ARCHIVED - Survey and Health Risk Assessment of Background Levels of Melamine in Infant Formula Allowed for Sale in Canada

Bureau of Chemical Safety
Food Directorate
Health Products and Food Branch
November, 2008
ISBN: 978-1-100-11355-5
Cat.: H164-104/2008E-PDF

Table Of Contents

• Background
• Sampling Plan and Analytical Methodology
• Melamine Levels in Infant Formula Products
• Health Risk Assessment
  • Hazard Characterization
  • Exposure Assessment
    • Consumption figures and body weights
    • Probable Daily Intake (PDI) Values
  • Risk Characterization
  • Conclusions
• Appendix 1

Background

Melamine (CAS No. 108-78-1) is the common name for 1,3,5-Triazine-2,4,6-triamine. Melamine is a synthetic chemical used in a variety of industrial applications including the production of resins and foams, laminates, glues, cleaning products, fertilizers and pesticides.

Melamine is not naturally occurring, has no accepted food uses, and is not permitted to be added to food in Canada. However, low levels of melamine may be present in the environment as the result of its many commercial uses, and trace amounts may occur in certain food commodities. The presence of low background levels (in the range of parts per billion or ppb) of melamine in food does not pose a health risk to consumers. High levels of melamine in any food product are thought to be due to intentional contamination in an attempt to artificially increase the protein content.

Recently, there have been numerous reports from several countries of the possible contamination of a variety of food products from China with melamine, specifically infant formula and other foods that contain milk and milk-derived ingredients.

In response to these reports, Health Canada conducted a survey to determine the background levels of melamine in infant formula products allowed for sale in Canada. It is important to note that infant formula products available for sale in Canada are subject to pre-market notification as required by Division 25 of the Food and Drug Regulations. These products have to satisfy stringent chemical, microbiological, nutritional, and quality requirements.

In this survey, samples of 80 different infant formula products were analysed for melamine, using a method with a detection limit of 4 ng/g of melamine. The results of this survey were used to estimate dietary intakes of melamine for infants of different ages.

Sampling Plan and Analytical Methodology

The survey covered 80 infant formula products that are marketed under 14 brands, by 10 different companies. The products were purchased between September 22, 2008 and October 6, 2008 from retail and grocery stores in the Ottawa area. Among the 80 products, 65 products were milk based, 13 products were soy based, 1 product was milk and soy based, and 1 product was made from casein derivatives and soy oil; 52 products were powdered and 28 were liquid of which 7 products were ready to use and 21 were concentrated.

Notes:

• Infant formula samples were tested as sold. Note that powdered and concentrated infant formulas are diluted with drinking water prior to consumption. Powdered infant formula is diluted with drinking water by up to an 8 fold factor. Concentrated liquid infant formula is usually diluted by a factor of 2 prior to consumption.
• It should be noted that the absence of any particular brand from this survey means only that the brand was not included in the survey. No particular inference should be drawn from the presence or absence of any brand.
• In most cases, samples of each product were taken from only one production lot. Therefore, variations from one production lot of a product to another lot were not addressed.
• When there was more than one production lot tested for a product, the results from that product are differentiated in the table by an arbitrary lot letter (e.g. Lot A, Lot B, etc). When multiple samples from the same lot were tested, the results were averaged and reported as a single sample (an asterisk * is added to denote averaged results).
• Samples represent a "snapshot" of the market and do not represent market share. Differences between brands do not necessarily reflect differences in consumer exposure to background levels of melamine and should not be used as indices of good product choice for consumers.

Melamine Levels in Infant Formula Products

Appendix 1 summarizes the concentration of melamine determined in samples of infant formula products. Melamine was detected in 60 of the 80 products sampled. Detectable concentrations of melamine ranged from 4.31 to 346 ng/g, well below Health Canada's interim maximum level of 1000 ng/g of melamine in infant formula.

Health Risk Assessment for Background Levels of Melamine in Infant Formula Found on the Canadian Market

All infant formula products must undergo a rigorous pre-market evaluation and be approved by Health Canada before they are allowed for sale in Canada. No infant formula produced in China is approved for sale in Canada. To date, no melamine-contaminated infant formula has been discovered in Canada and no melamine related illnesses associated with any food has been reported in Canada.

For the purposes of differentiating between the presence of low background levels of melamine in food and the problem of intentional adulteration of infant formula, the Bureau of Chemical Safety's Food Research Division of Health Canada conducted this survey to determine the background levels of melamine in infant formula products allowed for sale and available on the Canadian retail market. This survey has allowed Health Canada scientists to quantify the background occurrence of melamine in infant formula, assess the potential risks associated with exposure to melamine through the consumption of infant formula, and make any necessary changes to current risk management strategies.

Hazard Characterization

The toxicological reference dose established by Health Canada scientists for melamine of 0.35 mg/kg bw/day, and applied for the development of the interim standards for both melamine and cyanuric acid in foods, was used in this assessment to evaluate Canadian infant's exposure to melamine through the consumption of infant formula.

A complete description of the hazard characterization is provided in the document: Health Canada's Human Health Risk Assessment in Support of Standard Development for Melamine in Foods.

Exposure Assessment

Consumption figures and body weights

Infant formula intake by infants for specific growth periods were obtained from "A Practical Guide to Baby Care" prepared by Institute National de Santé Publique du Quebec (2001). Frequency of consumption was assumed to be daily. The average infant weights utilized in this assessment were based on the growth charts from the " Pediatricians Guide to your Children's Health and Safety".

Analytical results

This survey analysed 94 samples from 80 different infant formula products available on the Canadian retail market. Of the 80 different products analysed 20 were below the analytical detection limit of 4 ng/g (ppb; parts per billion), representing approximately a 75% incidence of positive detections of low background levels of melamine. Detectable melamine concentrations ranged from 4.31 to 346 ng/g (ppb) in infant formula products "as purchased", and from 0.53 to 68.86 ng/g "as consumed". None of these products contained melamine at levels above the interim standard of 1 ppm (1 ppm is equal to 1000 ppb) of melamine and cyanuric acid applied to all infant formula and other sole source nutrition products in the form in which they are sold.

Reconstitution factors were applied to concentrations of melamine found in concentrated and powdered infant formula (that is, those products requiring dilution or reconstitution prior to consumption). The reconstitution factors, based on typical product label instructions, were applied in order to obtain melamine concentrations in these products "as consumed". Reconstitution factors were 1, 0.5 and 0.124 respectively for ready-to-use, concentrated and powdered formulas.

Mean and maximum concentrations of melamine "as consumed" were then calculated for infant formula products for each age group taking into consideration the age for which each infant formula product is directed (ie. those products directed at infants of particular age ranges). Table 1 shows the average and maximum concentrations of melamine in infant formulas consumed by infants of different age groups, utilized in the exposure assessment. Non-detects were assumed to contain melamine at a concentration equal to half the detection limit of 4 ng/g. The highest levels of melamine on an "as consumed" basis were found in a ready-to-use infant formula at 68.86 ng/g.

Probable Daily Intake (PDI) Values

A deterministic exposure assessment was used to calculate Probable Daily Intakes (PDI) of melamine for different age groups of infants based on the survey results. In particular, the average and maximum infant formula consumption intakes and average and maximum concentrations of melamine in infant formulas "as consumed" were utilised to estimate melamine intakes.

Calculations are based on the following exposure scenarios for each age group:

A summary of the PDIs of melamine for infants of different age groups is shown in Table 2.
The PDIs of melamine varied from as low as 0.542 µg/kg body weight per day (0.000542 mg/kg bw/day) for 12-18 month infants to as high as 19.07 µg/kg body weight per day for (0.01907 mg/kg bw/day) 0-1 month old infants assuming a maximum infant formula intake at the maximum melamine concentration "as consumed" found in the survey.

Risk Characterization

As an interim risk management measure, Health Canada has set an interim standard for combined residues of melamine and cyanuric acid in infant formula and sole source nutrition products, including meal replacements of 1.0 ppm. All infant formulas "as sold" and adjusted to an "as consumed" basis, sampled in the survey, were well below this interim standard.

Infant formula samples (ready-to-use, concentrated, and powdered) were collected from the Canadian market (n=94) and analyzed using Liquid/Liquid Extraction and Cation Exchange Solid Phase Extraction to prepare samples, and Electrospray Positive Ionization Liquid Chromatography Tandem Mass Spectrometry to quantitate melamine (detection limit of 4 ng/g). Average and maximum melamine levels detected in the infant formula "as consumed" were applied to the average and maximum infant formula intake rates by children 0-18 months of age.

At the maximum concentration of 68.86 ng/g and at maximum infant formula intakes, less than 10% of the current Health Canada's Toxicological Reference Dose (TRD) would be accounted for in all infant age groups consuming infant formula. As such, it is concluded that these background levels of melamine in infant formula "as consumed" do not represent a health risk to consumers.

Conclusions

The Bureau of Chemical Safety in Health Canada's Food Directorate conducted a human health risk assessment based on the observed background levels of melamine in infant formula. All of the infant formulas analysed contained either very low levels of melamine or did not contain melamine at detectable levels (using a very sensitive methodology detecting as low as 4 ng/g (4 ppb)). Based on the low levels of melamine found in infant formula, Health Canada scientists have concluded that infant formula products allowed for sale in Canada are safe to consume.

It remains Health Canada's policy that levels of potential contaminants in infant foods should be kept as low as reasonably achievable. While it is not unexpected to detect such low levels of melamine in foods, as those reported in this survey, Health Canada will apply a zero tolerance to any food product where it is determined there had been deliberate adulteration.

Appendix 1

Concentration (ng/g sample) of melamine in infant formula as sold

• It should be noted that the absence of any particular brand from this survey means only that the brand was not included in the survey. No particular inference should be drawn from the presence or absence of any brand.
• Samples represent a "snapshot" of the market and do not represent market share. Differences between brands do not necessarily reflect differences in consumer exposure to melamine.
• When there was more than one production lot tested for a product, the results from that product are differentiated in the table by an arbitrary lot letter (e.g. Lot A, Lot B, etc). When multiple samples from the same lot were tested, the results were averaged and reported as a single sample (an asterisk * is added to denote averaged results).
• "nd" means that melamine was not detected in that sample based on a detection limit of 4 ng/g.