n r T"'i----r~,--r i -5 -4 -3 -2 -1 0 1 In P F O A 1"1 I I ! 2345 In A L T LN ALT (IU/L) by LN PFOA (ng/mL) In PFOA AST/ALT by PFOA (ng/mL) AST/ALT by LN PFOA (ng/mL) LN AST/ALT by LN PFOA (jig/mL) ln PFOA GGT (IU/L) by PFOA (jig/mL) PFOA LN GGT (IU/L) by LN PFOA (jxg/mL) Total Bilirubin (mg/dL) by PFOA (jig/mL) 2- c_ 1 --* !^ -- \ 1 1; --* * i i i i i i i i i i 0 10 20 30 4 0 5 0 60 70 80 SO PFOA LN Total Bilirubin (mg/dL) by LN PFOA Gig/mL) i ' i i 1 i ' i i 1 i i ' i 1 i ' -5 -4 -3 -2 -1 0 1 2 3 4 5 In P F O A LN Total Bilirubin (mg/dL) by LN PFOA (ug/mL) Direct Bilirubin (mg/dL) by PFOA (jig/mL) I -7: 0 .6 - : . 0 .5 - I ! 0 .4 - CO ! g j u vo 0 . 3 j Q 0 .2 - * 0- -- r-- , , r-- , . , 1 . 1 i 1 r - p - r i i J I 1 0 10 20 30 40 50 60 70 80 90 PFOA LN Direct Bilirubin (mg/dL) by LN PFOA (jig/mL) LN Direct Bilirubin (mg/dL) by LN PFOA (jxg/mL) ln P F O A TSH (jiIU/mL) by PFOA (jig/mL) PFOA LN TSH (uIU/mL) by LN PFOA Og/mL) ln P F O A LN TSH (jig/mL) by LN PFOA Og/mL) T4 (ng/dL) by PFOA (jig/mL) PFOA LN T4 (n g/dL) by PFOA (jig/mL) 2 .3 - . .. 2 .1 - .* * ... 1 .9 - ** : . * - , w- * 1 .7 1 .5 - 1 .3 - 1 .1 - - 5 -4 -3 -2 -1 0 12 3 4 5 In P F O A In T 4 LN T4 (jig/dL) by LN PFOA (jig/inL) In P F O A In T A Free T4 (ng/dL) by PFOA (jxg/mL) LN Free T4 (ng/dL) by LN PFOA (jig/mL) LN Free T4 (ng/dL) by LN PFOA (ng/mL) T3 (ng/dL) by PFOA (jig/mL) LN T3 (ng/dL) by LN PFOA (ng/mL) In P F O A In T 3 LN T3 (ng/dL) by LN PFOA Gig/mL) In P F O A Appendix B Table Bl. Demographic and Clinic Chemistry Comparisons (Mean, Standard Deviation (SD). Median and Range) of Employees Who Self-Reported Prescribed Cholesterol-Lowering Medications (N = 46) Versus Those Who Did Not (N = 506). 2000 Fluorochemical Medical Surveillance Program Prescribed Medication (N=46) Not Prescribed (N=506) Mean (SD) Median Range Mean (SD) Median Range PFOA 1.98 2.12 1.21 0.14-10.30 PFOS 1.69* 1.72 1.27 0.11-10.06 Age 49* 7 50 31-60 BMI 28.8* 4.5 28.4 19.9-39.3 Alcohol 0.4 0.6 0.0 0.0-2.0 % Antwerp 21* - - - % Cottage Grove 20* - - - % Decatur 59* - - - Glucose 103* 38 99 54-331 Cholesterol 221 45 217 144-384 LDL 134 40 130 59-222 HDL 47 14 43 31-91 Triglycerides 226* 160 194 35-792 Aik Phos 73* 21 73 30-126 AST 27* 8 25 7-48 ALT 36* 20 32 9-96 GGT 36* 24 28 10-144 Total Bilirubin 0.8 0.2 0.8 0.4-1.5 2.21 6.40 1.10 0.01-92.03 1.05 0.97 0.72 0.02-6.24 40 9 39 21-67 27.4 4.6 26.6 17.2-52.1 0.6 0.9 0.3 0.0-6.4 39 - - - 24 - - - 37 - - - 91 19 91 31-251 214 41 211 105-331 136 36 133 37-235 49 13 46 18-121 159 112 128 24-796 66 18 64 21-160 25 7 24 10-69 30 15 26 6-103 28 22 22 6-314 0.9 0.3 0.8 0.3-2.3 Table Bl. (continued) Prescribed Lipid-Lowering Medication (N=46) Not Prescribed (N=506) Mean (SD) Median Range Mean (SD) Median Range Direct Bilirubin 0.1 0.06 0.1 0.0-0.3 AST/ALT 0.89 0.42 0.82 0.22-2.22 TSH 2.8 3.1 2.1 0.4-21.5 T4 8.1 1.4 7.9 5.8-12.9 Free T4 1.1 0.2 1.1 0.8-1.5 T3 125 19 123 93-190 * p <.05 (prescribed vs. non-prescribed) 0.1 0.06 0.1 0.0-0.7 0.98 0.12 0.97 0.70-1.82 2.4 3.4 1.9 0.03-65.3 8.2 1.4 8.2 4.2-12.0 1.1 0.2 1.1 0.6-1.8 127 23 125 78-300 Tabic 132. Mean and Standard Deviation (SD) of PFOA, PFOS, Demographics and Clinical Chemistry Results by Location by Cholesterol-Lowering Medication Status Antwerp Cholesterol-Lowering Med Yes No Mean (SD) Mean (SD) Cottage Grove Cholesterol-Lowering Med Yes No Mean (SD) Mean (SD) Decatur Cholesterol-Lowering Med Yes No Mean (SD) Mean (SD) PFOA 1.16 (1.53) PFOS 1.25 (0.91) Age 50* (5) BMI 25.9 (3.4) Alcohol 0.7 (0.8) Glucose 96 (30) Cholestrol 232 (38) LDL 152 (43) HDL 52 (18) Triglycrides 195* (125) Alk Phos 58 (20) 1.02 (1.06) 0.95 (0.97) 37 (8) 24.7 (3.0) 1.1 (.i) 84 (17) 218 (41) 139 (37) 55 (15) 120 (83) 60 (14) 2.94 (3.66) 4.63 (12.53) 0.76 (0.74) 0.86 (0.98) 50* (7) 29.5 (4.2) 41 (9) 29.9 (4.8) 0.7 (0.7) 0.7 (0.7) 103 (13) 100 (23) 214 (38) 210 (39) 126 (47) 130 (32) 52 (17) 180 (83) 46 (11) 187 (139) 77* (23) 65 (15) 1.97 (1.52) 2.18* (2.01) 48* (7) 29.7 (4.6) 0.2 (0.4) 106* (46) 220 (50) 130 (35) 44 (10) 253* (187) 77 (19) 1.89 (1.61) 1.29 (0.92) 42 (9) 28.6 (4.6) 0.1 (0.3) 93 (14) 214 (41) 136 (36) 44 (10) 182 (110) 73 (20) Antwerp Cholesterol-Lowering Med Yes No Mean (SD) Mean (SD) Fable B2 (continued) Cottage Grove Cholesterol-Lowering Med Yes No Mean (SD) Mean (SD) Decatur Cholesterol-Lowering Med Yes No Mean (SD) Mean (SD) ASF 25 (6) 23 (6) ALT 22 (9) 23 (10) 30 (8) 25 (8) 43 (25) 34 (17) AST/ALT 1.3 (0.6) 1.2 (0.4) 0.8 (0.3) 0.8 (03) GOT 25 (13) 23 (17) 53 (40) 31 (32) Tot Bilirubin 1.0 (0.3) 1.0 (0.3) 0.9 (0.2) 0.9 (0.3) Dir Billirubin 0.1 (0.03) 0.1 (0.05) 0.1 (0.03) 0.1 (0.02) TSH 1.9 (0.7) 2.0 (1.6) 2.5 (0.6) 2.4 (1.4) T4 8.6 (1.7) 8.2 (1.4) 7.2 (1.2) 7.9 (1.1) Free T4 1.2 (0.2) 1.1 (0.2) 1.1 (0.1) 1.1 (0.1) T3 128 (18) 131 (19) 117 (16) 125 (30) * p < .05 cholesterol lowering medication (yes vs. no within each location) 27 (9) 39 (19) 0.8 (0.3) 35 (18) 0.7 (0.2) 0.1 (0.07) 3.2 (4.0) 8.2 (1.3) 1.1 (0.2) 127 (20) 26 (8) 34 (16) 0.9 (0.4) 30 (17) 0.7 (0.2) 0.1 (0.08) 2.8 (5.2) 8.4 (1.4) 1.1 (0.1) 125 (22) Table B3. Mean, Standard Deviation, Median and Range of PFOA, PFOS, Demographic Factors and Clinical Chemistry Results, By Location, 2000 Fluorochemical Medical Surveillance Program, For Employees Prescribed Cholesterol-Lowering Medications Antwerp (N= 1O') Mean (SD) Median Range Cottage Grove (N=9) Mean (SD) Median Range Decatur 01=27) Mean (SD) Median Range PFOA 1.16 1.53 PFOS 1.25 0.91 Age 50 5 BMI 25.93 3.4 Alcohol 0.73 0.8 Glucose 96 30 Cholestrol 232 38 LDL 152 44 HDL 52 18 Triglycrides 195 124 Alk Phos 582,3 20 AST 25 6 ALT 222'3 9 AST/ALT 1.3 0.6 0.62 0.14-5.31 0.98 0.23-2.77 51 41-56 24.7 22.4-34.2 0.5 0.0-2.0 89 54-168 243 179-280 160 85-222 45 31-87 179 35-463 58 30-84 24 17-37 22 9-38 1.2 0.6-2.2 2.95 3.67 1.55 0.29-10.30 1.97 1.52 1.76 0.15-4.97 0.763 0.74 0.36 0.11-2.10 2.182 2.01 1.58 0.15-10.06 50 7 52 39-60 48 7 47 31-60 29.5 4.2 30.9 23.4-35.5 29.7' 4.6 29.4 19.9-39.3 0.73 0.7 1.0 0.0-2.0 0.22'3 0.4 0.0 0.0-1.6 103 13 104 82-122 106 46 99 70-331 214 38 216 154-280 220 50 216 144-384 126 47 124 59-203 130 35 129 66-216 52 17 49 37-91 44 10 42 31-73 180 83 193 67-294 253 187 230 52-792 a ' 23 81 40-105 i f 19 74 44-126 30 8 27 20-41 27 9 25 7-48 431 25 33 16-96 39' 19 34 17-94 0.8 0.3 0.8 0.4-1.3 0.8 0.3 0.7 0.2-1.2 Tabic B3 (Continued) Antwerp (N=9) Mean (SD) Med Range Cottage Grove (N=9) Mean (SD) Med Range Decatur CN=27) Mean (SD) Med Range GGT Total Bil Direct Bil TSH T4 Free T4 T3 2521 13 1.03 0.3 0.1 0.03 1.9 0.7 8.62 1.7 1.2 0.2 128 18 21 10-49 0.9 0.7-1.5 0.1 0.1-0.2 1.8 0.7-3.1 8.3 5.9-11.5 1.2 0.9-1.5 128 98-162 53u 40 41 14-144 0.9 0.2 0.9 0.6-1.2 0.9 0.03 0.1 0.0-0.1 2.5 0.6 2.1 1.7-3.5 7.21 1.2 7.4 5.8-9.1 1.1 0.1 1.0 0.8-1.3 117 16 117 96-141 352 18 28 13-89 0.7' 0.2 0.7 0.4-1.1 0.1 0.07 0.1 0.0-0.3 3.2 4.0 2.5 0.4-21.5 8.2 1.3 8.0 6.2-12.9 1.1 0.2 1.1 0.8-1.4 127 20 123 93-190 1. Statisticallysignificantly (p < .05) different than Antwerp 2. Statistically significantly (p < .05) different than Cottage Grove 3. Statistically significantly (p < .05) different than Decatur Table B4. Non-adjusted and Adjusted* Regression PFOA Coefficients for Lipid Clinical Chemistry Results. 2000 Fluorochemical Results Restricted to Employee Taking Cholesterol-Lowering Medications (N = 46) Non-adjusted Ln of Response Ln PFOA Variable Coefficient SE Cholesterol LDL HDL 0.0453 0.0253 0.0003 0.0263 0.0494 0.0364 Triglycerides 0.0681 Aik Phosphatase 0.0612 0.0968 0.0420 p value .09 .61 .99 .49 .15 AST 0.0305 0.0437 .49 ALT 0.1040 0.0672 .13 GGT 0.0591 0.0734 .43 Total Bilirubin -0.0818 0.0356 .03 TSH T4 Free T4 T3 0.0928 -0.0223 -0.0219 0.0007 0.0904 0.0226 0.0197 0.0199 .31 .33 .27 .72 Adjusted* Ln PFOA Coefficient SE 0.0501 0.0303 0.0140 0.0292 0.0536 0.0346 0.1015 0.0972 0.0533 0.0420 0.0347** 0.0428 0.0434 0.0475 0.0340** 0.0469 0.0862 0.0674 0.0544** 0.0669 0.0653 0.0812 0.0404** 0.0746 -0.0649 0.0369 -0.0623** 0.0361 0.1803 0.0918 -0.0415 0.0236 -0.0307 0.0213 0.0041 0.0207 p value .09 .58 .69 .30 .21 .42 .37 .47 .21 .42 .43 .59 .09 .09 .06 .09 .16 .84 * Ln of PFOA coefficient adjusted for Ln Age, Ln BMI and Ln Alcohol unless otherwise (**) noted ** Ln PFOA coefficient adjusted for Ln Age, Ln Triglycerides and Ln Alcohol Appendix C Table Cl Number of Employees. Mean. 95% Confidence Interval, Median and Range by PFOA Decile Distribution , 2000 Fluorochemical Medical Surveillance Program for Antwerp, Cottage Grove and Decatur (N = 552) PFOA Decile 1 2 3 4 5 6 7 8 9 10 N 51 58 54 55 59 50 54 60 53 58 _______________________ PFOA Mean 95% C.I. Median 0.068' 10 0.05-0.07 0.06 0.208*10 0.19-0.21 0.19 0.369,10 0.35-0.37 0.36 0.559,10 0.53-0.55 0.55 0.909,10 0.87-0.94 0.88 1.26* 1.23-1.28 1.26 1.6410 2 jyl.2,10 1.60-1.67 2.12-2.23 1.63 2.16 3.0M51 11.271"9 2.90-3.10 6.98-15.56 2.96 4.94 Range 0.007-0.13 0.13-0.29 0.30-0.44 0.44-0.71 0.71-1.10 1.11-1.40 1.42-1.85 1.86-2.50 2.51-3.69 3.71-92.03 M0 Statistically significantly (p < .05) different than PFOA decile(s) 1,2, 3 . . . and/orlO Table C2. Adjusted Odds Ratios (O.R.) and 95% Confidence Interval (95% C.I.) for Lipid Clinical Chemistry Reference Points, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program (N = 552) PFOA Choi >= 200 mc/dl Choi >= 240 mc/d 1 LDL >== 130 mc/dl Decile O.R. 1 95% C.I. O.R.1 95% C.I. O.R.' 95% C.I. 1 1.0 2 0.4 0.2-0.9 3 0.8 0.4-1.8 4 1.0 0.4-2.2 5 1.7 0.8-3.9 6 1.0 0.4-2.3 7 0.9 0.4-1.9 8 1.1 0.5-2.4 9 1.5 0.7-3.5 1.0 1.1 0.4 -2.4 0.9 0.3-2.2 1.2 0.5-3.0 1.6 0.7-3.8 1.2 0.5-3.0 1.0 0.4-2.5 1.5 0.6-3.4 1.4 0.6-3.5 1.0 0.6 0.3-1.3 0.7 0.3-1.5 0.8 0.4-1.7 1.1 0.5-2.3 0.8 0.4-1.7 0.7 0.3-1.5 1.0 0.5-2.2 1.2 0.5-2.6 10 1.1 0.5-2.5 1.0 0.4-2.4 0.9 0.4-1.8 Adjusted for age, BMI and alcohol IIDL<= 40 mc/dl O.R.' 95% C.I. 1.0 1.4 0.5-3.6 0.4 0.1-1.3 1.9 0.7-4.9 1.1 0.4-3.0 1.7 0.6-4.6 0.6 0.2-1.8 1.2 0.5-3.2 0.9 0.3-2.3 2.9 1.2-7.5 Tricl veerides >= 150 mc/dl O.R.' 95% C.I. 1.0 0.8 0.4-2.0 1.0 0.4-2.5 1.4 0.6-3.3 1.1 0.5-2.5 1.7 0.7-4.2 1.0 0.4-2.5 2.7 1.2-6.4 2.4 1.0-5.7 2.7 1.2-6.3 PFOA Decile 1 2 3 4 5 6 7 8 9 10 Table C3. Nonadjusted and Adjusted Odds Ratios (OR.) and 95% Confidence Intervals (95% CM.) for IIDI, (< 40 mg/dL) and Triglycerides (>= 150 mg/dl.) by PFOA Decile, 2000 Fluorochcmical Medical Surveillance Program (N = 552) ___________HPI, < 40 mg/dl,______________ _________Triglycerides >" liQ rog/dl, O R .1 95% C.l O R .2 95% CM. O .R.1 95% CM. O.R.2 95% C.l. 1.0 1.0 1.0 - 1.0 - 1.4 0.6-3.6 l.l 0.4-2.' 0.9 0.3-2.0 0.8 0.3-1.8 0.4 0.1-1.3 0.3 0.1-1.1 0.8 0.4-1.9 0.8 0.3-1.9 2.0 0.8-5.0 1.6 0.6-1.2 1.1 0.7-3.2 1.2 0.5-2.8 1.3 0.5-3.3 0.9 0.3-2.4 l.l 0.5-2.5 0.9 0.4-2.0 1.8 0.7-1.5 1.3 0.5-3.6 1.7 0.8-3.9 1.5 0.6-3.5 0.7 0.3-2.0 0.5 0.2-1.3 1.0 0.4-2.3 0.8 0.3-1.9 1.5 0.6-3.8 0.9 0.3-2.4 2.6 1.2-5.8 2.0 0.9-4.5 0.6 0.5-3.5 0.6 0.2-1.8 2.6 1.2-5.9 1.7 0.7-4.0 _____ M 1.5-8,5 1.8 0.7-4.6 2.8 _ 1.3-6.3 1.8 0.8-4.1 'Not adjusted 'Adjusted for location Table C4. Odds Ratios (O.R.) and 95% Confidence Intervals (95% C.I.) for Hepatic Clinical Chemistry Results, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program (N = 552) PFOA Decile 1 2 3 4 5 6 7 8 9 10 O.R1 95% C.I. 1.0 0.7 0.2-2.5 0.6 0.1-2.2 0.8 0.2-2.7 0.4 0.1-1.6 0.2 0.0-0.9 0.8 0.2-2.7 0.7 0.2-2.4 1.1 0.4-3.8 1.8 0.6-5.5 ALT > 50 IU/L O.R12. 95% C.I. 1.0 0.8 0.2-3.1 0.7 0.2-2.9 0.7 0.2-2.7 0.3 0.1-1.6 0.1 0.1-0.9 0.9 0.2-3.5 0.6 0.1-2.4 0.9 0.3-3.3 1.7 0.5-5.5 O.R3. 95% C.I. 1.0 0.9 0.2-3.4 0.6 0.1-2.2 0.8 0.2-2.7 0.4 0.1-1.7 0.1 0.0-0.9 0.7 0.2-2.7 0.5 0.1-2.0 1.0 0.3-3.4 1.1 0.4-3.7 1. Not adjusted 2. Adjusted for age, BMI and alcohol 3. Adjusted for age, triglycerides and alcohol O.R1. 95% C.I. 1.0 0.3 0.0-1.2 0.8 0.2-2.7 l.l 0.3-3.6 0.4 0.1-1.6 0.3 0.4-1.4 0.9 0.3-3.2 0.7 0.2-2.4 1.7 0.6-5.5 0.7 0.2-2.5 GGT > 50 IU/L O.R2. 95% C.I. O.R3 95% C.I. 1.0 0.2 0.0-l.l 0.8 0.2-3.0 1.0 0.3-3.5 0.4 0.1-1.6 1.0 0.3 0.0-1.2 0.7 0.2-2.7 l.l 0.3-3-6 0.4 0.1-1.5 0.3 0.0-1.3 1.0 0.3-3.4 0.3 0.0-1.3 0.9 0.2-3.1 0.6 0.2-2.3 0.5 0.1-1.9 2.0 0.7-6.4 2.0 0.7-6.6 0.7 0.2-2.6 0.4 0.1-1.7 Table C5. Non-adjusted and Adjusted Regression PFOA Coefficients for Lipid Clinical Chemistry Results. 2000 Fluorochemical Medical Surveillance Program for Participants Who Did Not Self-Report Taking Cholesterol Lowering Medications (N = 506) and All Participants (N = 552) Ln PFOA Non-adjusted Coefficient SE p value Ln PFOA Adjusted1 Coefficient SE p value All Locations Not Prescribed 0.0059 All Participants 0.0083 Antwerp Not Prescribed 0.0051 All Participants 0.0050 Cottage Grove Not Prescribed 0.0034 All Participants 0.0045 Decatur Not Prescribed 0.0221 All Participants 0.0310 All Locations Not Prescribed 0.0005 All Participants 0.0018 Antwerp Not Prescribed -0.0037 All Participants -0.0041 Cottage Grove Not Prescribed -0.0036 All Participants -0.0001 Decatur Not Prescribed 0.0258 All Participants 0.0276 0.0060 0.0058 0.0106 0.0104 0.0089 0.0087 0.0139 0.0131 0.0089 0.0088 0.0157 0.0156 0.0138 0.0140 0.0199 0.0189 Ln Cholesterol .32 0.0076 0.0059 .15 0.0099 0.0058 .63 0.0130 0.0096 .63 0.0121 0.0094 .70 0.0021 0.0100 .61 0.0051 0.0091 .11 .02 Ln LDL .96 .84 0.0266 0.0348 0.0021 0.0029 0.0141 0.0133 0.0091 0.0089 .81 0.0106 0.0147 .79 0.0092 0.0146 .79 0.0049 0.0145 .99 0.0071 0.0146 .20 0.0302 0.0200 .15 0.0294 0.0191 .20 .09 .18 .20 .83 .58 .06 .01 .81 .75 .47 .53 .73 .63 .13 .12 Table C5 (continued) Ln HDL Ail Locations Not Prescribed -0.0307 All Participants -0.0295 0.0079 0.0077 .0001 .0001 -0.0183 0.0069 -0.0167 0.0067 .01 .01 Antwerp Not Prescribed -0.0057 All Participants -0.0030 0.0136 0.0137 .68 .83 -0.0095 0.0131 -0.0078 0.0130 .47 .55 Cottage Grove Not Prescribed All Participants -0.0153 -0.0154 0.0122 0.0121 .21 .20 -0.0192 0.0120 .11 -0.0199 0.01170 .09 Decatur Not Prescribed -0.0256 All Participants -0.0200 0.0149 0.0140 .09 .16 -0.0207 0.0141 -0.0159 0.0133 .14 .23 Ln Triglvcerides All Locations Not Prescribed 0.0892 All Participants 0.0917 0.0185 0.0185 .0001 .0001 0.0711 0.0738 0.0169 0.0168 .0001 .0001 Antwerp Not Prescribed 0.0840 All Participants 0.0734 0.0288 0.0294 .004 .01 0.0980 0.0920 0.0270 0.0274 .0004 .0001 Cottage Grove Not Prescribed 0.0343 All Participants 0.0317 0.0316 0.0306 .28 .30 0.0280 0.0269 0.0314 0.0300 .28 .37 Decatur Not Prescribed 0.0715 All Participants 0.0983 0.0400 0.0394 .08 .01 0.0689 0.0106 0.0376 0.0373 .07 .01 1. Multiple Regression Model: LN (lipid clinical chemistry) = Intercept + LN (age) + LN (BMI) + LN (alcohol) + LN (PFOA) Table C6. Non-adjusted and AdjustedI'0 Regression PFOA Coefficients for Hepatic Clinical Chemistry Results, 2000 Fluorochemical Medical Surveillance Program (N = 552) Non-adjusted Ln PFOA Coefficient SE p value Adjusted1-2 Ln PFOA Coefficient SE_______ p value Ln Alkaline Phosphatase All Locations 0.0155 Antwerp -0.0025 Cottage Grove -0.0141 Decatur 0.0394 Not Prescribed fN = 506) 0.0082 .06 0.0093' 0.0081 0.00372 0.0081 0.0137 .85 -0.0060 -0.0170 0.0139 0.0140 0.0113 .21 -0.0127 -0.0140 0.0117 0.0117 0.0191 .04 0.0460 0.0394 0.0192 0.0192 All Participants fN = 552) .25 .65 .67 .22 .28 .24 .02 .04 All Locations 0.0189 0.0082 .02 Antwerp -0.0082 0.0139 .56 Cottage Grove -0.0091 0.0115 .43 Decatur 0.0445 0.0176 .01 0.0117 0.0006 -0.0109 -0.0218 -0.0098 -0.0100 0.0492 0.0429 0.0080 0.0080 0.0140 0.0139 0.0119 0.0120 0.0179 0.0181 .14 .45 .44 .12 .41 .40 .01 .02 Ln AST Not Prescribed (N = 506) oo All Locations -0.0018 0.0086 Antwerp -0.0048 0.0137 .73 Cottage Grove -0.0281 0.0141 .05 Decatur 0.0205 0.0200 .31 -0.0051 -0.0089 -0.0029 -0.0066 -0.0258 -0.0271 0.0114 0.0062 0.0086 0.0087 0.0138 0.0142 0.0146 0.0145 0.0203 0.0203 .55 .31 .83 .64 .08 .07 .57 .76 All Particinants fN = 552) All Locations 0.0001 0.0085 .92 Antwerp -0.0050 0.0135 .71 Cottage Grove -0.0254 0.0139 .07 Decatur 0.0239 0.0194 .22 -0.0019 -0.0059 -0.0034 -0.0069 -0.0240 -0.0250 0.0179 0.0104 0.0085 0.0087 0.0135 0.0137 0.0144 0.0145 0.0198 0.0200 .82 .50 .80 .62 .10 .09 .37 .60 Ln ALT Not Prescribed (N = 506) All Locations 0.0402 Antwerp -0.0122 Cottage Grove -0.0131 Decatur 0.0954 0.0143 .005 0.0249 0.0115 0.0220 .58 -0.0085 -0.0293 0.0215 .54 -0.0096 -0.0008 0.0300 .002 0.0704 0.0581 All Participants (N = 552) 0.0132 0.0136 0.0222 0.0222 0.0209 0.0208 0.0287 0.0287 .06 .40 .70 .19 .65 .69 .02 .04 All Locations 0.0457 0.0141 .001 Antwerp -0.0151 0.0218 .49 Cottage Grove -0.0143 0.0215 .50 Decatur 0.1081 0.0278 .0001 0.0316 0.0170 -0.0118 -0.0311 -0.0089 -0.0134 0.0898 0.0691 0.0131 0.0135 0.0219 0.0218 0.0207 0.0215 0.0270 0.0274 .02 .21 .59 .15 .67 .53 .001 .01 Table C6 (continued) Ln GGT Not Prescribed (N = 506) All Locations 0.0409 Antwerp 0.0170 Cottage Grove -0.0088 Decatur 0.0754 0.0174 .02 0.0326 0.0097 0.0307 .58 0.0269 -0.0047 0.0292 .76 -0.0198 -0.0233 0.0344 .03 0.0800 0.0599 Not Prescribed (N = 552) 0.0166 0.0163 0.0294 0.0295 0.0286 0.0270 0.0344 0.0329 .05 .55 .36 .87 .49 .39 .02 .07 All Locations 0.0454 0.0170 .01 Antwerp 0.0114 0.0302 .71 Cottage Grove -0.0062 0.0295 .83 Decatur 0.0851 0.0319 .01 1. Adjusted for Ln Age. Ln BMI, Ln Alcohol 2. Adjusted for Ln Age. Ln BMI. Ln Alcohol 0.0380 0.0135 0.0214 -0.0097 -0.0183 -0.0290 0.0974 0.0673 0.0163 0.0158 0.0290 0.0286 0.0289 0.0278 0.0320 0.0308 .02 .39 .46 .73 .53 .30 .003 .03 Table Cl. Non-adjusted and Adjusted1Regression PFOA Coefficients for Thyroid-Related Clinical Chemistry Results, 2000 Fluorochemical Medical Surveillance Program (N = 552) Non-adjusted Ln PFOA Coefficient SE p value Ln TSH Adjusted1 Ln PFOA Coefficient SE p value All Locations Not Prescribed 0.0395 All Participants 0.0439 0.0204 0.0199 .05 .03 0.0360 0.0400 0.0207 0.0201 .08 .05 Antwerp Not Prescribed 0.0509 All Participants 0.0568 0.0329 0.0319 .12 .08 0.0391 0.0463 0.0333 0.0322 .24 .15 Cottage Grove Not Prescribed -0.0016 All Participants 0.0031 0.0310 0.0295 .96 .92 -0.0111 0.0322 -0.0066 0.0306 .73 .83 Decatur Not Prescribed 0.0343 All Participants 0.0319 0.0497 0.0477 .49 .50 0.0365 0.0328 0.0513 0.0490 .48 .44 Ln T4 All Locations Not Prescribed -0.0037 All Participants -0.0049 0.0054 0.0053 .50 .35 -0.0057 0.0054 -0.0072 0.0052 .29 .17 Antwerp Not Prescribed -0.0022 All Participants -0.0035 0.0099 0.0098 .83 .72 -0.0041 0.0099 -0.0047 0.0098 .68 .63 Cottage Grove Not Prescribed -0.0124 All Participants -0.0149 0.0072 0.0072 .09 .04 -0.0093 0.0072 -0.0121 0.0072 .20 .12 Decatur Not Prescribed -0.0012 All Participants -0.0015 0.0126 0.0150 .92 .90 -0.0083 0.0127 -0.0057 0.0116 .51 .63 All Locations Not Prescribed -0.0138 All Participants -0.0144 Antwerp Not Prescribed -0.0108 All Participants -0.0113 Cottaee Grove Not Prescribed -0.0093 All Participants -0.0012 Decatur Not Prescribed -0.0138 All Participants -0.0100 Table Cl (continued) Ln Free T4 0.0044 0.0042 .002 .001 -0.0117 0.0043 -0.0124 0.0042 0.0078 0.0077 .17 .15 -0.0140 0.0078 -0.0139 0.0078 0.0058 0.0057 .11 .04 -0.0071 -0.0091 0.0059 0.0057 0.0103 0.0096 .18 .30 -0.0184 0.0105 -0.0146 0.0098 .01 .003 .07 .07 .23 .12 .08 .14 All Locations Not Prescribed 0.0107 All Participants 0.0103 0.0052 0.0050 Antwerp Not Prescribed 0.0222 All Participants 0.0207 0.0077 0.0076 Cottaee Grove Not Prescribed 0.0026 All Participants -0.0012 0.0096 0.0092 Decatur Not Prescribed 0.0317 All Participants 0.0331 0.0117 0.0108 1. Adjusted for Ln Age, Ln BMI, Ln Alcohol Ln T3 .04 .04 .005 .007 .79 .90 .008 .003 0.0105 0.0099 0.0053 0.0051 .05 .05 0.0216 0.0204 0.0077 0.0076 .01 .01 0.0006 -0.0001 0.0099 0.0095 .95 .99 0.0271 0.0293 0.0119 0.0109 .02 .01 Table C8. Number and Percent Subjects Above or Below Reference Points for Thyroid-Related Clinical Chemistry Results, by PFOA Decile, 2000 Fluorochemical Medical Surveillance Program (N = 552) PFOA Decile TSH (ulU/mL) T4 (uc/dL) <0.25 >5.5 <4.5 >12.0 N (%) N (%) N(%) N (%) 1 2 3 4 5 6 7 8 9 10 p value 0(0) 0(0) 1(2) 0(0) 0(0) 0(0) 1(2) 2(4) 0(0) 4(7) 0(0) 1(2) 0(0) 2(4) 0(0) 6(10) 0(0) 4(8) 1(2) 1(2) .67 .03 0(0) 0(0) 0(0) 0(0) 1(2) 0(0) 0(0) 0(0) 0(0) 1(2) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) .41 .49 Free T4 (ng/dL) <0.70 >1.53 N (%) N (%) 0(0) 1(2) 0(0) 1(2) 0(0) 0(0) 1(2) 1(2) 1(2) 0(0) 0(0) 0(0) 0(0) 0(0) 1(2) 2(3) 0(0) 0(0) 0(0) 0(0) .68 .47 T3 (ne/dL) <60 >181 N(%) N (%) 0(0) 1(2) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 2(4) 0(0) 1(2) 0(0) 2(3) 0(0) 0(0) 0(0) 2(3) _ .38 Table C9 . Predicted Thyroid-Related Clinical Chemistry Results Based on Multiple Regression Models for 40 Year Old Male with BMI = 28 and Drinks 0.5 Alcohol Beverages per Day (N = 552) Predicted Predicted Predicted Predicted TSH (pIU/mL) T4 (pg/dL) Free T4 (ng/dL) T3 (ng/dL) Reference Ramie (0.25-5.5) (4.5-12.0) (0.70-1.53) (60-181) PFOA Serum Concentration (pg/mL) 0.005 1.57 8.28 1.16 119 0.01 1.61 8.24 1.15 120 0.10 1.77 8.10 1.12 123 0.50 1.89 8.00 1.09 125 1.00 1.94 7.97 1.08 125 5.00 2.07 7.88 1.06 127 10.00 2.13 7.84 1.05 128 50.00 2.27 7.74 1.03 130 100.00 2.33 7.71 1.02 131 Appendix D Table D l. Correlation Coefficients (same as Table Al) Ln (Variable) Ln (PFOA) LniPFC PFOA PFOS Age BMI Alcohol Cholesterol LDL HDL Triglycerides Glucose Aik Phos AST ALT AST/ALT GGT Total Bilirubin Direct Bilirubin TSH T4 Free T4 T3 1.0 0.55**** 0.03 0.11* -0.12** 0.04 0.00 -0.17**** 0.210**** -0.01 0.08 -0.01 0.12** -0.15*** 0.10* -0.18**** -0.06 0.09 -0.03 -0.15** 0.09* 1.0 0.06 0.00 -0.10** 0.08 0.06 -0.07 0.13** -0.13** 0.08 0.00 0.08 -0.09* 0.10* -0.13** -0.03 0.04 0.002 -0.06 0.03 *p < .05 ** p < .01 ***p<.001 ****p< .0001 Ln(Age) 1.0 0 .2 0 **** -0.15*** 0.15*** 0.13** -0.13** 0.18**** 0.26**** 0.08 0 .0 0 0.03 -0.05 0.15*** -0.08 -0.03 -0 . 0 1 -0.09 -0.17*** -0 .10* Ln(BMP Ln(Alcohol) 1.0 -0.26**** 0.00 -0 .0 2 -0 43 **** 0 4 4 **** 0.28**** 0 .1 2 ** 0.15*** 0.41**** -0.39**** 0.31**** -0.18**** -0.03 0.08 -0.05 -0.17*** 0.02 1.0 0 .10* 0.01 0.37**** -0 .12** -0.19**** -0.25**** -0.04 -0.16*** 0.17*** 0 .2 0 **** -0 .0 2 -0 .0 2 -0 .0 2 -0.17*** 0.08 -0 . 0 1 Age (years) by PFOS (ng/mL) LN Age (years) by LN PFOS (jig/mL) In PFOS BMI by PFOS (ng/mL) 50-1 * 45- 0 1 2 3 4 5 6 7 8 9 10 PFOS LN BMI by LN PFOS (^ig/mL) ln P F O S Alcohol (drinks/day) by PFOS (jig/mL) 6- 5- 0 123456 PFOS LN Alcohol (drinks/day) by LN PFOS (jig/mL) Glucose (mg/dL) by PFOS (ug/mL) G LU C O SE i In G L U C O S E Cholesterol (mg/dL) by PFOS (ng/mL) LN Cholesterol (mg/dL) by LN PFOS (jig/mL) LDL (mg/dL) by PFOS (jig/mL) 01 23456789 PFOS 10 LN LDL (mg/dL) by LN PFOS (|ig/mL) ln PFOS HDL (mg/dL) by PFOS (jig/mL) LN HDL (mg/dL) by LN PFOS (ng/mL) Triglycerides (mg/dL) by PFOS (jig/mL) LN Triglycerides (mg/dL) by LN PFOS (ng/mL) Alkaline Phosphatase (IU/L) by PFOS (ug/'mL) 160 " 1 34 PFOS LN Alkaline Phosphatase (IU/L) by LN PFOS (pg/mL) AST (IU/L) by PFOS (jig/mL) LN AST (IU/L) by LN PFOS (|ig/mL) In PFOS ALT (IU/L) by PFOS (jig/mL) LN ALT (IU/L) by LN PFOS ftig/mL) / AST/ALT AST/ALT by PFOS (ng/mL) LN AST/ALT by LN PFOS (jig/mL) In PFOS InA ST/A LT GGT (IU/L) by PFOS (|xg/mL) LN GGT (IU/L) by LN PFOS (ug/mL) Total Bilirubin (mg/dL) by PFOS (jig/mL) 1-- -- i-- -- i-- -- i-- 1-- i-- 1-- i-- 1-- r 0 1234 5 6 PFOS LN Total Bilirubin (mg/dL) by LN PFOS Qig/mL) -4-3-2-1012 In P F O S Direct Bilirubin (mg/dL) by PFOS (.ug/mL) LN Direct Bilirubin (mg/dL) by LN PFOS (ug/mL) TSH (fiIU/mL) by PFOS (jig/mL) LN TSH (nIU/mL) by LN PFOS (ng/mL) T4 (ug/dL) by PFOS (jig/mL) *. 11-. -.y _ :- .- .i- . is s i - y w - -'e-?- : . . - 6 -f` \ \ 5-T ' ; 4 ^ U 123 4 PFOS LN T4 (ug/dL) by LN PFOS (jig/mL) In PFOS Free T4 (ng/dL) by PFOS Qig/mL) LN Free T4 (ng/dL) by LN PFOS (ng/mL) 0.5 0.3 0.1 -0.3 -0.5 -0.7 -4 -3-2-10 In P F O S 12 T3 (ng/dL) by PFOS (|xg/mL) LN T3 (ng/dL) by LN PFOS (gg/mL) Appendix E Table El. Non-adjusted and Adjusted Regression PFOS Coefficients for Lipid Clinical Chemistry Results, 2000 Fluorochemical Medical Surveillance Program (N = 506) Non-adjusted Ln PFOS Coefficient SE p value Adjusted Ln PFOS Coefficient1 SE p value Ln Cholesterol All Locations 0.0151 Antwerp 0.0091 Cottage Grove 0.0151 Decatur 0.0247 0.0081 0.0144 0.0130 0.0173 .06 .53 .25 .15 Ln LDL 0.0157 0.0049 0.0141 0.0235 0.0081 0.0128 0.0134 0.0173 .05 .70 .30 .18 All Locations 0.0153 Antwerp -0.0068 Cottage Grove 0.0115 Decatur 0.0457 0.0120 0.0211 0.0198 0.0249 .20 0.0135 .75 -0.0137 .56 0.0187 .07 0.0421 Ln HDL 0.0120 0.0194 0.0203 0.0245 .26 .48 .36 .09 All Locations -0.0161 Antwerp -0.0014 Cottage Grove -0.0018 Decatur -0.0242 0.0108 0.0185 0.0179 0.0186 .14 .94 .92 .20 -0.0093 -0.0020 -0.0114 -0.0166 0.0094 0.0175 0.0174 0.0172 .33 .91 .52 .34 All Locations 0.0778 Antwerp 0.1066 Cottage Grove 0.0453 Decatur 0.0416 Table El. (continued) 0.0255 0.0381 0.0463 0.0501 Ln Triglycerides .002 0.0752 .006 0.1063 .33 0.0393 .41 0.0304 0.0230 0.0357 0.0455 0.0463 .001 .003 .39 .51 1. Adjusted for Ln Age. Ln BMI, Ln Alcohol Tabic E2. Non-adjusted and Adjusted1-2Regression PFOS Coefficients for Hepatic Clinical Chemistry Results. 2000 Fluorochemical Medical Surveillance Program (N = 506) Non-adjusted Ln PFOS Coefficient SE p value Adjusted Ln PFOS Coefficient1-2 SE p value Ln Alkaline Phosphatase All Locations 0.0189 Antwerp -0.0051 Cottage Grove -0.0137 Decatur 0.0405 0.0112 0.0185 0.0165 0.0238 .09 .78 .41 .09 Ln AST 0.0130' 0.00742 -000281 -0.01402 -0.00091 0.00912 0.0381' 0.03462 0.0110 0.0109 0.0186 0.0185 0.0170 0.0171 0.0237 0.0234 .24 .50 .88 .45 .59 .59 .11 .14 All Locations -0.0006 Antwerp -0.0356 Cottage Grove 0.0081 Decatur 0.0111 0.0117 .96 0.0183 .05 0.0210 .70 0.0250 .66 0.00051 -0.00632 -0.0370' -0.04202 0.01401 0.01232 0.0103' 0.00822 0.0116 0.0117 0.0182 0.0184 0.0214 0.0216 0.0247 0.0245 .97 .59 .04 .02 .51 .57 .68 .74 All Locations 0.0354 Antwerp -0.0490 Cottage Grove 0.0510 Decatur 0.0927 All Locations 0.0534 Antwerp 0.0287 Cottage Grove 0.0417 Decatur 0.0583 Table E2. (continued) Ln ALT 0.0195 .07 0.0295 .10 0.0311 .10 0.0376 .01 0.0343' 0.01062 -0.04991 -0.07342 0.04901 0.04092 0.0826' 0.07802 0.0179 0.0183 0.0293 0.0289 0.0299 0.0310 0.0351 0.0347 .06 .58 .09 .01 .10 .19 .02 .03 Ln GGT 0.0237 .02 0.0416 .49 0.0425 .33 0.0433 .18 0.0558' 0.02512 0.0260' -0.00882 0.0321' 0.01522 0.0612' 0.05242 0.0225 .01 0.0217 .25 0.0291 .51 0.0389 .82 0.0414 .44 0.0401 .71 0.0424 .15 0.0401 .19 Table E2. (continued) Ln Total Bilirubin All Locations -0.0415 Antwerp -0.0344 Cottage Grove -0.0142 Decatur -0.0531 0.0138 .003 0.0240 .15 0.0201 .48 0.0254 .04 -0.03561 -0.027512 -0.03541 -0.03242 0.01071 0.01422 -0.05651 -0.05252 0.0136 0.0135 0.0241 0.0247 0.0206 0.0205 0.0255 0.0250 .01 .04 .14 .19 .61 .49 .03 .04 1. Adjusted for Ln Age, Ln BMI, Ln Alcohol 2. Adjusted for Ln Age, Ln Triglycridcs, Ln Alcohol Table E3. Non-adjlisted and Adjusted1Regression Coefficients for Thyroid Clinical Chemistry Results. 2000 Fluorochemical Medical Surveillance Program (N = 506) Non-adj usted Ln PFOS Coefficient SE p value Adjusted1 Ln PFOS Coefficient SE p value Ln TSH All Locations 0.0227 Antwerp 0.0306 Cottage Grove 0.0224 Decatur -0.0132 All Locations 0.0029 Antwerp -0.0183 Cottage Grove 0.0010 Decatur 0.0101 All Locations -0.0080 Antwerp -0.0217 Cottage Grove 0.0108 Decatur -0.0015 0.0275 0.0447 0.0442 0.0620 0.0073 0.0133 0.0104 0.0156 0.0059 0.0105 0.0083 0.0129 .41 0.0232 .49 0.0369 .51 0.0094 .83 -0.0125 Ln T4 .69 0.0004 .17 -0.0179 .92 0.0067 .52 0.0062 Ln Free T4 .18 -0.0064 .04 -0.0208 .20 0.0126 .91 -0.0019 0.0228 0.0446 0.0451 0.0626 0.0072 0.0131 0.0105 0.0155 0.0058 0.0104 0.0082 0.0129 .40 .41 .83 .84 .95 .18 .51 .69 .27 .05 .13 .88 Table E3. (continued) All Locations 0.0053 0.0071 Antwerp 0.0173 0.0106 Cottage Grove -0.0200 0.0135 Decatur 0.0325 0.0147 1. Adjusted l'or Ln Age. Ln BMI, Ln Alcohol Ln T3 .46 .10 .14 .03 0.0061 0.0186 -0.0190 0.0304 0.0071 0.0104 0.0138 0.0146 .39 .08 .17 .04