Who Cares Which E. coli Makes Someone Ill?

There’s been some hoopla in the food news this week. Six more E. coli serotypes (in addition to E. coli O157:H7) are now to be treated as adulterants by USDA when they are found in raw beef trim.

The addition of these toxin-producing E. coli serotypes to the pantheon of named adulterants is largely due to the efforts of Bill Marler. And I applaud his desire to advance food safety.

But what happens when an eighth serotype causes an illness outbreak? For example, E. coli O104:H4 the serotype that caused last summer’s massive outbreak in Germany, is not one of the “super six” newly named adulterants.

I was reminded of this problem yesterday while speaking with Dr. Raoult Ratard, State Epidemiologist with the Louisiana Department of Health and Hospitals. We were discussing the illness outbreak in several southern US states, and the shiga-toxin-producing E. coli O145 strain that was recovered from patients. As of today, there are 12 confirmed illnesses in Georgia (5), Louisiana (3, including a 21-month-old girl who did not survive), Alabama (2), Florida (1) and Tennessee (1).

I asked Dr. Ratard whether he could confirm that the Louisiana illnesses were due to E. coli O145. With a verbal shrug, he replied that he did not know; Louisiana does not do an immunological identification of E. coli strains, beyond determining whether or not they are E. coli O157:H7. He opined that this would be a waste of time and resources, given the number of different strains in circulation.

Instead, Louisiana looks for shiga-toxin producting E. coli, determines whether or not the strain is E. coli O157:H7, and runs a genetic profile (known as PFGE). The PFGE result is emailed to the CDC, and the culture is purified and shipped to the agency labs. As far as Louisiana is concerned, the exact identification of the E. coli serotype is interesting from an academic perspective, but not something that they care to spend time on.

After thinking about this for a couple of minutes, I found myself agreeing with Dr. Ratard. There was a time when determining the serotype was a useful tool in tracing the source of a disease outbreak. That tool has been supplanted by a much more precise and reliable tool, in the form of genetic profiling.

Which brings me back to USDA and the “super six serotypes” that are in the media spotlight. What the agency should have done – and what I proposed back in 2009 – was to simply declare ANY shiga-toxin producing E. coli as an adulterant.

  • The toxin doesn’t care which serotype is producing it.
  • The patients don’t care which serotype is making them ill.
  • The epidemiologists no longer rely on serotyping to define an outbreak.

So why should USDA set up seven individual small targets (E. coli O157:H7 and the “super six”) instead of a single inclusive target known around the world as “shiga-toxin producing E. coli?

Beats me!

4 thoughts on “Who Cares Which E. coli Makes Someone Ill?

  1. The way Dr. Ratard studies the strains isolated from patients is quite logical. Yet, the way the industry has to protect the consumer is different. Not all “shiga-toxin producing E. coli” (so called stx+ strains) are harmful. Only some of them are potentially pathogenic, and even less are potentially highly pathogenic. This is why the list of those designated as “adulterant” cannot include all stx+ E. coli. If all food batches in which stx+ strains are detected had to be discarded, the amount of food of animal origin still on the market would be reduced dramatically…


    1. Olivier, I understand your point, and it has some merit. But one could make the same argument about Salmonella. Not all salmonellae are equally pathogenic or equally virulent. Yet all salmonellae are treated equally from an enforcement, HACCP and quality/safety assurance perspective.


      1. Hi
        That is not always the case for Salmonella’s, the EU2073 Guidelines 2009 for Raw Poultry meat only requires you to look for Salmonella typhimurium and enteridis.
        Similar could be applied to Listeria monocytogenes as not all have the same virulence and some are not pathogenic. I suppose for practicability you have to start and finish somewhere.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.