The pet food recall: Perspectives from the trenches (Proceedings)

What was the sequence of events leading to the recall?

Late February – consumer complaints about cat pet food

February 28 – CAHFS received what turned out to be an index case from veterinary clinic in So. CA

March 6 – feed study conducted by Menu Foods

March 7 – testing of food by Menu Foods; initial testing negative

March 12 – testing of wheat gluten requested by Menu Foods

March 16 – Menu Pet Foods issues a recall

March 17 – FDA Kansas Laboratory starts testing for broad array of chemicals

March 22 – FDA Forensic Chemistry Center in Cincinnati begins testing; melamine detected. CAHFS also finds melamine

March 23 – NYS Agricultural Laboratory reports finding aminopterin; not detected by FDA and CAHFS

March 26 – other compounds related to melamine were also detected: cyanuric acid, ammelide, ammeline and ureidomelamine (MARCs or melamine and related compounds)

April 2 – "wheat gluten" determined to be wheat flour

April 9 – CAHFS confirms another case involving a pet food that was not on the recall list; recall list expands

April 16 – a second commodity, rice protein concentrate, was identified as being contaminated and additional pet food was recalled

Late April – investigation of exposure of hogs, poultry and fish conducted

What are MARCs (melamine and related compounds)?

Melamine and cyanuric acid are relatively simple compounds:

Melamine and Related Compounds

Melamine is used combined with formaldehyde to produce melamine resin, a very durable thermosetting plastic, and of melamine foam, a polymeric cleaning product. The end products include countertops, fabrics, glues, housewares and flame retardants. Melamine is one of the major components in Pigment Yellow 150, a colorant in inks and plastics. Melamine is also used to make fertilizers. Melamine derivatives of arsenical drugs are potentially important in the treatment of African trypanosomiasis. Melamine use as non-protein nitrogen (NPN) for cattle was described in a 1958 patent. In 1978, however, a study concluded that melamine was not an acceptable nonprotein N source for ruminants because its hydrolysis in cattle was slower and less complete than other nitrogen sources such as cottonseed meal and urea.

Melamine is reported to have an oral LD50 of > 3000 mg/kg based on rat data, which makes it only minimally toxic (table salt has a similar LD50 value). It is also an irritant when inhaled or in contact with the skin or eyes. The reported dermal LD50 is >1000 mg/kg for rabbits. In a 1945 study, large doses of melamine were given orally to rats, rabbits and dogs with no significant toxic effects observed.

Cyanuric acid is used as a stabilizer in recreational water treatment to minimize the decomposition of hypochlorous acid by sunlight in outdoor swimming pools and hot tubs. It achieves this by absorbing UV radiation. Chlorinated derivatives of cyanuric acid, such as trichloro-s-triazinetrione and sodium dichloro-s-triazinetrione, are used as algacides or microbiocides in swimming pool water and large scale water systems in industry. FDA permits a certain amount of cyanuric acid to be present in some non-protein nitrogen (NPN) additives used in animal feed and drinking water. Cyanuric acid has been used as NPN. For example, Archer Daniels Midland manufactures an NPN supplement for cattle, which contains biuret, triuret, cyanuric acid and urea.

There is very little toxicity information available for cyanuric acid.

A clue to the potential formation of crystals of melamine cyanurate in renal tubules: testing for cyanuric acid concentration is commonly done with a turbidometric test, which uses a reagent, melamine, to precipitate the cyanuric acid. The relative turbidity of the reacted sample quantifies the CYA concentration. This test works because melamine combines with the cyanuric acid in the water to form a fine, insoluble, white precipitate that causes the water to cloud in proportion to the amount of cyanuric acid in it.

Where did MARCs come from?

It is believed that melamine had been added to various commodities (in the case of the pet food, both "wheat gluten" and rice protein concentrate) to increase their apparent protein content. The compounds other than melamine are believed to have been inadvertent contaminants as a result of using melamine "scrap".

What was the pathogenesis of ARF in affected animals?

Assessment of Melamine and Cyanuric Acid Toxicity in Cats

Birgit Puschner, Robert Poppenga, Patricia Pesavento, Elizabeth Tor, Linda Lowenstein and Michael Filigenzi

California Animal Health and Food Safety Laboratory (Puschner, Poppenga, Tor, Filigenzi) and Department of Pathobiology, Microbiology and Immunology (Pesavento, Lowenstein), School of Veterinary Medicine, University of California, Davis, CA 95616

In a pilot study, melamine was added to the diet of cats at 0.5% and 1%. No effect on renal function was observed after feeding two cats for 10 days. Subsequently, melamine and cyanuric acid were administered in the diet at 0% (control), 0.2%, 0.5% and 1% each. Approximately 12 hours after dosing, the three dosed cats developed slight depression, vomiting and anorexia. The cats did not eat the second spiked food sample offered. Renal function was significantly impaired (marked increase in serum urea nitrogen and creatinine) in all three dosed cats when assessed at 36 hours after the initial melamine and cyanuric acid exposure.

The three dosed cats were euthanized at 48 hours after dosing and a full post-mortem examination was performed. Atypical gross findings were limited to the kidneys (all cases) and lungs (2 cats). In all cats, typically fan-shaped, birefringent crystals were detectable by gross analysis, urinalysis, touch impression, and histology. Histology on either alcohol fixed or formalin fixed tissue revealed crystals, often occlusive, within the distal nephron extending from the renal papilla to the cortex. Renal interstitial edema was severe, and present throughout the kidneys. At the cortico-medullary junction, hemorrhage and tubular damage were present in regions remote from detectable crystals. In two cats there was pulmonary edema.

Presented at the 50th Annual Conference of the American Association of Veterinary Laboratory Diagnosticians, October 18-23, 2007.

AAVLD Survey of Cases

AAVLD survey of pet food-induced nephrotoxicity in North America, April to June 2007

Wilson Rumbeiha, Dalen Agnew, Grant Maxie, Michael Scott, Brent Hoff, Barbara Powers

Diagnostic Center for Population and Animal Health, College of Veterinary Medicine, Michigan State University, Lansing, MI 48910-8104 (Rumbeiha, Agnew, Scott); Animal Health Laboratory, Laboratory Services Division, University of Guelph, Guelph, Ontario, Canada N1H 6R8 (Maxie, Hoff); Veterinary Diagnostic Laboratory, Colorado State University, Fort Collins, CO 80523 (Powers)

Following the outbreak of pet-food induced nephrotoxicity in cats and dogs around March 15, 2007, the AAVLD commissioned a voluntary survey of AAVLD-accredited laboratories that was eventually expanded to cover commercial laboratories and veterinary clinics across North America. Following this outbreak, there were numerous unsubstantiated reports of thousands of pets having been affected and/or died as result of having consumed melamine-contaminated food. There was also confusion surrounding the case definition of what truly counted as pet-food induced illness or death. This ongoing survey was designed to collect data from veterinary professionals across North America to determine: a) the numbers of confirmed cases of pets affected by melamine-contaminated food; b) the geographical distribution of cases across North America; c) common threads among affected animals; and d) to develop a case definition of what actually constituted a case of pet food-induced illness or death.

To achieve the above objectives, a questionnaire consisting of 17 questions was designed. Collection of data started on April 5, 2007 and is still ongoing using a Web-based format via the on-line tool Web Monkey™. For purposes of this abstract, analysis has been performed on data collected up to June 6, 2007; 486 cases had been entered in the survey database. Of those cases, 139 (29) were excluded because the data did not meet the criteria for further analysis.

Of the 347 cases that met the criteria, 235 (68) were cats and 112 (32) were dogs. The sex of 8 cats was not identified, but of the remaining 227 cats, 48 were male and 52 were female. Fifty nine of the dogs (53) were male and 47 were female. The age range for cats was 0.17-19 years and for dogs it was 0.33- 19 years. Of all the cases, 98 were reported from the US, with the remaining 2 reported from Canada. From North America as a whole, the majority of cases were reported from Texas, California, Ontario, Illinois and Michigan in descending order of frequency. It was interesting to note that of all the cat and dog cases, 61 and 74 respectively were dead and the rest were either ill or had recovered at the time of reporting. Among US cats, the top 2 foods responsible for illness were lams and Special Kitty US in descending order of frequency. The top 2 food products most responsible for illness among dogs in the US were Alpo Prime foods followed by 0l' Roy US. In Canada, 20 of 27 cases reported were cats, and Special Kitty Canada and President's Choice were the 2 foods most frequently associated with illness. Another interesting finding was that of all the dog and cat cases, an equal proportion (18) used 3 or more of the 8 diagnostic criteria used to confirm pet-food-induced nephrotoxicity. These criteria were as follows: a history of having ingested recalled food; finding characteristic crystals in urine; gross findings of characteristic yellow crystals in the kidneys; histologic findings of distinct crystals in the kidneys; analytical confirmation ofmelamine in urine; analytical confirmation ofmelamine in tissues; analytical confirmation ofmelamine in food; and analytical confirmation ofcyanuric acid, ammelide, or ammeline in tissues, urine, or food.

Only a minority of cases reported in the survey had pre-existing conditions that would be considered to make affected pets more vulnerable to pet food-induced nephrotoxicity. This was surprising considering that the vast majority of cases reviewed by authors of this abstract had preexisting conditions considered to render affected pets more vulnerable to pet food-induced nephrotoxicity, including glomerulonephropathy and tubulointerstitial nephritis.

Presented at the 50th Annual Conference of the American Association of Veterinary Laboratory Diagnosticians, October 18-23, 2007.

What lessons were learned?

• Both pet and human food supplies are vulnerable to chemical contamination.

• There is a new awareness on the part of regulatory agencies about food vulnerabilities.

• Sophisticated analytical methodology is capable of quickly detecting thousands of potential contaminants.

• While there was good communication between state and federal agencies in many instances, additional effort is needed to improve flow of the free flow of information.

• Programs such as the Food Emergency Response Network (FERN) proved their worth in responding to an emergency.

References

Chemical and Engineering News Summary at http://pubs.acs.org/cen/science/86/8619sci3.html

FDA Risk Interim Risk Assessment at http://www.cfsan.fda.gov/~dms/melamra.html

Brown, C.A., Jeong, K., Poppenga, R.H., Puschner, B., Miller, D.M., Ellis, A.E., Sum, S., Cistola, A., and Brown, S.A. (2007): Outbreaks of melamine associated renal failure (MARF) in dogs and cats in 2004 and 2007. Journal of Diagnostic Investigation, 19(5):525-531.

Puschner, B., Poppenga, R.H., Lowenstine, L.J., Filigenzi, M.S., and Pesavento, P.A. (2007): Assessment of melamine and cyanuric acid toxicity in cats. Journal of Diagnostic Investigation, 19(6):616-624.