FDA Warning Letters This Week 2/22/17

warning-letters

The week goes by and new letters come out. Looks like we’ve only got one food letter this week so let’s get into it.

Don’t know what warning letters are? Check out this post for a brief overview of what they are and why FDA sends them out.

WL# 17-PHI-04

CDC Public Health Image library #2287 Credit Elizabeth White (2002)
CDC Public Health Image library #2287 Credit Elizabeth White (2002)

Subject: CGMP/Food/Prepared, Packed or Held Under Insanitary Conditions/Adulterated/L. monocytogenes

Ohhhhh it’s an L. mono letter, always interesting. Already we have to pause for some background.

At the end of 2013, CDC and FDA partnered up to stop foodborne outbreaks of Listeria monocytogenes through the utilization of whole genome sequencing (WGS). This technology used to be effectively impossible 30 years ago, and prohibitively expensive and time consuming even 10 years ago.

What this means is that when L. mono is found in foods or a sick person in the hospital, they can sequence it’s entire genome to determine how closely it might be related to other L. mono cultures found in the network. The network consists of genome databases Pulsenet, GenomeTrakr, and The Listeria Initiative, which are jointly managed by CDC, FDA, and local health departments.

Once L. mono has been found and sequenced, in tandem with a traditional epidemiological investigation (interviews, additional sampling of products and environments, illness reporting), the sequence is compared to those in the database to determine if the strain that was found is “closely related” to others and see if there is a link. This is how the listeria found in the Blue Bell products/plant was traced to 10 illnesses that occurred as early in 2010, even when the listeria in the ice cream wasn’t found until 2015. Across several states and 5 years no firm link was established until WGS sequencing revealed the relationship.

There’s some debate about how transparent CDC is about determining when related isolates are causal, or to what level this data should be considered definitive in the absence of confirmed positives in products. However, FDA is going full steam ahead with using WGS as a new enforcement and Epi tool, and CDC has some good data to support the effect on outbreaks it may have had. Whether that’s because manufacturers are more careful in light of the enforcement activities or because the enforcement activities have prevented outbreaks is unclear. But no matter the end result, it’s a win for public health.

The last bit of debate is as to the extent that L. mono can be considered an adulterant. Here in the U.S. we’ve made the determination that if there is any present in the food, it’s adulterated. This is different from several other countries, including Australia/New Zealand and the UK, which allow a non-zero legal limit if L. mono is not expected to be able to grow in the product over it’s shelf life. There’s debate over that as well.

So back to the warning letter:

FDA’s laboratory analysis of fifty environmental swabs collected on September 12, 2016, confirmed that eighteen of the fifty environmental swabs were positive for L. monocytogenes. Of most importance:

Three positive environmental swabs were collected from the following direct food contact surfaces in your cheese processing room during the production of your RTE cheeses:
  • The top of the cheese slicer
  • The cheese slicer string
  • The inside of a plastic crate used to store finished cheese before packaging
–    The remaining fifteen positive environmental swabs were collected from locations adjacent to food contact surfaces and from non-direct food contact surfaces.
–    One positive sample was collected from your RTE feta cheese

They also found a positive in unpasteurized cheese during the facility visit.

This is a pretty solid connection given they found Listeria in the product itself, especially a post-pasteurization product, and that they found it on product contact surfaces where it could continuously inoculate that pasteurized product.  There are other warning letters that claim the products are adulterated when “we found it in the threshold of your entryway to the plant” that don’t always feel solid, but this one is not one of them.

On September 12, 2016, the lift arm and bowl support brackets of the mixer contained areas which appeared to be rusted and contained rough surfaces. In addition, the beater shaft housing area of the mixer, directly above the bowl support brackets, was observed to contain areas which appeared to be rusted and contain food particulates and/or foreign matter.
Food processing tools like my hand mixer have hard to clean areas that can collect food particles like cookie dough, you’ve got to check those areas every time!

Rust happens, and plants have to do regular walkthroughs and replace what needs to be replaced as it ages. But even with the best of intentions, not everything always gets replaced by the time FDA walks through. This makes it a common violation (do you have any rusty or damaged bowls or pots in your cupboard that you plan to replace soon?). But the food particulates are a good find and an indicator of a less-than-robust sanitation standard. Small, hard-to-clean areas around bolts and inside moving parts like the underside of a kitchen-aid (see picture to left) don’t get properly inspected and cleaned unless you have a dedicated program to seek these areas out.

The floors in the processing room and walk-in cooler were observed to be in disrepair, containing areas where the concrete is cracked, rough, and peeling

This one is also common, as mentioned on this site before, cracked/pitted/porous floors are a potential point where environmental pathogens like listeria and salmonella can hide from getting properly cleaned, waiting to jump back out. It’s also a common finding because refinishing or extensive repair of floors is expensive, normally planned on a long-term basis (maybe the slow season so the plant can close or when the contractor has availability), and because in other manufacturing industries it isn’t always a problem (e.g. machine shops).

We have reviewed your written responses to the Form FDA-483, received October 24, 2016, and December 7, 2016…We will ascertain the adequacy of your corrective actions during our next inspection.

Also as discussed here in the past, this is a good sign that FDA knows this company is taking the findings seriously and making real change. Otherwise they would respond with “this is inadequate” and request additional action or proof of change before closing the letter.

Greater than twenty flies landing on the floor, food processing equipment, food processing utensils, and other food contact surfaces and non-food contact surfaces…Three fly catcher tapes containing multiple flies hanging in different areas of the processing room… Dead flies on the window sills near the batch pasteurizer and three bay sink areas.

Well, they had fly catchers, so they know that flies aren’t supposed to just keep flying around, so there’s that. The manufacturer responded that they would purchase a fly zapper and new tapes, but FDA wasn’t satisfied. With pest control, you’re supposed to keep them out of the building and away from your products. So your interventions need to prevent access (find access points, block holes and unscreened windows, create breezeways and air curtains, etc.), and keep them from getting to products with interventions like ILTs.

Your maintenance of the grounds is inadequate to protect against contamination of food, as required by 21 CFR 110.20(a)… Live chickens and pigs coming within approximately one foot of the main door to the production facility and what appeared to be remnants of dead chickens and goats in close proximity to the production facility…Multiple items within approximately twenty feet of the outside perimeter to your production facility which may constitute an attractant, breeding place, and  harborage areas for pests, including, but not limited to, a chicken coop, an abandoned truck, a small four-wheeled loader, wood paneling, vegetation over six feet tall, and other small items which appear to be refuse.

Yep, sounds like a small farm/dairy. I can just picture it, can’t you? Some animals wandering, tall grass, chicken coop, and a couple old trucks and tractors getting overtaken by weeds

Outside of the animal carcasses (obvious pest and disease attractant), this is FDA throwing a little muscle at this farm to clarify that they need to treat it as a food processing facility and not as just a farm. The letter indicates that the company took action to clean up all of these items, however FDA was unsatisfied because they apparently didn’t send them pictures of the cleaned up areas nor discuss how they would keep livestock from hovering around the entrances of the food plant. FDA again noted that they would confirm the changes made were effective at the next inspection, continuing to show that that the inspection itself and initial 483 response must have gone well and the company is doing the right things post-inspection.

Sounds like this company has some work left to do, and that in this case FDA found some tangible and realistic findings of sanitation and facility problems. It helps that the findings were also supported by the L. mono data to really drive home to the company why they should be doing these things. Always nice when cause and effect tie together with GMP’s to drive positive food safety changes home.