[NOTE: The original letter submitted to the FDA contains appendices which a to request a hard copy of this document with appendices attached, please contact Consumers Unions Yonkers Office.]
June 5, 1998
Joseph A. Levitt
Director, Center for Food Safety
and Applied Nutrition, HFS-1
U.S. Food and Drug Administration
200 C Street, SW
Washington, DC 20204
Dear Mr. Levitt:
We are writing to bring to your attention the results of recent tests conducted by Consumers Union that bear on the safety of foods wrapped in certain types of plastic packaging.
As part of a project on endocrine disrupters in foods (see CONSUMER REPORTS, June 1998, pages 52-55), we tested cheeses wrapped in several kinds of plastic, to determine whether plasticizers present in the wrapping had migrated into the cheese. We found various results with different types of plastic packaging. Samples of individually-wrapped slices of American cheese and of products in laminated foil/plastic wrapping (e.g., Cracker Barrel cheddar cheese) contained no plasticizers attributable to packaging. But cheeses wrapped in PVC film in stores contained very high levels of the plasticizer DEHA [di(2-ethylhexyl)adipate]. Seven samples with this type of wrapping contained 51 to 270 ppm (average 153 ppm) of DEHA, which clearly had leached from the PVC film.
Many of our samples contained phthalate plasticizers, in particular dibutylphthalate (DBP) and di(2-ethylhexyl)phthalate (DEHP). The DEHP concentrations were as high as 3.7 parts per million (ppm). However, the wraps used on these cheeses did not contain either phthalate as a primary plasticizer, and concentrations in the cheeses could not be clearly attributed to migration from the wraps. Other possible sources include glues and inks used on the printed labels, and background environmental contamination in the cheese itself. Both DBP and DEHP have been shown to be present in dairy foods independent of contact with plastic wraps.
Our sampling was limited; we tested only 19 products, purchased in a variety of supermarkets and smaller stores in the New York metropolitan area. Despite that limited scope, we found evidence that plasticizers occur widely in cheeses. In particular, the consistent high levels of DEHA we found in cheese packaged in PVC cling-wrap deserve priority attention.
While there is no conclusive evidence that the levels of DEHA we found in cheese are harmful, per se, there are reasons for concern about the potential health effects of this plasticizer. DEHA is a teratogen and causes reproductive effects in animal studies. In the NTP carcinogenesis bioassay, it induced liver tumors in mice. It has not been adequately evaluated for its potential endocrine effectsa risk we consider potentially more significant than carcinogenicity. Because of its reproductive and teratogenic effects at high doses, we believe it needs careful evaluation for possible effects on the endocrine system, and that a precautionary approach is appropriate until better data are available on that question. Frankly, we consider 153 ppm a startlingly high level of a packaging migrant in any food, let alone one like cheese that is consumed in large quantities by children.
In Europe, the EEC Scientific Committee for Food has recommended a tolerable daily intake of 0.3 mg/kg/day for DEHA, and the Commission of the European Communities has established a provisional specific migration limit of 18 ppm for DEHA migration from plastic wraps into foods. DEHA levels we found in cheeses exceed this health-based SML by a wide margin, and could result in many consumers’ exceeding the TDI.
The problem of DEHA migration from PVC wrap into cheeses has been documented in several other countries. In the UK, where the issue first was reported in 1987, the problem was largely eliminated in a few years by substituting polymerized plasticizers for DEHA in PVC film.
The phthalates, principally DEHP, that we found in most cheeses at levels up to 3.7 ppm, also raise some health concerns. Among other effects, some of the phthalates have shown estrogenic activity in animal studies. It is not possible, given the limited state of current knowledge, to specify a safe level of exposure to hormone-mimicking chemicalsespecially, again, in foods like cheese, which is consumed in large quantities by children. The levels of DEHP we found cannot be presumed to be free of risk, and in our view efforts to keep phthalates out of foods like cheese are worthwhile. On the other hand, since there are multiple sources of potential contamination of cheeses with these compounds, simple changes in packaging materials are unlikely to solve this problem. Nevertheless, our results suggest that it may be possible to reduce phthalate migration by paying attention to factors like the inks and adhesives on packages and labelssome of which may be sources of the phthalate contamination we found.
We offer you here the details behind our magazine report. We request that the FDA and responsible sectors of the food and packaging industries promptly investigate the problems our testing has brought to light, especially DEHA migration from PVC cling-wrap into cheeses, and take appropriate steps to eliminate these avoidable exposures to plasticizers in our food supply.
A report on our tests and findings is attached. We will be happy to meet with the FDA and other interested parties to discuss effective solutions to the problems our tests have revealed. If you have questions or would like additional details, please call Dr. Groth at 914-378-2301.
Edward Groth III, PhD Mark Silbergeld
Director, Technical Policy Co-Director
And Public Service Washington Office
cc: Jerome Heckman, Keller & Heckman
Dane Bernard, NFPA
Courtney M. Price, CMA
Dagmar Farr, FMI
BACKGROUND ON CUS TESTS
BACKGROUND ON CUS TESTS
During the period May-September 1997, we purchased samples of 19 cheeses wrapped in several different types of plastic packaging. All samples were obtained in New York City or nearby suburbs. We bought cheeses with similar fat content; most were cheddar, to minimize variability due to cheese rather than packaging. We aimed to compare different packaging materials for plasticizer migration; we were not concerned with comparing brands of cheese for plasticizer content. Any brand names mentioned here are noted only to identify the packaging material by providing specific examples.
Samples were sent to a contract laboratory experienced in the analysis for migration of plasticizers from packaging into foods. Three products, in three different types of packaging, were tested in a pilot phase in June 1997, and 16 additional products were tested in October 1997.
Plastic wrappers were analyzed to identify the plastic polymer type and to determine the identity and quantity of plasticizers present. Cheeses were then analyzed for plasticizer content. Details on the analytical methodology followed are provided in an excerpt from the contract lab’s report, attached as Appendix A.
In the pilot phase, the plastic packages were extracted with solvents and the extracts were analyzed for plasticizers first by gas chromatography-mass spectrometry (GC-MS). A confirming analysis was done using nuclear magnetic resonance spectroscopy. Plasticizers were identified by comparing the resulting spectra to standard library spectra of additives. In subsequent full-scale tests, GC-MS alone was used, as the initial round of tests affirmed its accuracy.
The cheeses were extracted and the extracts analyzed by GC-MS (in the pilot study) or by liquid chromatography-mass spectrometry (LC-MS) in the full study. In the pilot phase, each cheese product was homogenized, and the average plasticizer content of the cheese as a whole was determined. In the full-scale tests, the outer surface was cut from each sample to a depth of 0.5 to 1 cm, and "surface" and "core" portions were analyzed as separate samples. Comparable portions from three samples of each product were combined into composite "surface" and "core" samples for this analysis.
Packaging Types and Materials Examined
Our samples included a variety of brands or products packaged in the following types of plastic wrapping:
Individually wrapped slices. For example, Kraft Singles, Borden Singles. The cheese is pasteurized cheese food, formed into square slices. Each slice is individually wrapped in a plastic envelope, and an outer plastic package with a printed label contains 16 or 24 slices.
Laminated foil wrapper. A plastic/mylar laminated wrapper with a metallic appearance, vacuum-packed, as used on Kraft Cracker Barrel Sharp Cheddar.
Vacuum-packed clear plastic film. A variety of products wrapped in a semi-rigid plastic envelope with a printed label, which may be in tight contact with the cheese surface (as Land O Lakes Sharp Cheddar, or Kraft Natural Monterey Jack) or more loosely packed (as with Sargento Sliced Muenster).
Cling-wrap stretch film. Plastic stretch film with an adhesive label, typically used to wrap random-weight blocks of cheese cut in stores. Those we tested included several varieties of New York State Cheddar, purchased in supermarkets and delicatessens; a Wisconsin Colby Cheddar, a Black Diamond Canadian Cheddar, and an Ilchester Double Gloucester imported from England.
Red plastic film. Vacuum packed, with adhesive printed label, used on Millers Kosher Baby Gouda. Looks like red wax, but is plastic.
Wax coating with plastic outer wrap. We included two products in which
the cheese itself is coated in wax, and a film of plastic shrink-wrap
is on the outside of the plastic, with either a printed label
on the plastic (as for Bonbel Semi-Soft Cheese) or an adhesive-backed
printed paper label (as for Classic Vermont Cheddar).
Plastics Used in Packaging Materials/Abbreviations Used
EVA ethylene/vinyl acetate copolymer
HDPE high-density polyethylene
PET polyethylene terephthalate
p-PVC plasticized poly(vinylchloride)
RCF regenerated cellulose film
Plasticizers Tested For/Abbreviations Used
Tables presenting complete data on individual samples tested in both the pilot phase and the full-scale tests are attached, as Appendix B. In this section, results are summarized and highlights are presented.
(1) Analysis of Packaging Materials
Individually wrapped slices. The outer wrappers were PP; the inner wrappers were PP (Borden) or HDPE (Kraft). The outer wrappers contained DBP and DEHP at low levels, indicating that the phthalates were not present as plasticizers; they might be associated with inks on the labels.
Laminated foil wrapper. Outer (clear) layer is PE; inner (foil-like) layer is EVA. Plasticizers detected included DBP, DEHP and DEHA, all at low levels, not present as primary plasticizers.
Vacuum-packed clear plastic film. Five different products in this type of package were tested. All five packages were laminated, with EVA on the food contact side. Four of the five had PET outer layers; the fifth had a PP outer layer. All five packages contained traces of DBP and DEHP, and one also contained DiBP; as above, not present as primary plasticizers.
Cling-wrap stretch film. One sample of this packaging type was tested in the pilot phase, and five more samples in the full study. In all cases the plastic film was p-PVC, and a primary plasticizer, DEHA, was present at levels ranging from 5,499 to 21,729 ppm. Low levels of DBP were found in five samples. Four of the five also had low levels of DEHP, while the fifth had 825 ppm of this phthalate. One sample had a trace of DiBP.
Red plastic film. One product of this type was tested. The plastic wrap was EVA. It contained low levels of DBP and DEHP.
Wax coating with plastic outer wrap. The outer coating on the Bonbel Semisoft cheese was identified tentatively as RCF. It contained low levels of DBP and DEHP. The wax-coated Vermont Cheddar had a clear outer wrap of p-PVC, plasticized with DEHA at 5,539 ppm. This outer film also contained low levels of DBP and DEHP.
(2) Analysis of Cheeses
Individually wrapped slices. Very low levels of DBP, DiBP and DEHP (all well below 1 ppm) were found in the three products of this type that we tested.
Laminated foil wrapper. Very low levels of DBP and DEHP (less than 1 ppm) were found in the one product tested with this kind of wrap.
Vacuum-packed clear plastic film. The five samples of cheeses in this packaging all contained DBP at levels well below 1 ppm, and DEHP at levels up to 2.1 ppm. One sample also contained a trace of DiDP. DEHP levels in all cases were greater in surface samples (those that had been in contact with the wrapper) than those in "core" samples, by factors ranging from 2- to 7-fold.
Cling-wrap stretch film. We tested two samples of Ilchester Double Gloucester wrapped in this PVC film in the pilot phase, and found 200 and 270 ppm of DEHA, respectively, in the homogenized samples. Five more samples, four domestic cheddars and one Canadian cheddar, were tested in the full study. They contained the following amounts of DEHA:
Sample DEHA Concentrations (ppm)
These samples also contained traces of DBP (five samples) and DEHP (all seven). One sample contained a trace of BBP, two had minute amounts of DiBP, and one sample had a trace of DiDP. Phthalate concentrations in surface samples were generally higher than those in core samples.
Red plastic film. Like most other products tested, the single Gouda cheese wrapped in this packaging contained traces of DBP and DEHP, and levels in surface samples tended to be higher than in core samples.
Wax coating with plastic outer wrap. The product with the RCF outer wrapping contained low levels of DBP and DEHP, most likely due to background contamination of the cheese itself. The wax-coated cheddar with a p-PVC outer wrap contained a trace of DBP and 3.7 ppm of DEHP, the highest average concentration of DEHP found in any sample, and the concentration in the surface layer (6.3 ppm) was much higher than in the core sample (1.8 ppm). DEHP may have leached through the wax from some component of the outer packaging, such as ink on the printed label. But the wax proved a barrier to DEHA; none was detected in the cheese, despite the 5,539 ppm of that plasticizer found in the plastic outer wrap.
Five of the six types of plastic packaging we tested contained only low levels of the plasticizers we looked for, indicating that most forms of cheese packaging are not plasticized with those compounds. Cheeses in two of those package types, the individually wrapped slices and the laminated foil wrappers, had only minimal traces of plasticizers (all well below 1 ppm).
Cheeses in two other packaging types, the vacuum-packed clear films and the red plastic film covering one sample of Gouda, plus the two cheeses coated in wax with a plastic outer wrap, had traces of DBP (all below 1 ppm), but also contained DEHP concentrations up to 3.7 ppm. Packaging on these cheeses did contain low levels of these phthalates, from 4 to 38 ppm of DBP and from 7 to 73 ppm of DEHP. And the generally higher DEHP levels in surface, as opposed to in core, cheese samples suggest that migration from the packaging was a possible source of this contamination. But factors other than plasticizers in the plastic wrap may also be involved. Some inks and/or adhesives used on labels, or to seal packages, could be a source of phthalates. Our analysis of the packages included sampling label portions, both those printed on the plastic film and adhesive-backed paper labels. Some of the DEHP levels detected were associated with label portions. Background contamination of milk with these compounds (which can be concentrated by 10x or more in the cheese-making process) may also contribute to the total.
In contrast, the plasticized PVC film wrap on seven products had a high DEHA content (up to 2 percent by weight), and correspondingly high levels of DEHA were found in all samples of cheese wrapped in this plastic. In this case, there can be no doubt that the plastic wrap was the source of the DEHA in the cheeses.
The levels we detected (51 to 270 ppm for the cheese samples as a whole; up to 484 ppm in surface samples) are quite consistent with what has been previously reported by investigators in other countries. Sandberg et al. (1982) reported up to 1000 ppm of DEHA in surface layers of Swedish cheeses. The British Ministry of Agriculture, Fisheries and Food (MAFF) reported DEHA levels of 28 to 135 ppm in PVC-wrapped cheeses in the UK (Castle et al. 1987; Startin et al. 1987). An Australian study found DEHA content ranging from 31 to 429 ppm in cheeses wrapped in PVC film at the point of sale (Kozyrod and Ziaziaris 1989), while a Canadian survey found levels up to 310 ppm in similar products (Page and Lacroix 1995). A recent Danish study (Petersen et al. 1995) measured DEHA migration from PVC wrap into cheeses under lab conditions, and found that DEHA levels reached 44-58 ppm within the first two hours, and 53-150 ppm after 5 days.
Our calculated average DEHA levels (based on a weighted average of surface and core samples) may understate potential exposure. If consumers were to cut slices progressively from a block of cheese, re-wrapping it in the original plastic film between servings, many slices in turn would for a time be an outer surface, in contact with the plastic. Given the rapid migration rate found by Petersen et al., the average DEHA level for an entire package of cheese could be higher than what we found by averaging the outer and inner portions sampled at the same time. On the other hand, one can calculate the amount of DEHA in a given piece of PVC wrap (the plasticizer is typically used at levels of 15-25 mg/dm2). From the amounts we found in cheeses, we estimated that most of the DEHA in the wrap had apparently migrated into several of our cheese samples, suggesting that average levels much higher than those we found may be rather unlikely.
What are the health implications of DEHA in cheeses at the levels we found? Definitive human data are not available. In animal studies, DEHA has caused reproductive effects and birth defects, and it caused cancer (liver tumors) in mice, but not in rats (NTP 1996). Its possible endocrine effects have not been adequately evaluated, and need to be.
The Scientific Committee for Food (SFC) of the European Economic Community reviewed the literature on toxic effects of DEHA and suggested a Tolerable Daily Intake (TDI) of 0.3 mg/kg body weight (EEC Commission 1993, cited in Petersen et al. 1995). For a 60-kg adult, the TDI would be 18 mg; for a 20-kg child, 6 mg. Amounts of cheese containing 150 ppm DEHA that would provide the TDIs are 120 g/day for the adult, and 40 g/day for the child. These intakes are well within the normal range of cheese consump-tion, especially for children. It appears therefore that the levels of DEHA we found in cheeses wrapped in PVC film could readily cause many people who eat cheese to exceed the EEC SFCs recommended maximum safe daily intake. Petersen et al. calculated that the maximum DEHA concentration in cheese consistent with the SFCs TDI based on Danish cheese-consumption patterns is 18 ppm. The Commission of the European Community adopted a provisional "specific migration limit" (SML) of 18 ppm for DEHA (CEC, 1994). Our tests suggest that levels far above these health-based European limits are widespread in cheeses in the US.
In the UK, soon after the initial reports of high DEHA levels in PVC-wrapped cheeses, MAFF research showed that use of polymeric plasticizers, rather than DEHA alone, in PVC film greatly reduced migration of DEHA into cheese (Castle et al. 1988). A consensus developed that dietary DEHA intake from PVC cheese wraps should be substantially reduced (Harrison 1988). A survey by MAFF in 1990 found that the "old style" of PVC wrap was no longer in use in the UK; it had been replaced by film with polymeric plasticizers, or polymeric plasticizers mixed with DEHA, and leaching into cheeses was much lower (MAFF 1990). However, Petersen et al. reported, in 1995, that the old style PVC film was still in widespread use in Denmark, as it appears to be in the U.S. today, based on our limited tests. The same wrap may be used on other fatty foods prepared in supermarket "deli" departments, in addition to on cheeses.
An EEC directive that took effect in 1993 (Directive 90/128/EEC) limits the overall concentration of migrated plastic components in foods to 60 ppm, and imposes other restrictions to prevent transfer of plasticizers to foods (Petersen et al. 1995). As noted, the CEC has a provisional SML for DEHA specifically of 18 ppm (CEC 1994). There is no comparable regulation or limit on DEHA migration at present in the United States; instead, individual packaging migrants are evaluated on a case-by-case basis (Begley 1997). As part of its "Sensitivity of Method" policy, the FDA considers that packaging migrants at concentrations below 0.5 part per billion pose de minimis risks, while those that exceed 0.5 ppb are subject to the case-by-case assessment policy. The average DEHA level we found in PVC-wrapped cheeses exceeds this "threshold of regulatory concern" by a factor of 300,000-fold.
Summary: Our tests showed substantial migration of the plasticizer DEHA from PVC cling-wrap, used to wrap random-weight blocks of cheese cut in stores, into cheeses. The average level we found, 153 ppm, exceeds by a wide margin the provisonal maximum permitted level for DEHA migration into food in the EEC, which is 18 ppm. Many consumers of cheeses like the ones we tested could readily exceed the European SFCs TDI for DEHA by eating cheese servings well within the normal range. Such excessive intake is especially likely for children.
We agree with the European authorities who concluded a decade ago that this source of DEHA in the diet should be reduced or eliminated. There appear to be feasible alternative plasticizers for PVC film that would permit this public-health goal to be easily attained. There also are alternative kinds of plastic wrap that we found contain minimal amounts of plasticizers, some of which might be possible substitutes for PVC film for wrapping pieces of cheese in stores. We therefore call on the FDA and the food industry sectors involved to investigate this problem promptly, and to pursue and implement a strategy that will eliminate this avoidable source of exposure to DEHA as quickly as possible.
The lower levels of phthalates we found in cheeses also raise some concerns, and warrant efforts to track down and eliminate the sources of contamination involved. Although our tests indicate that the phthalates we detected are not used as plasticizers in cheese wraps, the possibility that they may be present in other packaging components, such as inks and adhesives, needs to be investigated.
Begley, T., 1997. Methods and approaches used by FDA to evaluate the safety of food packaging materials. Food Additives and Contaminants 14: 545-553.
Castle, L., Mercer, A.J., Startin, J.R. and Gilbert, J., 1987. Migration from plasticized films into foods. 2. Migration of di(2-ethylhexyl)adipate from PVC films used for retail food packaging. Food Additives and Contaminants 4: 399-406.
Castle, L., Mercer, A.J., and Gilbert, J., 1988. Migration from plasticized films into foods. 4. Use of polymeric plasticizers and lower levels of di(2-ethylhexyl)adipate plasticizer in PVC film to reduce migration into foods. Food Additives and Contaminants 5: 277-282.
Commission of the European Community (CEC), 1994. Synoptic Document N.7. Draft of provisional list of monomers and additives used in the manufacture of plastics and coatings intended to come into contact with foodstuffs. CS/PM/2356. Brussels: CEC, D-G III.
EEC Commission, 1993. Provisional compilation of the SCF opinions on materials and articles intended to come into contact with foodstuffs, Volume No. 2 (199101992). Brussels: The Commission of the European Communities, Directorate-General Industry (III/E/1).
Harrison, N., 1988. Migration of plasticizers from cling film. Food Additives and Contaminants 5: 493-499.
Kozyrod, R.P., and Ziaziaris, J., 1989. A survey of plasticizer migration into foods. Journal of Food Protection 52: 578-580.
Ministry of Agriculture, Fisheries and Food (MAFF), 1990. Plasticizers: Continuing Surveillance. Food Surveillance Paper No. 30. London: Her Majestys Stationery Office.
National Toxicology Program (NTP), 1996. NTP Technical Report No. 212: Di(2-ethylhexyl)adipate. Produced from Chemtrack Database 07/23/96. (See http://ntp-server.niehs.gov/htdocs/Levels/Tr212levels.html.)
Page and Lacroix, 1995, Food Additives and Contaminants 12: 129-151. (Cited by Castle, L., personal communication, 1997.)
Petersen, J.H., Naamansen, E.T., and Nielsen, P.A., 1995. PVC cling film in contact with cheese: Health aspects related to global migration and specific migration of DEHA. Food Additives and Contaminants 12: 245-253.
Sandberg, E., Vaz, R., Albanus, L., Mattsson, P. and Nilsson, K., 1982. Migration of plasticizers from PVC films to food. Var Foda 34: 470-482 (in Swedish).
Startin, J.R., Sharman, M., Rose, M.D., Parker, I., Mercer, A.J., Castle, L. and Gilbert, J., 1987. Migration from plasticized films into foods. 1. Migration of di(2-ethylhexyl)adipate from PVC films during home use and microwave cooking. Food Additives and Contaminants 4: 385-398.