Document MomRKxx0La2gkkgQkaqYvx9Yz

slh(ymdHwml Robolt A. Tdl ldontorkr 4676 Cdumbla Pwkmy Cindnnrti OH 452261900 July 31,2001 HETA 20010235 Brian Lewis PACE International Union 3340 Perimeter Hill Drive Nashville, Tennessee 3721 1 Dear Mr.Lewis; This letter is in response to your request to the National Institute for OccupationalSafety and Health (NIOSH) for a Health Hazard Evaluation (HHE). Your request concerned exposure to pefluorooctanoicacid (PFOA)and pedluorooctanesulfonate(PFOS)among employees ofthe 3M Corporationin Cottage Grove, Minnesota. The request stated that employees had elevated blood fluoride levels. No health effectswere noted: "he request apparently stemmed h m work~rcon& ngarding recent institutionofenginaring contmb amithe use o f p a s o d protective equipmemt (PPE) among employees inthe fluorochemicaldcp9lrtmglt. In respOnse.to this request, I conductedtelephoneinterviewswith the SafetyDirector fbr the flmrochemid . ' department, the local Unionpresident, and the Director of OccupgtionalMediciae for 3M Corporation. This letter summarh the findings of my evaluation and gives you information addressing your concam. Background 3M Cottage Grow, Mmesota The 3M f d t y in CottageGrovehouses several divisions,one of.which is the Specialty Materials Markets Group (pnwiously known as the Chemical Division). The fluorochemical operationsare one part of the Specialty Materials Markets Group. Production of ammoniumperfluorooctanoate (APFO),a synthetic surfactant, began in Cottage Grove in 1947. APFO dissociates rapidly in aqueous sdution to PFOA. APFO is a white powderproducedby an electrochemical process. It is then isolated and Converted to a salt slurry, which in turn is converted to a salt cake, which is dried and It was reportedthat the electrochemical cell used in the production of APFO is no longer running, but that drying and packaging of APFO still occurs in the Cottage Grove plant. Other fluomch&micalsmanhtured at Cottage Grove have perfluorooctane sulfonyl fluoride (POSF) as a precursor in a process similar to that described for PFOA. These compounds are transformed after absorption in the body to PFOS. Primary production of PFOS compounds has been at 3M facilities in Antwerp, 3M Cottage Grove. HETA 20010235 Page 2 Belgium and Decatur, Alabama, with production in Cottage Grove being a much smaller operation. Therefore, published literature regarding health effects of PFOS have primarily included workers fromAntwerp and Decatur, but should be applicableto Cottage Grove POSF production workers as well. POSF production was recently stopped in the Decatur facility and both POSF and APFO production are being phased out by 3M.There are currently approximately 33 workers in building 15, which is the main APFO production building and 10 to 20 workers in building 25, which the POSF production area Voluntary medical surveillancehas been performed biennially since the 1970son fluoqxh&cal workers. Participation rates in the medical surveillance program during the 1990s HW 50-?oO/0.2 Suweillancecorisists ofa medical questionnaire, sphmetry, a chemistry panel, urinalysis,and' vital signs. Total serum fluorinelevels were measured until theearly 1990swhen methods became available to a n a l p serum fir specific fluorochcmi&. f .. Serum PFOA levels in CottageGrove employeesmged fiom 0-80 parts per million (ppm) with a mean of5 ppm and a median of 1.1 ppm in 1993, fiom 0-114.I ppm with a mean of6.8 ppm and a median of 1.2 ppm in 1995, and from 0.1-81.3 ppm, with a mean of 6.4 ppm and a median of I .3 pprn in 1997.2 Serum PFOS levels in workers in 3M's Antwerp,Belgium, and Decatur, Alabama facilitiesranged h m 0-12.8 ppm (mean 2.49 ppm) in 1995 and 0.1-9.9p@n(mean 1.75 ppm) in 1997? Levels in 3M employees in corporate officesavrrBgcd 0.05 ppm." Serum fluorinelevels in the generalpopulation are reported to range fiom0.01-0.13 ppm? Health Eflects of PFOA Animal studies demonstrate that PFOA is absorbed by ingestion, inhalation, and dennal exposure and is not metabolized. It is eliminated primarily throughthe urine and feces. PFOA has an estimated half-life of 18-24 months in humans? PFOA increases the incidence of liver, +mas, and Leydig cell adenomas in lab animds.'3 The paucrcatic denomas may be due tomild elevations in cholecystokinin. PFOA also increased estradiol levels and decnased kstostmne levels in rats? it c.aused marked hepatic effects in animals, including liver enlapment, focal necrosis, changes in lipid metabolism,and reduced levels of lipids6 Several epidemiologic studies have been conducted on PFOA workers at the Cottage Grove facility utilizing biennial medical surveillance data. Cholecystolcinin levels were inversely related to serum PFOA levels, not elevated as seen in animal studies? While one study found positive associations between serum fluorine levels andzstradiol, prolactin, and thyroid stimulating hormone (TSH), and a negative relationship between serum fluorine and testosterone,' another did not find a significant relationship between serum PFOA levels and these hormones.' Gilliland and Mandel reported that in unadjusted analyses, total serum cholesterol, high density lipoprotein O L ) , low density lipoprotein (LDL);'serum glutamic pyruvic transaminase (SGPT), gamma glutamyl transferase (GGT), or serum glutamic oxaloacetic-rt @GOT)were not correlated with total serum fluorine levels! ubel et. al. found no evidence of adverse hepatic effects related to exposure to fluorochemi~als,~ 3M Cottage Grove. HETA 20010235 Page 3. ,4 retrospective cohort mortality study of Chemical Division workers employed during the time period 1948-1978 was performed.' This included bothAPFO and POSF workers. There was no excess of all-cause mortality, cancer-specificmortality,or any other cause of death. An update of this cohort through 1983revealed an excess of prostate cancer mortality among men employed at least 10 years.' This was based on 4 prostate cancer deaths in the exposed workers. A M e r update of only PEOA workers through 1997revealed no excess prostate cancer mortality risk, but did show a statistically nonsignificant increase in cerebrovasculardisease mortality,although there was no dose-responserelationship? Mortality rates for all cams and for all cancers wcrc lower than expected in the general population. Health Efects of?FOS PFOS concentrates in the liver and.isexcreted in both the rniiieandfeces?.PFOS has an C estimated half-life in serum of 1OOO-1500days in humans.' Exposure in animalsresults in decreased serum cholesterol levels and metabolic wasting? A study ofPFOSproduction workers did not reveal statistically significant changes in hepatic emzymes, cholesterol, or other lipoproteins at serum PFOS levels less than 6 ppm? Very few employees had serum PFOS levels higher that 6ppm, therefore it was not possible to draw conclusions regarding exposuxe at those levels. In a cohort mortality study ofworkersemployed in the Alabama plant during theperiod 1961- . 1991, the allcause mortality mtes were statistically significantly lower thanthegeneral population.lo Rates were also lower for all cancers combined but elevated for cancer of the lung, bronchus, and trachea, however, this elevation was not statistically significant Cohort mortality was updated through 1998, and the all-cause mortality ratcswerc still significantlylower thaathe general population." However, there were threedeaths from bladder cancer, resultingin a statistically significantly elevated rate. The author stated that it was unclear ifthese can be attributed to fluorochemicalexposure, other occupationalexposures, or non-occupationalfktors. E X P O S CU p~ itdi The American C o d i c e of Governmental Industrial Hygienists' (ACGMm) Threshold Limit Value (TLV@) for APFO is 0.01 milligrams per cubic meter of air (mg/m3)as an 8-hour timeweighted average (TWA).'* There is no NOSH Recommended Exposure Limit (REL) or U.S. Department of Labor, Occupational Safety and Health Administration (OSHA)Permissible Exposure Limit (PEL)for APFO. The exposure limitfar POSF set by 3M is 0.1 mg/m3. 3M developed an Occupational Biologic Limit Value (BLV) of 5 ppm for PFOA in March of 2OOO." A BLV is defined as the best estimate of a level of a chemical substance or its rnetabolite(s) in a biological fluid that if present, even on a chronic basis, d u l d not be expected to pose, or correlate with,a significant risk of adverse health effects. Workers with s e m PFOA levelsbetween 5 and 10pprn undergo medical surveillance every 6 monthscompared to every 2 years for routine surveillance, but are allowed to remain in the work area as long as serum levels 3M Cottage Grove. HETA 200 10235 Page 4 decline on subsequent testing. Workers with levels > 10 ppm are removed from the area and monitored. 3M recently undertook an industrialhygiene assessment of the fluorochemical manufacturing process at the Cottage Grove facility. They identified work areas and tasks with potential exposures above the TLV for APFO. In response to these findings, process operator's job tasks were modified so that they were removed from the process areas and stationed primarily in an isolated control room. Entry into the process areas now requires PPE consisting of full-face respirators (either supplied air or air purifling, depending on the area), coveralls (eitheq Tp&@ or rubber, depending on the area), nitrile gloves, and rubber boots. Access to the prcrcess areas is througha decontamination room. A large numberoftasks withpotential exposures above the TLV for APFO were identified, and action taken to reduce exposure iacluding cxhaust ventilation, machineenclosure,improved housekeeping, and'improvedWOWtraining. None of the environmental air samples in mn-process areas of the building, including the contml room and decontamhation rooms,had APFO concentrations above 0.002 mg/m3. Enghccring controls and PPE similar to those added in the APFO area are utilized in the POSF production area. Conclusions and Recommendations 3M has an extensive medical suxveillance and industrial hygiqne monitoring program for fluorochemical workem at the Cottage Grove facility. Work arcas have a combination of engineering ControIs, personalprotective equipment, and administrativecontrols in place to reduce exposure to fluomchcmicals. 3M has conducted or spoasorcd nummus animal and human health effects studies related to fluorochemical exposure. Most importantly, these fluorochemicai mandactmhg processes at Cottage Grove ani b e i i actively phased out.. Exposures to PFOA and PFOSat the levels encounted at 3M facilitieshave not been Bssoci(lfcd with ad- health effkcts. An epidemiologic study of POSF workers has identified a pomile relationship be& occupdonal exposureto POSF and bladdcr cancer. It shouldbe noted that mortality studies may u&mtm.ak the incidence ofbladder caner since it is often not fittal. Conversely, other literature on the health effects of POSF has not indicated that it is a bladder carcinogen. Given the above, no conclusion can be drawn at this time concerning occupational exposures to POSF and bladder cancer. I recommend that 3M continuecurrent medical surveilhce even after the phase-out of these products is complete, in order to more clearly ascertain the potential health effects of exposure to PFOA and PFOS. In addition, 3M should continue to update cohort mortality studies of these employees. 3M should closely follow available epidemiologic and clinical information regarding the potential risk of bladder cancer in workers exposed to POSF.''Informaton' available in the hture may help determine if screening for bladder cancer among exposed employees would be useful. . ,. ...?.. 3M Cottage Grove. HETA 20010235. * Page 5 I hope this information is helpful to you. This letter serves to close the file for this'project. For the purpose of inform& affected employees, copies of thisreport should be posted by the employer in a prominent place accessible to the cmployecs for a period of 30 calendar days. If you haveany fbthcr qucstioly please call me at (5I3) 8414 386. Sincerely youq .. Elma H.Page, MD.,M.P.H. . .i . Divisionofs~ance,Hapad Evaluitions and Field Studies .. -.. , References: 1. Olsen GW, Gilliland FD, Burlew MM, Burris JM, Mandel JS, Mandel JH (19981. An epidemiologic investigation of reproductive hormones in men with occupational exposure to peffluorooctanoic acid. J Occup Environ Med. 40(7):6 14-622. 2. Olsen GW, BurrisJM,Burlew MM, Mandel JH [2000]. Plasmacholecystokininlevels and hepatic enzymes, cholesterol, and lipoproteins in ammonium perfluorooctanoate production workers. Drug and Chemical Toxicology. 23(4):603-620. 3. Olsen GW,Bunis JM,Mandcl JH,Zobel LR [1999]. Serum perfluorooctanesulfonatcand hepatic and lipid clinical chemistry tests in fluorgchemical production workers. J &a& Envi-on Med. 41(9):799-806. -$&: 4. 3M [2000]. Dratt initialassessmentreport: pcrfl&ooc& sulfonc acid and its*s&. Submitted to the United States EnvironmentalProtectionAgency. - 5. UbeI'FA,Sorenson SD,Roach DE [1980]. Hcalth status ofplant workers exposed to fluomchemicals a preliminary report. Am Ind Hyg Assoc J 41584589. 6. Gilliland FD,Mandel JS [19961. Serum perfluoroQctanoic acid and hepatic enzymes, lipoproteins, and cholesterol: a study of occupationallyexposedmen. Am J Ind Med. 29560- 568. 7. Gilliland FD [1992]. Fluorocarbonsand human heal*. studies in an occupationalcohort. Doctoral dissertation, University ofMinnesota In: Olsen GW, W a a d FD, B d c w MM, Burris JM,Mandel JS,Mandel JH [1998]. An epidemiologic investigationof reproductive hormones in men with occup8tionalexposure to perfluorooctanoicacid. J 0ccup EnvirOnMad. 40(7):614-622. 8. Gilliland FD,Maude1JS [19931. Mortality among employees of aperfluomoctanoicacid production plant. J Occnp Environ Med. 35(9):950-954. 9. Alexander BH [2001]. Mortality study of workers employed at 3M Cottagc Grove Mlity. University of Minnesota School of Public Health. 10. Mandel JS, Johnson RA [1995]. Mortality study of employees at 3M plant in Decatur, Alabama. h:3M [2000]. Draftinitial assessment report: perfluorooctanesulfonic acid and its salts. Submittedto the United StatesEnvironmental P-rptectionAgency. 11. Alekinder BH [20011. Mortality study of workers employed at 3M Decatur facility. University of Minnesota School of Public Health. 12. ACGIH [2OOl). 2001 TLVs@and BEIS@:threshold limit values for chemical substances and physical agents. Cincinnati, OH: American Conference of Governmental Industrial Hygienists. 13. 3M [2000].Documentation of occupational biologic limit value for perfluorooctanoate.