Want to know more about the origin of Shigatoxin-producing E. coli?

 Bill Marler is a very rare individual: he is a lawyer with an ability to grasp the principles of microbiological food safety and to write about them in a way that is accessible to the general public.  Bill has been closely involved in food borne illness litigation since 1993, when he obtained a $15.6 million settlement for Brianne Kiner, a nine year old girl who suffered HUS and ultimately kidney, liver and pancreas failure after eating Jack-in-the-Box hamburger.  He has become a fiercely active food safety campaigner.

Recently, Bill conducted a literature review on the origins of Shigatoxin-producing Escherichia coli  and has published the first two parts on his food safety blog.  These publications are well worth reading - they are not written in highly technical language, but the reference list allows those readers sufficiently interested and capable to follow up his sources and read the original material.

Links to the first two parts are shown below:

http://www.marlerblog.com/lawyer-oped/where-the-hell-did-shiga-toxin-e-coli-come-from-a-literature-review---part-1/

http://www.marlerblog.com/case-news/where-the-hell-did-shiga-toxin-e-coli-come-from-a-literature-review---part-2/

There is just one problem with this review; Bill has taken away my option to set such a study of literature as an assignment for my postgraduate students in my course "Frontiers of Food Microbiology".  The students will still be expected to read and discuss the material, but I can't use their reviews for assessment purposes.  Thanks Bill!

Big One becomes Big Seven

Escherichia coli O157:H7 has now been joined on the FSIS most unwanted list by the "Big Six" - other strains of E. coli capable of producing Shigatoxin.

As we saw in May and June 2011 in the German outbreak, Shigatoxin-producing E. coli can cause a potentially deadly food borne infection that can leave survivors damaged for life.  The outbreak also resulted in huge economic loss in several European countries as fresh produce was either banned by authorities or shunned by consumers.

As of the 13th September 2011, these seven strains - O157:H7, O26, O11, O103, O121, O45 and O145 - will not be permitted in non-intact raw beef in USA.  If they are found to be present, the meat must either be destroyed or cooked before sale.

It has been a long road to get to this point. In October 2009, Bill Marler, a US attorney, filed a petition with USDA/FSIS for an Interpretive Rule declaring all enterohaemorrhagic Shigatoxin-producing serotypes of E. coli, including non-O157 serotypes, to be Adulterants within the meaning of the Federal Meat Inspection Act.

Not much has been heard since, though there has been a certain amount of correspondence.  Even the regulatory authorities didn't seem to have a united view.  On the 3rd June 2011, the Deputy Director of CFSAN, Donald Kraemer, stated on the FDA website "FDA considers any disease-causing strain of E. coli in food to be illegal”.  The FSIS deems only O157:H7 to be an adulterant.  Tellingly, Kraemer’s comment was removed from FDA website on 5th July 2011.

The new declaration is being hailed as a victory.  Elisabeth Hagen, head of food safety at the Department of Agriculture, said that this was "one of the biggest steps forward in the protection of the beef supply in some time.  We’re doing this to prevent illness and to save lives.”  A worthy cause.

However, I have some concerns about this optimism.  Will the reclassification of the Big Six make meat safer?  I'm not so sure.  Certainly, if these bacteria are detected in meat, the product will not be allowed on the market unless it is diverted to cooked products.  This may be a challenge for the meat processors - since 1994, O157:H7 in raw ground beef has been declared an adulterant with zero tolerance by FSIS.  In October 2007, the Topps Meat Company recalled 21.7 million pounds of ground beef, bringing the total recalls in the US between April and October 2007 to over 30 million pounds of red meat, mostly hamburger.  A company manufacturing frozen hamburger patties is unlikely to have the capacity to redirect this much meat to a cooking process and there may be difficulties in finding a buyer for the product, so it may have to be destroyed.

Secondly, as far as I am aware, there is no requirement for processors to test for these bacteria.  If this situation doesn't change, the first indication that something is wrong may still be when people start showing up at the hospital with gastrointestinal disease.

Thirdly, the cost of testing is currently very high and the testing may take up to 5 days, even when things go well.  I have already written about the impossibility of guaranteeing safe food by testing.

Finally, testing for the Big Seven will miss any Shigatoxin-producing non-members, as was the case with the German outbreak of O104:H4

The FSIS move is a good start and is motivated by good reasons.  However, the only way that safety of food can be improved is by development of risk management plans and rigorous application of critical control points throughout the food chain, including food service outlets, i.e. farm to fork.  Consumers should not receive contaminated food, but they too must play their part by prevention of cross contamination in the home and proper cooking of foods.  Future posts will deal with some of these approaches.

Where does that leave us with foods like sprouts?  Food for thought.

Since I wrote this article, Shawn Stevens has written an article "Big six declared Adulterants: Is it a good thing?" in Meatingplace.com, a blog for the meat industry in the US.  It seems that he too has some concerns. You may have to register to read the article, but registration is free.

Can we control toxigenic E. coli?

Last week at the New Zealand Institute of Food Science and Technology annual conference, I presented a paper on non-O157:H7 shigatoxigenic E. coli.  That's a bit of a mouthful, if you'll pardon the unfortunate pun.  These bacteria have gained the ability to produce a toxin that can destroy intestinal epithelial cells and also damage the kidney.  E. coli O157:H7 is probably the best known of the Enterohaemorrhagic E. coli  (EHEC) types.  The strain causing the outbreak in Germany and other parts of Europe is a non-O157 type and indeed appears to be from a different group - the Entero-Aggregtive EC.

During the session, two other papers were presented, one on Cronobacter sakazakii - the neonate nasty, and one on Salmonella.  All three presenters talked about control of these food borne disease-causing organisms.

During the discussions, it became apparent that there was some difference of opinion on the effectiveness of control measures during production and processing of foods.

For example, the poultry industry in New Zealand has been remarkably successful in controlling Salmonella in poultry flocks by strict management of biosecurity, and thus eggs and chicken meat are essentially free of Salmonella.  Unfortunately, these controls have not worked for Campylobacter and there is still a significant rate of C. jejuni illness that can be traced to poultry, though this rate has fallen over the past two years.

Similarly, when I suggested that control of Salmonella would likely also control E. coli, a friend and colleague stood up and said that these two organisms are significantly different and controls would need to be tailored to each.  I invited him to come outside and we'd sort it out by fisticuffs!  However, it became obvious that we were talking at cross purposes and we were both right.

Salmonella, Cronobacter and E. coli are all vegetative bacteria.  That is, they do not form spores and are not particularly heat or chemical resistant.  Thus, during food processing or in the hands of the consumer, control is easy - heating to about 75-80C for a few seconds will kill them all.  If we prevent cross contamination, then the finished food will be safe to eat.

It is much more difficult to set conditions during primary production that will control these bacteria.  Vegetables are usually grown in soil.  They therefore become contaminated with soil organisms.  Even hydroponic growing systems may become colonised by bacterial biofilms and thus contaminate the products.  Animals and birds all carry populations of bacteria in their guts or on hides, hair, feathers and feet.  It is impossible to eradicate these bacteria, so they must be controlled during processing. 

But what do we do with vegetable sprouts?  These products must be regarded as hazardous - they are grown in conditions of high humidity and at temperatures that support the rapid growth of bacteria.  If pathogens, such as E. coli O104:H4 are present in the seeds, in the water or the equipment, they can grow rapidly.  Stir-frying will probably kill the bacteria, provided the temperature gets high enough.  Sprouts are also eaten raw, so there is no controlled lethal process step and the consumer ingests the bacteria, sometimes with fatal results.  If people are to continue to eat raw sprouts, we need to develop some means of decontaminating them, such as a rinse in a bactericidal chemical solution.  Nothing is perfect, but control of production, processing and distribution, together with consumer education, should decrease the likelihood of a similar outbreak to that currently occurring in Germany.

Colonel Mustard, in the Dining Room, with the Sprouts

The last couple of weeks have been like the old game of Cluedo, with everyone running round accusing various participant of foul play and trying to find the culprit.  It would be amusing if it were not so deadly serious.

It now appears that the outbreak of Escherichia coli O104:H4 in Germany, which has now reached 16 countries, was probably caused by fresh vegetable sprouts grown in a farm in Uelzen, near Hamburg.  Unfortunately,  cucumbers from Spain, lettuces and tomatoes also had the finger pointed almost indiscriminately at them.  That caused huge financial losses for the suppliers and had knock-on effects on vegetable sales throughout Europe.  In some ways, I can sympathise with the authorities and the media.  This is one of the largest and most serious outbreaks of food poisoning ever and finding the source as quickly as possibly was imperative.

At the time of writing, there have been 3517 cases of EHEC infection, resulting in an unprecedented 839 cases of Haemolytic Uraemic Syndrome and 39 deaths.  The rate of new case reporting has slowed, but it is likely that more people will become ill before this is all over.

This shows just how difficult it is to pinpoint the source of an outbreak of food poisoning in our highly integrated and widespread food supply chain.  The sprout farm purchased seeds for sprouting from both European and Asian countries.  The seeds were sprouted, using what seems to be standard and well designed conditions, and the 18 different sprout mixtures were sent to many different points in Germany.

The German authorities have been accused of being in disarray and having no proper response prepared.  There may be some truth in this - if the responsibility for food safety is spread across many local authorities and agencies, setting up a coordinated and timely response is fraught with difficulty.  But it is generally recognised that epidemiological investigations may have a success rate as low as 33% in tracking down the source of infections and the chance of success falls as time goes on.

At times like this, we often hear calls for increased testing of products before they are released onto the market.  Good try, but no cigar!  For a number of reasons, microbiological testing to assure safety of food is just not possible.  Testing is expensive and time consuming.  In some cases, the testing period exceeds the shelf life of the product.  A simple statistical calculation shows that when contamination levels are low, the number of samples that must be tested to get even a 95% chance of detection is prohibitively large and even then, 5% of contaminated samples will be accepted as safe.  Testing for E. coli O157:H7 would not have picked up the German strain and even proposed widened testing for "The Big Six" Entero-Haemorrhagic E. coli strains would have missed this one.

Regulators must be strong in the coming months.  Microbiological testing gives only a retrospective view and a poor one at that.  Imposing increased mandatory testing will not assure the safety of foods.

The only way we can ensure the safety of our food supply is to introduce controlled lethal steps in processing, such as heating or irradiation, or to put in place rigorous control of every potentially hazardous ingredient, process step, processing facility and distribution chain.  This is particularly important with high risk products such as sprouts.  What every food producer and distributor needs is a Hazard Analysis based Food Safety Programme of Risk Management.

How do you trace an outbreak? With difficulty!

The outbreak of E. coli O104:H4 food poisoning in Germany is possibly one of the worst outbreaks on record and certainly one of the most deadly.

Public health officials still don't know the source.  Spanish cucumbers were initially blamed, but it appears that might have been wrong, though it does appear that salad vegetables may have been involved.

The investigation of outbreaks like this is a science in itself, but must be frustrating for epidemiologists.  Our modern supply chains are extremely complex - foods are sourced from all over the world and suppliers may make up orders from many different primary producers.

André Picard, — Public Health Reporter of Canadian newspaper "The Globe and Mail" wrote a neat article with associated links and videos on the investigative process and its difficulties.  I have reproduced part of it below:

"...real life does not have tidy TV endings. Despite determined efforts and great expense, the precise source of poisoning is found in fewer than one-third of outbreaks. But here is a quick look at how public officials try to resolve those puzzles. 

First, there is a trigger event – usually people showing up in emergency rooms with violent cases of diarrhea and vomiting that are the hallmarks of food poisoning. Samples – stool, saliva, sometimes blood – are taken. Serious illnesses like E. coli get special attention: 
Reporting them to public health authorities is mandatory.

Data from lab tests are routinely sent to local public health authorities and to PulseNet, an electronic database. While most cases of food poisoning are sporadic, patterns can be spotted quickly – there can be a cluster in a single city, or a sudden spike over a wide geographic area. Labs do DNA fingerprinting of pathogens like E. coli that can show if the illnesses have a common source. 

In this instance, three deaths of patients with E. coli infection in Germany set off alarm bells. They all had the same DNA fingerprint, strongly suggesting a common food source.
When an outbreak is suspected, epidemiologists (disease detectives) start questioning those who are sick. In these “hypothesis-generating” interviews, patients are asked about specific high-risk foods (such as sprouts or unpasteurized milk) and for detailed recollections of what they have eaten in recent weeks. 

This can produce, fairly quickly, a list of suspected culprits. E. coli is a fecal bacterium, for example, so raw foods are a focus. In Germany, a day after the deaths, consumers were warned to not eat tomatoes, cucumbers or lettuce. Three of four cucumbers tested came from Spain, so the country was quickly identified as a culprit in media reports. 

Public health officials share information gathered from patients with food safety regulators (such as the Canadian Food Inspection Agency), who try to trace the commercial supplier of a common food source. 

In their interviews, epidemiologist collect information on where food is purchased; if many of the patients shopped at the same food chain, for example, the CFIA would identify their vegetable suppliers and try and trace the movement of food through distributors and back to the farm. 

But there can be hundreds, even thousands, of suppliers and contamination of food can occur at any point in the process of getting food from the farm to the table – during production, processing, storage, distribution or preparation.  Finding the “locus of contamination” becomes an almost insurmountable process of elimination. Currently, farms in Europe that produce vegetables that are consumed raw are being set upon by all manner of investigators who will test products, water sources and so on. 

The paradox is that the more people get sick, the easier it is to ultimately find the culprit because the number of food sources they have in common becomes smaller. Often, despite mountains of data, the leads run cold. But the demand for answers does not wane.
The challenge for public health officials and regulators is finding the right balance. They issue speculative warnings like “Don’t eat Spanish cucumbers,” but more often than not they are wrong, and that can cause grievous damage to food producers. 

At the same time, acting slowly and waiting for a definitive answer on the source can result in countless cases of illness and death, and still no definitive answer. 

To read the full article and access the video clips, click here.

Not all reporting is of this quality; even the BBC has broadcast a news item labeling E. coli as "a virus".  Didn't the reporter talk to an expert, or even a Level 6 food science student?