Keeping E. coli O121 Out of Flour

As kids, we used to help our mom clean the kitchen while the cookies were baking in the oven. We would diligently wipe the beaters and bowls clean of cookie dough and fight over any chocolate chips. Those days are sadly gone. The FDA recently sent warnings that raw cookie dough must not be consumed. Not due to the raw eggs in the dough, but gasp, cough, the flour! Yes, a 30 million pound recall of flour has expanded in scope several times due to the presence of E. coli O121.

E. coli non–O157 STEC Background

Most members of Escherichia coli (E. coli) are harmless and live in a symbiotic relationship in the intestinal tracts of humans and animals. However, a few strains are pathogenic and can cause serious disease. Certain E. coli, called Shiga-toxin E. coli (STEC), cause disease through the production of a “Shiga” toxin that is excreted in the intestine once the organism is ingested. E. coli O157:H7 is the most notorious of this group. One egregious property of the STEC group is the low infectious dose. It has been reported that doses as low as 10-100 colony-forming units can lead to symptoms, which include bloody diarrhea and can lead to kidney failure, especially in the very young, elderly, or immunocompromised individuals. Beyond E. coli O157, approximately 70 percent of non-O157 STEC infections in the U.S. are caused by six other species and have been coined the Big 6: O26, O45, O103, O 111, O121, and O145. Of those, approximately 6 percent of the overall infections were caused by O121.

Cattle have been identified as the major source for O157 and non-O157 STEC. Thus bovine intestinal matter cross-contaminating onto raw meat is the predominant vehicle of transmission. Nevertheless, non-meat foods have also been implicated in outbreaks and include milk, produce, and water. Now, flour can be added to the list. It seems doubtful that these food groups are sources in and of themselves, rather they have been cross-contaminated likely from a bovine source. For instance, contaminated irrigation water may be a source for agricultural products. Person-to-person transmission has also been identified as a potential route of exposure.

In regards to testing, most STEC testing methods are validated for use with meat and poultry products, but unfortunately not other food matrices. The current STEC testing methods involve an initial rapid screening that will identify if one of the Big 6 non-O157 STEC is present. Testing may be stopped at this point and corrective/preventative actions taken. If testing continues, the screening data will be confirmed through a series of complex assays. Technically trained scientists are needed to carry out the confirmation methods due to their complexity. The results, depending on the test method, will not distinguish between the species but merely confirm that a non-0157 STEC is present in the sample.

The recent flour recall may prompt test methodology to expand from its current focus of meat-based products to other food matrices. For those non-meat or poultry-based producers, if testing for non-O157 STEC is conducted, testing options should be discussed. It is also important that the test be validated for the test matrix. Validating the test method with the matrices is a critical component for data integrity. Furthermore, due to the confirmation complexity, the producer may want to verify that the lab be certified to perform these test methods. One certification tool is ISO 17025 and most labs will publish the test methods pertaining to their certification.

In-Plant Procedures for Keeping Non-O157 E. coli Out of Finished Product

Millers. When lots can be segregated because of daily validated wet cleaning and sanitation procedures, the “clean-up to clean-up timeframe” effectually breaks a production cycle. This will provide three benefits: 1) reduce microbial loads; 2) remove transient organisms, and 3) provide lot segregation. In cases where wet washes are not used, and not wanted, an alternative method to reduce the microbial load is to apply chlorine dioxide gas as a dry sanitizer. Pure Line has developed a patented chlorine dioxide (CLO2) development process wherein water, as humidity, is absent from the gas. Commissioned research has demonstrated validated log reductions for E. coli along with other food-based pathogens. Current research is ongoing for the application of gas onto or throughout dried products and its ability to reduce the microbial populations. For the plant environment, the CLO2 product can be used on hard surfaces as a sanitizer. For those surfaces that are coated with product, the gas will not be able to penetrate through to the surface beneath without pressure. In these instances, Pure Line has developed a Blower Box that will dispense the gas with pressure.

Sampling plans. The validation and ongoing verification of suppliers is a component of a comprehensive food safety program and in keeping with the Food Safety Modernization Act. This program should be established to include an initial, robust testing regime of in-bound ingredients and incorporating skip lot testing as part of on-going supplier qualification programs.

One sampling plan, considered a rule of thumb, is the Association of Official Agricultural Chemists square root of N plus one for sampling lots of wheat, flour, dried fruit, or other bulk agricultural products wherein N equals the lot size. Acceptability limits for each supplier or raw ingredient category and subsequent lot qualification protocols are components of the program. For example, a supplier would be qualified after the initial sampling regime wherein testing of each lot is conducted for a pre-defined period of time or number of inbound lots. The testing is conducted per unit and is performed on an individual sample basis with a criterion of zero defective/out-of-specification results. Once the supplier is qualified, subsequent shipments of inbound goods would be verified with the square root of N plus one testing regime, but using a composite of the units rather than individually tested. Since the organism tested in this case is non-O157 STEC, a pathogen, a standard of 100 percent of the samples must meet the criteria set. Inbound lots can be tested on a skip-lot basis.

Sampling. Once in production, producers should analyze their manufacturing environment and equipment by taking both environmental and in-process product samples. In-process product sampling means that a test and hold program is in place and corrective/preventative actions are established before testing is initiated. In the best-case scenario, product sampling is conducted by an auto-sampler, particularly at the packaging step with each lot tested. Although end product testing is not considered a representative sample, taken together, the testing of inbound raw-ingredients, plant environmental and process equipment sampling, in-line, and finished product testing provide a picture of the microbiological landscape of the process over time. Tracking and then trending the data will provide a depiction of events such as seasonal variations or the effects of supplier or process changes.

Corrective/preventative actions. Out-of-specification responses have the best outcome when they are developed and documented as written programs before testing programs are conducted. In most cases, an out-of-specification result stemming from environmental samples should serve as an early warning or detection mechanism. Often when product is implicated, it is a sign of a condition that has been manifesting for some time. The corrective action/preventative action, or CAPA, program, works to immediately minimize the risk. This will often involve maintaining a hold on the product if a food contact surface or product is involved. If a non-food contact environmental site is involved, the site is immediately spot cleaned, sanitized, and dried. A documented investigative process is then conducted by the HACCP, aka Hazard Analysis and Critical Control Points, team. The multidisciplinary team views the site and looks for root causes. Once a root cause is identified, further corrections and preventative measures can be targeted and then implemented. Although difficult, the process is often a test in patience as it can often span over a multi-week timeframe especially if there are construction events t as part of the corrective/preventative actions.

Using Flour as a Raw Ingredient

During the milling process, flour usually does not undergo a microbial kill-step, but it is expected that further down the production chain, a kill-step is included in the baking process. If a producer has used recalled/implicated flour, demonstrating the use of a validated thermal lethality step may be of importance to reduce the potential or scope of a recalled ingredient. It is, however, incumbent on the ability to demonstrate that a validated thermal lethality has been conducted. To validate an oven, the key processing parameters that should be considered are as follows:

  • Using an identified and consistent line speed that is verified;
  • Demonstrate uniform heating throughout oven (no cold spots) using thermocouples;
  • Inoculate batches with most heat resistant organism that is identified in the Hazard Analysis—if a surrogate is used, provide rationale why the surrogate chosen is relevant to the product;
  • Inoculate batches with a mixture (cocktail) of identified strains; and
  • Replicate study with at least two, preferably three trials.


Pathogenic organisms continue to surprise us by finding their way into products that we once considered safe. There are processes that can be done to minimize their occurrence once we understand their origin and how they may be cross-contaminated. We may not be able to help mom clean up the kitchen by wiping the bowls clean of raw cookie dough when the rest of the cookies are baking, but we can do our part in helping everyone enjoy safe, quality products.

Dr. Deibel is director of microbiology at Covance Food Solutions and leads the consulting divisions. Reach her at Baldus works at Covance Food Solutions in the Microbiological Consulting division leading the Food Safety Training. Reach her at

Leave a Reply

Your email address will not be published. Required fields are marked *