Number 64 June 1995
MAFF UK - SURVEY OF AFLATOXIN M1 IN RETAIL MILK AND MILK
PRODUCTS
Index to MAFF UK Food Surveillance Information
Sheets, 1995
See also:
22: MAFF, UK - Aflatoxins in herbs and
spices (January 1994)
37: MAFF, UK -
Survey of "ethnic foods" for mycotoxins (September 1994)
45: MAFF, UK - Aflatoxin contamination of
peanut butter and peanuts (November 1994)
56: MAFF, UK - Aflatoxin contamination of
peanut butter and peanuts (February 1995)
78:
MAFF, UK - Survey of aflatoxin M1 in farm gate milk (January 1996)
81: MAFF, UK - Aflatoxin surveillance of retail
and imported nuts, nut products, dried figs and fig products (March 1996)
Summary
MAFF has carried out a survey of retail milks and milk products to assess
the levels of aflatoxin M1 (AFM1) in these commodities.
Most milks contained very low or undetectable amounts of AFM1. The
amounts detected in milk products such as yoghurt and cheese were consistent
with the low levels of AFM1 found in milks. These findings provide
similarly reassuring results as earlier surveys.
Background
MAFF has been monitoring aflatoxin contamination of food for a number of
years. Aflatoxins are a group of toxicants produced predominantly by two
species of mould, Aspergillus flavus and A. parasiticus, which
may grow on a number of crops. The formation of aflatoxins depends on the foods
on which the moulds grow and the conditions of heat and humidity during crop
growth and storage. Aflatoxins have been found in a wide variety of foodstuffs
around the world and particularly in countries where climate or poor storage
conditions favour the growth of moulds. The aflatoxins are a group of
structurally related difuranocoumarin derivatives of which 18 types have been
identified. Aflatoxins B1, B2, G1, G2
and their metabolites M1 and M2 are the most common, and
of these B1 and G1 occur most frequently in food. AFM1
is found in the milk and dairy products of animals that have consumed feed
contaminated by aflatoxin B1 (AFB1).
Research on laboratory animals has shown that aflatoxins are liver
carcinogens in animals and there is evidence to suggest that they are also human
carcinogens, with AFB1 being the most potent and AFM1
being probably about 10 times less potent. The Food Advisory Committee (FAC)
recommended statutory limits following advice from the Committee on Toxicity of
Chemicals in Food, Consumer Products and the Environment (COT) that aflatoxin
contamination should be reduced to the lowest level that is technologically
possible.
Regulations to limit the levels of aflatoxins in certain foodstuffs were
introduced in the UK at the end of 1992. The 'Aflatoxins in Nuts, Nut
Products, Dried Figs and Dried Fig Products Regulations 1992' (S.I. 1992 / 3236)
sets limits for aflatoxins in these foods. The 'Feeding stuffs Regulations
1991 (S.I. 1991/2840) set maximum levels for AFB1 in animal feed
and thus restrict the amount of AFM1 carried over into milk.
Revised limits come into force on June 30th 1995 (S.I. 1995/1412). The UK is
currently involved in EC discussions over possible harmonised legislation to
control certain mycotoxins in some foods, including AFM1 in milk and
dairy produce. These discussions are at a very early stage.
Current Survey
This survey of milk and milk products obtained from retail outlets
throughout England and Wales determined the level of AFM1 present in
88 full fat, 3 semi-skimmed and 71 skimmed milks, 73 UK produced cheeses, 62
infant formulae, 31 dried milks and 30 yoghurts. Milk samples were obtained
during the summer, when cows are at pasture, and in the winter when cows consume
more concentrated feedingstuffs.
Methods of analysis
Analysis for AFM1 was performed at the CSL Food Science
Laboratory, Norwich, using immunoaffinity column clean-up and reversed phase
high performance liquid chromatography with fluorescence detection. The limit
of determination was 0.01 µg/kg for milks and 0.02 µg/kg for milk
products.
Analyses were conducted using spiked samples (equivalent to 0.1µg/kg)
to assess recoveries, those falling between 70 - 110 per cent being classed as
valid. The average recovery for the analysis of skimmed and semi skimmed milks
was 92 per cent, with a range of 81 - 98 per cent and a coefficient of variation
(CV) of 5.5 per cent. The average recovery for the analysis of full fat milk
was 92 per cent, with a range of 70 - 100 per cent and a CV of 7.0 per cent.
The average recovery values for milk products ranged between 74 - 92 per cent.
The largest spread of data was observed for Wensleydale cheese which had an
average recovery of 86 per cent with a range of 68-104 per cent and a CV of 21.6
per cent, which can probably be explained by sample inhomogeneity. All results
were corrected using recovery values obtained daily.
Repeatability figures were obtained for a number of unspiked samples. The
CV of one milk sample, analysed 4 times and containing a mean value of 0.22 µg
AFM1/kg was 9.5 per cent. At the 0.01 µg/kg level the degree
of uncertainty would be greater, with repeatability possibly as large as 50 per
cent. The significance of values quoted as 0.01 µg/kg and <0.01 µg/kg
thus should be treated with caution. All cheese samples were analysed in
duplicate and the average CVs for all positive cheese samples was 8.0 per cent.
Results
Table 1 shows the results for full fat,
semi-skimmed and skimmed milks. Fifty-five percent (47/86) of summer milks
contained less than 0.01 µg AFM1/kg . Forty-four percent
(38/86) of summer milk samples contained between 0.01 - 0.04 µg AFM1/kg
. One further sample of full fat summer milk was found to contain the highest
level of 0.22 µg AFM1/kg. When a second sample of this milk
was analysed (originating from the same producer and sold at the same retail
point) no AFM1 was detected (less than 0.01 µg/kg).
Thirty-four percent (26/76) of the winter retail milks contained less than
0.01 µg AFM1/kg. Sixty-six percent (50/76) of winter milks
contained between 0.01 - 0.04 µg AFM1/kg.
Table 2 shows the results for dried milk, infant
formulae and yoghurts. AFM1 was not detected above 0.02 µg/kg
in 96 per cent (89/93) of dried milk and infant formulae. The highest level
(0.05 µg/kg) was found in an infant formula sample . AFM1 was
not detected above 0.02 µg/kg in 80 per cent (24/30)of yoghurt samples.
The highest level of 0.04 µg AFM1/kg was found in a plain
yoghurt sample.
Table 3 shows the results for a variety of UK
cheeses. Low levels of AFM1 were detected in all samples. The
highest level detected was 0.22 µg/kg, in a sample of Wensleydale.
Forty-seven per cent (34/73) of cheeses contained AFM1 in the range
of 0.06 - 0.1 µg/kg.
Interpretation
Summer retail milk samples contained very low aflatoxin levels in milk, with
the majority (99 per cent) of samples containing <0.05 µg AFM1/kg.
Only one sample exceeded this level, but when a second sample of this milk was
bought and analysed (originating from the same producer and sold at the same
retail point) no AFM1 was detected (<0.01 µg/kg). All
winter milks analysed in this survey contained <0.05 µg AFM1/kg.
The probable higher use of concentrated feedingstuffs in the winter months does
not appear to lead to any significant increase in the incidence of AFM1
in milk.
A comparison with previous AFM1 surveillance of farm gate milk
in 1988 and 19891 reveals a reduction in the
proportion of samples exceeding 0.05 µg/kg from 3.3 per cent to 0.6 per
cent. An apparent increase in the proportion of samples containing a very low
level of AFM1 in the range of 0.01 - 0.04 µg AFM1/kg
must be treated with caution because of the degree of uncertainty in measuring
AFM1 at the 0.01 µg/kg. level in milk. The observed
differences are probably due to small fluctuations above and below the reporting
limit.
None of the samples of dried milks, infant formulae or yoghurt contained
levels of AFM1 exceeding 0.05 mg/kg. This indicates that the milk
used to manufacture these products contained negligible levels of AFM1.
Cheeses contained the highest levels of AFM1, probably because AFM1
becomes concentrated in the curd when cheese is made. However, the levels of
AFM1 in cheese were consistent with the levels of AFM1
in milks found in this survey.
These results demonstrate the effectiveness of the Feeding Stuffs
Regulations in keeping AFM1 levels low in milk by regulating the
levels of AFB1 in cattle feed. The significance of these data will
be considered by the Working Party on Naturally Occurring Toxicants in Food when
it next meets in Autumn 1995. A further survey of farm gate milk currently
underway will be reported in the Bulletin this Autumn and these results will
also be considered by the Working Party.
- Ministry of Agriculture, Fisheries and Food (1993). Mycotoxins: Third
Report. Food Surveillance Paper No. 36. HMSO.
- Mortimer, D. N., Gilbert, J. and Shepherd, M. J. (1987). Rapid and highly
sensitive analysis of aflatoxin M1 in liquid and powdered milks using an
affinity column cleanup. Journal of Chromatography, 407,
393-398.
- Sharman, M. and Gilbert, J. (1991). Automated aflatoxin analysis of foods
and animal feeds using immunoaffinity column cleanup and high-performance liquid
chromatographic determination. Journal of Chromatography, 543,
220-225.
- Sharman, M., Patey, A. L., and Gilbert, J. (1989). Application of an
immunoaffinity column sample cleanup to the determination of aflatoxin M1 in
cheese. Journal of Chromatography, 474, 457-461.
Contact point
Mrs Chelvi Leonard
MAFF, Food Safety and Science Group
Food
Contaminants Division
Room 210 Ergon House
c/o Nobel House
17
Smith Square
London SW1P 3JR
Tel:- +44 (0)171 238 5734
Fax:-
+44 (0)171 238 6591
Return to Index to Surveillance Information Sheets, 1995 page
Go to top of page
These pages were last updated on 1 October 1996
