Aflatoxin Concerns

On November 21, 2006, the FDA announced the start of a nationwide survey of aflatoxin levels in distiller’s grains, a high-protein, low-carbohydrate byproduct of ethanol production commonly used in animal feeds, including horse feed. They stated that, because of the increased use of distiller’s grain byproducts in animal feeds, the "FDA’s Center for Veterinary Medicine (CVM) is concerned about the potential animal and human health hazards from the use of aflatoxin-contaminated distiller’s grains as ingredients in animal feeds." Aflatoxin is also a potential contaminant in damaged corn or other grains, seed meals and seed meal byproducts.

On May 7, 2008, the FDA released a Draft Framework of the FDA Animal Feed Safety System, a proposed guideline for getting the FDA more directly involved in overseeing the safety of animal feeds. The plan includes an emphasis on prevention and states, in part:

"Process control is a systematic approach designed to ensure feed safety though the identification and use of appropriate controls during the manufacturing, packaging, storage, and distribution of feed ingredients and mixed feed.  Feed process control entails measures seeking to prevent, or eliminate or reduce, to an acceptable level, risks to animals and humans.For example, these process controls can include following written procedures to ensure aflatoxins are not present at unsafe levels by testing incoming loads of feed ingredients known to be susceptible to the molds that produce this mycotoxin."

Implementation of those goals is still many years away. As detailed in our July 2007 article on feed safety, the manufacturing process and testing of raw feed ingredients for equine feeds is not regulated. With the exception of spot checks by individual state Departments of Agriculture, you’re basically at the mercy of the manufacturer.

Conspicuously absent on the FDA website, on the same date as the future plans were posted, was any mention of the massive recall of equine feeds along the East Coast. The recall involved Purina, Tractor Supply’s DuMOR and Farnam Platform feeds (DuMOR and Platform are milled by Purina), produced during specific time frames from three Purina plants (see chart).

By the time the recall notices went out to dealers, some of these feeds had been on the market for as long as three months. The recalled feeds were manufactured between November 2007 and March 2008. However, the recall notice on the Purina web site did not go up until the first week in May, six months after the first manufacturing date of the involved feed and two months after the manufacturing date of the last involved feed. Farnam posted a similar notice on their Platform Feed site a week later, three months after the last manufacturing date of their involved feeds. We could not find a notice on the Tractor Supply Company website. No press releases were issued.

Companies are attempting to calm owner concerns by using the term "voluntary retrieval" (see box) and emphasizing that no confirmed aflatoxin-related health problems have been reported. However, we believe that confirming that an illness is aflatoxin-related after the fact is virtually impossible when there’s no feed left to test to confirm it was the source. Even if liver biopsy and liver aflatoxin analysis proves an aflatoxin damaged liver, it would be difficult to prove it was the feed unless the feed was available for testing, too.

Most feed dealers inventory and order weekly, keeping no more than a two-week supply of feed on hand. Commercial processed feeds should generally be fed within three months of the day of manufacture. However, it’s not uncommon for distributors to supply stores with older feeds if they have overstocked, or for feed dealers in more remote areas to take longer to move their minimum bulk order.

Feed Amounts

We asked Purina what the total tonnage of feed involved was, and how much of it had actually been recovered. We were told: "The amount of horse feed involved in the retrieval varied by plant. I’ve been unable to obtain final numbers as of yet," and that they hadn’t finished determining how much feed was actually returned. However, the FDA Enforcement Report bulletin for May 21, 2008, had some information on the recall that involved the Harrisburg, Pa., plant. For a total of 68 different feeds (including a variety of animal feeds in addition to horse feed) produced during just a one-month period, January to February 2008, the total tonnage was 2,511.25 tons of feed. The recalls from the Statesville and Guilderland plants spanned feeds produced over a three-month period.

Quality Control

In an upcoming product-rating article on the quality of commercial horse feeds, we’ll ask manufacturers to give us details of their quality-control programs. Obviously catching problems like this before the feed is bagged and out the door is ideal. However, the fact of the matter is that when dealing with hundreds of tons of raw materials on a daily basis spot checks can’t be 100% accurate. The more safety checks built into the system, the better, of course, but the burning question with this recall is how it was handled.

The first mention of the problem on the FDA’s website, in any location, was found on May 21, 2008, over six months from the manufacture of the first suspect feed. We contacted the FDA on multiple occasions and received replies ranging from "We are working with Purina on a health-hazard evaluation" to no reply at all.

In their May 21 Enforcement Bulletin, the FDA rated this recall as Class II, which is defined as "recalls are for products that might cause a temporary health problem, or pose only a slight threat of a serious nature.  One example is a drug that is under-strength but that isn’t used to treat life-threatening situations."

We specifically asked the FDA why this was considered a Class II recall when it involved aflatoxin and feeds targeting pregnant mares, growing horses and seniors who may be in fragile health. We also wanted to know why there is still no FDA notice posted on the Statesville, N.C., or Guilderland, N.Y., recalls, and why it took five weeks from the date of the recall letter and five months from the date of the manufacturing of the earliest involved feed for the Harrisburg plant notice to appear on the FDA site. As of the date of the writing of this article, they have not answered. Our request for the name of the lead FDA investigator was not answered.

When we questioned Purina about the way this recall was handled, they told us, "Because our product retrieval effort needed to focus on products from three different plants produced during specific time frames, we determined that a localized campaign to get the word out through our dealers, who have established relationships with our customers and best know customers’ purchase patterns, would be the most effective way to accomplish a voluntary retrieval of specific products from specific manufacturing runs." Dealers were supposed to post a notice, and presumably personally notify their customers who use these feeds.

Despite this, we have found that this recall was not common knowledge among horse owners, in fact, far from it. Does your feed dealer know you by name, have your contact information and know your buying habits? If you feed Purina feeds, did your feed dealer tell you?

We called the Tractor Supply Company’s customer-service number and were told there was no recall. We visited a Tractor Supply store, saw no notice, asked the salesman if there was a recall and were told "No." When asked to double check, he was able to find mention of it on the computer, but had no details on what products or the dates.

Purina claimed that both their own in-house testing and the North Carolina Department of Agriculture picked up the problem in February at the Statesville plant.

We asked Purina why the February recall notice was for feeds dating all the way back to November. Their reply was: "We diligently reviewed past production lots at a higher sampling rate.We looked hard in order to establish whether there had been additional contamination earlier in the year, and based on subsequent feed product testing by the state and Purina Mills LLC we felt it was prudent as a precautionary measure to go back to November code dates." In other words, it appears when they looked harder, they found it in feeds dating as far back as November 2007.

We also asked why two more recalls, the most recent dated April, were necessary if the problem was found in February. Their answer was, "As we continued our testing we determined that we should initiate retrievals at the other two plants as well." We’d translate that to mean, "Because we continued to find aflatoxin contamination."

In response to what safeguards would be implemented to prevent this in the future, we were told they have suspended buying from the supplier of the contaminated ingredient (which is still undisclosed) and that they were "reviewing our pre-ingredient aflatoxin screening protocols."

Bottom Line

We’re disappointed in how this feed recall was handled. The companies and agencies involved should be able to do better than this. The public deserves to know immediately when a problem of this magnitude could be threatening the health of their horses.

Reporting potentially contaminated grain two to seven months after the feed is manufactured is unacceptable to us. At every level—state, federal, manufacturers/distributors—the public was kept in the dark. Purina has said, "We regret the anxiety this situation has caused for customers." Considering how this recall was handled, our anxiety is ongoing.

Article by Eleanor Kellon, VMD, our Veterinary Editor. She has extensive experience with high-performance horses. With her husband, she breeds, races and trains Standardbred harness horses in Pennsylvania. She has written countless articles and several books, including "The Older Horse."