Number 83 March 1996
Index to MAFF UK Food Surveillance Information
Sheets, 1996
see also:
60: MAFF UK - Phthalates in Paper
and Board Packaging (May 1995)
82: MAFF UK - Phthalates in Food (March
1996)
93: MAFF,UK - Multi element analysis
of infant foods - follow-up survey (August 1996)
167: MAFF UK - Plant Oestrogens in
Soya-Based Infant Formulae (November 1998)
168: MAFF UK - Phthalates in
Infant Formulae - Follow-Up Survey (December 1998)
190: MAFF UK - Metals and other
elements in infant foods (November 1999)
Summary
MAFF's Food Safety Directorate has carried out a survey of phthalates in
samples of retail infant formulae. Results indicate that dietary intakes
of individual phthalates by infants consuming these formulae are below
Tolerable Daily Intakes, where these have been set. The Department of
Health has advised that these initial surveillance data and the available
biological studies do not suggest that the concentrations of phthalates
found in infant formulae would have any effect. This advice takes account
of all available information on the possible effects of phthalates,
including recent studies concerning oestrogenic activity.
Phthalates were detected in all samples of infant formula that were
analysed. As in earlier unpublished work conducted at CSL Food Science
Laboratory, Norwich, the most abundant phthalate identified in all samples
was di-(2-ethylhexyl) phthalate (DEHP). Using the manufacturers' feeding
guides, average intakes of total phthalates (expressed as dimethyl
phthalate) from infant formulae would be 0.13 mg/kg bodyweight/day for a
new-born infant, falling to 0.10 mg/kg bodyweight/day at six months.
Further work is required to establish the sources of phthalates present in
infant formulae and the reasons for any systematic variation between
different brands, and, as a matter of prudence, to reduce phthalate levels
in infant formulae where possible.
Background
Phthalic acid diesters, commonly known as phthalates, are a group of
organic chemicals that have a variety of industrial uses, including use as
plasticisers in a wide range of household and consumer goods,1
uses in lubricating oils and uses as carriers for perfumes in cosmetics.
The use of phthalates in plastic food packaging is limited, for example,
to the manufacture of materials such as some adhesives and some printing
inks. Phthalates are no longer used in the manufacture of cling film and
most other food contact plastic materials.
The release of phthalates to the environment may occur during the
production and distribution of phthalates, during the manufacture or use
of products in which they are used, or after the dispersal of these
products. Although some phthalates occur naturally in coal, crude oil and
shale, the contribution of such sources to general environmental levels of
phthalates is likely to be insignificant.2
As a result of their extensive use and their moderate resistance to
degradation, phthalates are widely distributed in the environment and are
often found at low levels in food. The toxicity of these substances has
been considered by the EC Scientific Committee for Food (SCF), which has
established Tolerable Daily Intakes (TDIs) for several of the chemicals in
the group.
A recent survey by MAFF showed that two phthalates, dibutyl phthalate
(DBP) and DEHP, were often present in paper and board packaging and were
also present at generally low concentrations (typically less than 10
mg/kg) in foods packaged in paper and board.3
The concentrations of phthalates in samples of fat analysed in this recent
survey would be expected to be lower in the core than at the surface if
migration from packaging was the major source of contamination. In over
half the samples, the concentrations of the phthalates at the core were
about equal to or higher than the concentrations found at the surface.
These results suggested that phthalates in food originate at least in part
from general environmental contamination. Results from other surveys of
milk and milk products,4,5
pork fat 6 and fish7,8
also indicate that phthalates are present in food from general
environmental contamination rather than specific sources such as food
packaging.
The current survey was conducted to estimate the intakes of individual
phthalates and total phthalates by infants from infant formulae.
Information on the market share of different brands of infant formulae was
taken from a MAFF study of food and nutritional intakes of infants
conducted in 1986; 9 additional
information on current brand availability was provided by the Infant and
Dietetic Foods Association.
A total of 59 individual samples of 15 different brands of infant
formulae were purchased from retail outlets in five towns across the UK.
In most, but not all, cases the different samples of each brand
represented different manufacturing batches, as indicated by batch codes
on the retail packaging. For each of the nine market-leading brands of
casein-dominant, whey-dominant or soya-based products, samples were
composited together such that each sample for analysis contained samples
of a single brand. Three further composite samples were made up of (i) two
other brands of casein-dominant formulae; (ii) two other brands of
whey-dominant formulae; and (iii) two other brands of soya-based formulae.
All samples were analysed at CSL Food Science Laboratory, Norwich.
Identification and determination of individual phthalates
Infant formulae samples were extracted with acetone/hexane (1:1) and
were fractionated by size exclusion chromatography then concentrated prior
to analysis by gas chromatography/low resolution mass spectrometry (GC-MS)
to identify individual phthalates. The limit of detection of the GC-MS
instrument was typically 0.001 mg/kg for individual phthalates. The
analytical method has only recently been developed and insufficient
validation data are available for a robust quantitative measure of
repeatability to be given. Further analyses were conducted using gas
chromatography/high resolution mass spectrometry to confirm the identities
of both major and minor peaks. These showed that some small peaks with
relative retention times coincident with those of phthalate standards
were, in fact, interfering compounds that had been co-extracted.
Concentrations of dipropyl phthalate (DPP), DBP, di-iso-butyl
phthalate (DIBP), butylbenzyl phthalate (BBP) and DEHP were quantified
after appropriate calibration. The presence of five other phthalates (di-iso-propyl
phthalate, dihexyl phthalate, dicyclohexyl phthalate, diheptyl phthalate
and dinonyl phthalate) could be confirmed by gas chromatography/high
resolution mass spectrometry, but these compounds were not quantified.
Phthalates of higher molecular weight (didecyl, diundecyl and didodecyl
isomers) were not detected in any of the samples analysed.
Some commercially-important phthalate esters, such as di-iso-octyl
phthalate (DIOP) are complex mixtures of isomers that are dispersed over a
wide range of retention times in capillary column gas chromatography. Nine
of the twelve samples were therefore analysed by high performance liquid
chromatography/mass spectrometry (HPLC-MS) using an atmospheric pressure
chemical ionisation interface, under which conditions the three mixtures
of isomers represented by DIOP, di-iso-nonyl phthalate and di-iso-decyl
phthalate are each eluted as single peaks, thus allowing total dioctyl,
and dinonyl phthalate concentrations to be established. The limit of
detection of the HPLC-MS instrument was of the order of 0.01 mg/kg for
each of the mixtures of phthalate isomers
Determination of total phthalates
Infant formulae samples were extracted with acetone/hexane (1:1) and
phthalates were converted to dimethyl phthalate (DMP) by heating the
extracts with methanolic potassium hydroxide solution and then with boron
trifluoride/diethyl ether complex. These extracts were then fractionated
by size exclusion chromatography and analysed by gas chromatography-mass
spectrometry (GC-MS).
Preparation and analysis of the samples were completed in parallel, in a
batch which included five process blank determinations. The lower limit at
which DMP extracted from samples could be measured securely was set by the
presence of DMP in blanks rather than by instrumental sensitivity. The
limit of quantification of 0.10 mg/kg was the mean plus three times the
standard deviation of the process blanks.
Results for each of the twelve samples are the means of replicate
determinations:
- Eight determination were performed in duplicate; the overall mean
difference between duplicate analyses was 16 percent.
- For two further samples, differences between duplicate analyses were
excessive and a further pair of analyses was performed. In both cases
the larger of the original results was clearly anomalous and was
discarded. The result quoted is therefore the mean of the three other
determinations.
- For two samples, analyses were performed in quadruplicate to provide
an indication of repeatability. Relative standard deviations of the
quadruplicate analyses were 6.3 percent and 9.6 percent.
Results
Total phthalates and individual phthalates were determined in all twelve
composite samples. The concentrations of total phthalates, measured and
expressed as DMP, and of individual phthalates in each of the samples are
shown in Table 1. The concentrations of each of
the analytes were similar in casein-dominant, whey-dominant and soya-based
formulae.
Dietary intakes of individual phthalates have been calculated for
infants, based on the manufacturers' feeding guides and assuming that
infant formulae were the only source of nutrition. When expressed on a
bodyweight basis, consumption of infant formulae, and thence intakes of
phthalates, are higher in new-born infants than in older children. For
each compound, there will be a range of daily intakes between birth and
six months due to changes in feeding regime and body weight over this
period in the infant's life. These ranges are given in
Table 2.
The dietary intakes of total phthalates was estimated on the same basis
as for individual phthalates. Average intakes of total phthalates from
infant formulae would be 0.13 (range: 0.03 to 0.23) mg/kg bodyweight/day
in new-born infants, falling to 0.10 (range: 0.02 to 0.19) mg/kg
bodyweight/day at six months. A variable proportion (5 percent - 77
percent) of the measured total was attributable to the measured individual
compounds.
Interpretation
The toxicity of those phthalates permitted for use in food contact
materials has been considered by the SCF. Sufficient toxicological data
were available for the committee to set a TDI of 0.05 mg/kg bodyweight/day
for DEHP and a temporary TDI for four of the other phthalates included in
this survey: 0.1 mg/kg bodyweight/day for butylbenzyl phthalate; 0.05
mg/kg bodyweight for DBP; 0.1 mg/kg bodyweight/day for dicyclohexyl
phthalate; and 0.2 mg/kg bodyweight/day for diethyl phthalate. Less
toxicological data are available for the remaining phthalates used in food
contact materials. The SCF has recommended a "group restriction"
for the sum of these remaining compounds of 0.05 mg/kg bodyweight/day.
Unlike the TDI, the group restriction is not a precise measure of
potential risk to human health. Instead, it is a precautionary limit set
to guide the food contact materials industry and enforcement authorities
pending submission of further toxicological data.
The TDIs for any of the individual phthalates would not be exceeded at
any time by infants fed according to the manufacturers' instructions on
any of the infant formulae analysed in this study. Dietary intakes of the
sum of the other phthalates that were quantified in this study (DPP, DIBP
and dioctyl phthalates other than DEHP) would also be lower than the
recommended group restriction of 0.05 mg/kg bodyweight/day.
The measurements of total phthalates found in infant formulae give
average intake estimates falling from 0.13 to 0.10 mg/kg bodyweight/day
over the first 6 months of life. Making a "worst case"
assumption that all phthalates might have TDIs as low as DEHP or DBP,
which is unlikely, total phthalate intakes exceed the lowest TDI of 0.05
mg/kg bodyweight/day by two- to three-fold. However, the TDI incorporates
a 100-fold safety factor beyond the dose with no effect. The Department of
Health has advised that there are unlikely to be any risks to infant
health from total phthalate levels even using the above "worst case"
assumptions about the activity of phthalates. The Department of Health has
also commented that the lowest TDIs are based on the no-effect level for
increases in liver peroxisomes, an effect seen in rodents exposed to
certain phthalates, but man is either considerably less sensitive or may
be insensitive to this effect. As this effect has been taken into account
by the SCF when setting the TDIs and the group restriction for phthalates,
it represents a very cautious approach to safety.
The TDIs discussed above were established before the recent findings
that some phthalates may have oestrogenic activity. Two phthalates, BBP
and DBP, appear to have weak oestrogenic activity in vitro.10
That is, in "test tube" experiments they triggered some of the
same effects as the natural hormone, oestradiol. However, they were about
a million times less potent than oestradiol. The plasticisers,
di-(2-ethylhexyl) adipate and DEHP, were not found to be active. Activity
in vitro is not necessarily evidence that they will be active in
the human body. To date only one of these phthalates, BBP, has been tested
in laboratory rodents.11 Female rats
were given BBP at a fixed concentration of 1 mg/l in their drinking water
throughout pregnancy and lactation. Resulting intakes were estimated to
range from 0.1 to 0.4 mg/kg bodyweight/day during the treatment period.
This was reported to cause small reductions in testis weight and sperm
production in the male offspring, who would have been exposed to BBP via
the mother during the entirety of their most vulnerable period of testis
development. These studies suggest that BBP and DBP may have weak
oestrogenic activity but at this stage the data are very tentative and no
firm conclusions can be drawn. Studies to confirm these effects and to
establish a no-effect level would be needed.
Taking as a worst case the infant formula with the highest
concentrations of the two phthalates which may have oestrogenic activity
(BBP and DBP), the combined estimated intake of these two phthalates was
0.023 mg/kg bodyweight/day. This is between four- and seventeen-fold lower
than the dose range said to cause minimal effects in rodents. All other
formulae analysed had lower concentrations of these phthalates, ranging
down to 33 to 133 times below the dose range said to cause minimal effects
in rodents.
The Department of Health has commented that there are differences in the
timing of testis development between rats and humans that may be important
for risk assessment. The relevant aspects of testis development in humans
takes place over a much greater proportion of prenatal and postnatal life
compared with testis development in the rat. Thus it is not possible to
directly extrapolate from effects observed in rats to human new-borns and
infants. The Department has advised that these initial surveillance data
and the limited biological studies available so far do not suggest that
the concentrations of phthalates found in infant formulae would have any
effect in humans. Further work is required to establish the sources of
phthalates present in infant formulae and the reasons for any systematic
variation between brands and, as a matter of prudence, to reduce phthalate
levels in infant formulae where possible.
References
- International Programme on Chemical Safety (1992)
Diethylhexyl phthalate. Environmental Health Criteria 131,
publ. World Health Organisation, Geneva.
- International Programme on Chemical Safety (1992)
Diethylhexyl phthalate. Environmental Health Criteria 131,
publ. World Health Organisation, Geneva.
- Ministry of Agriculture, Fisheries and Food (1995)
Phthalates in paper and board packaging. Food Surveillance
Information Sheet 60.
- Sharman, M., Read, W.A., Castle, L. and Gilbert, J.
(1994) Levels of di-(2-ethylhexyl)phthalate and total phthalate esters
in milk, cream, butter and cheese. Food Addit. Contam. 11,
375-385.
- Castle, L., Gilbert, J. and Eklund, T. (1990)
Migration of plasticiser from poly(vinyl chloride) milk tubing. Food
Addit. Contam. 7, 591-596.
- Cerbulis, J. and Byler, D.M. (1986) Isolation and
detection of dialkyl phthalates from pork. J. Agric. Food Chem.
34, 198-200.
- Musial, C.J., Uthe, J.F., Sirota, G.R., Burns, B.G.,
Gilgan, M.W., Zitko, V. and Matheson, R.A. (1981) Di-n-hexyl phthalate
(DHP), a newly identified contaminant in Atlantic herring (Clupea
harengus harengus) and Atlantic mackerel (Scomber scombrus).
Can. J. Fish. Aquat. Sci. 38, 856-859.
- Mayer, F.L., Jr. Stalling, D.L. and Johnson, J.L.
(1972) Phthalate esters as environmental contaminants. Nature
(London) 238, 411-413.
- Mills, A. and Tyler, H. (1992) Food and nutrient
intakes of British infants aged 6-12 months, publ. HMSO, London.
- Jobling, S., Reynolds, T., White, R., Parker, M.G.
and Sumpter, J.P. (1995) A variety of environmentally persistent
chemicals, including some phthalate plasticisers, are weakly
oestrogenic. Environmental Health Perspectives 103,
582-587.
- Sharpe, R.M., Fisher, J.S., Millar, M.M., Jobling,
S. and Sumpter, J.P. (1995) Gestational and lactational exposure of
rats to xenoestrogens results in reduced testis size and sperm
production. Environmental Health Perspectives 103,
1136-1143.
Contact point
For further information, please contact:
Dr Alison Gleadle
MAFF, Joint Food Safety and Standards Group
Food Contaminants Division
Room 238, Ergon House, c/o Nobel House
17 Smith Square
London SW1P 3JR
Tel: +44 (0)20 7238 6227
Fax: +44 (0)20 7238 5331
Return to Index to Surveillance Information Sheets, 1996 page
Go to top of page
These pages were last updated on 1 October 1996
