Header graphic for print

Food Safety News

Breaking news for everyone's consumption

How Did E. coli O145 Contaminate Lettuce? Part II

A look at how E. coli O145 could have contaminated romaine lettuce on a farm in Yuma – Part II

As state and federal public health officials continue to investigate the E. coli O145 outbreak tied to bagged Freshway Foods romaine lettuce, which has sickened 19 in 3 states, many questions remain.

The supply chain from the field to the supermarket is a long one, with many potential points along the way for contamination to occur. Where did the lettuce pick up E. coli O145, a pathogen found primarily in cattle and wildlife feces? According to the latest out of the U.S. Food and Drug Administration (FDA), investigators are focusing on an undisclosed farm in Yuma, Arizona which could be linked to the outbreak. If the contamination did occur on the farm, how could it have happened?

lettuce-field1-featured.jpgUnlike Salinas Valley, America’s salad bowl, which has been the source of several E. coli outbreaks, including the Dole spinach outbreak in 2006, Yuma-grown leafy greens have never been implicated.

Food Safety News paid a visit to the Yuma area and talked with epidemiological experts to explore a number of hypotheses. This series will look at three ways the E. coli O145-contaminated lettuce–if it was grown in Yuma–could have picked up the bug. Yesterday, Part I explored dust and mud as possible modes of contamination. Today Part II looks at wildlife intrusions, and tomorrow Part III will discuss irrigation water.

Part II – Wildlife Intrusions

As we noted yesterday, one of the largest cattle feedlots in the country is located in Wellton, AZ, around 20 miles, as the crow flies, from the heart of leafy green production in the Gila and Dome Valleys near Yuma. Naturally, the high concentration of cattle and manure produces a fecal, muddy mix in which E. coli bacteria can thrive. That mud, or mud-turned dust, could be picked up by birds or other animals stopping through the area.

With the exception of migratory birds and desert rodents, Yuma has very few wild animals that could venture onto leafy green farmland carrying E. coli bacteria. There are no forests in close proximity to the fields, unlike the Salinas Valley, where feral swine and other wildlife could excrete or externally carry E. coli onto a greens field. The large, multistate E. coli spinach outbreak in 2006 linked to Salinas is thought to have been caused by a wildlife intrusion, possibly by a feral pig, but a mode of transmission was never found.

“The role of wildlife as a source of foodborne microbial contamination along the farm-to-fork continuum is a long-standing concern among public health and food safety agencies,” according to a 2008 report by Edward Atwill at the Western Institute for Food Safety and Security at the University of California at Davis. “[The 2006 spinach outbreak] heightened concerns about the ability of wildlife to forage within or to transit through the produce production environment, and what biosecurity measures are in place to prevent wildlife access to foods that are minimally processed and often consumed raw.”

Wild birds, in particular, are considered potential vehicles for carrying pathogens from farm to farm. A study conducted by Canadian researchers in 2001 found indistinguishable E. coli O157 subtypes at two different feedlots approximately 100 km apart, determining that wild birds were among the only potential common vehicles shared between the two feedlots.

Another study performed by Ohio State University researchers in 2008 yielded similar results.

“The patterns of bird movement . . . along with the isolation of the pathogen E. coli O157 from both birds and cattle, further support the hypothesis that birds play a critical role in the dissemination of important foodborne bacteria among farms.”    

When Food Safety News visited Yuma, we were struck by how many birds there were. It was easy to imagine the possibility of birds playing a role in contamination in a lettuce field (All leafy greens commercially grown in Yuma have been harvested, production has moved back to Salinas).

“It’s one of the things we’re concerned about,” says Arnott Duncan, a grower who serves on the Arizona Leafy Greens Marketing Agreement committee. “But we aren’t going to be shooting the birds. There are a number of ways greens could become contaminated.”

Zach Mallove co-wrote and contributed to the research for this article.

© Food Safety News
  • hhamil

    Please provide a source for your statement, “The large, multistate E. coli spinach outbreak in 2006 linked to Salinas is thought to have been caused by a wildlife intrusion, possibly by a feral pig, but a mode of transmission was never found.”
    I found no such conclusion in Dr. Atwill’s short paper, “Implications of Wildlife in E. coli. Outbreaks Associated with Leafy Green Produce” (http://www.marlerblog.com/uploads/file/005%20Atwill.pdf) you cited. My thanks to Bill Marler for posting it on his blog.
    Nor was there such a conclusion in “How FDA Works to Keep Produce Safe” (http://www.fda.gov/ForConsumers/ConsumerUpdates/ucm094555.htm) which focuses on the 2006 spinach outbreak.
    As feral pigs are voracious and messy eaters, it makes little sense to me that a recent intrusion by a pig into that spinach field in 2006 could have been the source of the contamination. Surely, the investigators would have asked whether there was any evidence of an intrusion and with pigs, it would have been obvious.
    In the 2006 spinach contamination, there was lab confirmed surface water and bovine contamination within, as I recall, 2 miles of the contaminated field.
    Clearly, the SOURCE of the contamination was the infected cows. In turn, it is clear to me that infected cows are infecting other hosts which spread the contagion. Thus, using a HACCP style approach, the hazard must be contained at the CAFOs.
    We need to address the source of the contagion before these very dangerous E. coli. become widely established in nature. It they do, the impact on public health will be cataclysmic due to the importance of raw vegetables to human health.

  • jmunsell

    The article above states that FDA investigators are focusing on an undisclosed farm (singular, not plural) in Yuma, Arizone. Likewise, FDA has previously successfully traced back to a farm, or small number of farms, in California and Mexico where E.coli contamination of produce had occurred. It is ironic that FDA displays an ability to traceback to a solitary entity where contamination occured, in spite of the fact that FDA, by its own admission, has inspectors in produce plants as infrequently as once every several years. In stark contrast, USDA/FSIS has inspectors in every meat plant every day. Yet, USDA seldom traces back to the slaughter house of origin of E.coli contamination, preferring to terminate its investigation at the downstream further processing plant which unwittingly purchased meat which had been previously contaminated at one of its source slaughter providers. Make no mistake about this: USDA’s failure to trace back to the source is not caused by the agency’s inability to conduct tracebacks! Lack of meat tracebacks are primarily caused by the agency’s UNWILLINGNESS to conduct tracebacks to the SOURCE. FDA WANTS to determine the source, and require corrective actions to prevent recurrences. USDA is lacking in this regard, by intentional agency design. John Munsell

  • Food Safety News

    In the second paragraph of the Atwill paper it talks about cattle and wildlife being prime suspects, though a mode was never found – though there was no consensus on the probable cause, many experts and media discussed feral pigs as a likely mode of transmission.
    In total, the article really says there is so little data to help weigh out these costs/benefits – that’s not what this story was about, it was simply to think about a range of possibilities.

  • dick raymond

    harry, feral pigs were sacrificed in 2006 and found to have the same E coli 0157:H7 PFGE as was found in 7 head of cattle and the surface water that ran between the cattle and the spinach field. Of course, this PFGE matched that of those persons suffering from the outbreak. And, there were portions of the fence surrounding the field that were not intact, and there were little hoof prints throughout the field. Of course none of this is definitive evidence that the feral pigs contaminated the field, but the Epi evidence is very strong. We need the E coli vaccine used in every head of cattle in this country to help reduce this environmental spread of 0157 that is sickening and killing people. The government should pay for the vaccine to protect children from illness, same as they pay for vaccines for Polio, Measles, Mumps, Diphtheria etc.

  • Harry Hamil

    Please provide a source for your statement, “The large, multistate E. coli spinach outbreak in 2006 linked to Salinas is thought to have been caused by a wildlife intrusion, possibly by a feral pig, but a mode of transmission was never found.”
    I found no such conclusion in Dr. Atwill’s short paper, “Implications of Wildlife in E. coli. Outbreaks Associated with Leafy Green Produce” (http://www.marlerblog.com/uploads/file/005%20Atwill.pdf) you cited. My thanks to Bill Marler for posting it on his blog.
    Nor was there such a conclusion in “How FDA Works to Keep Produce Safe” (http://www.fda.gov/ForConsumers/ConsumerUpdates/ucm094555.htm) which focuses on the 2006 spinach outbreak.
    As feral pigs are voracious and messy eaters, it makes little sense to me that a recent intrusion by a pig into that spinach field in 2006 could have been the source of the contamination. Surely, the investigators would have asked whether there was any evidence of an intrusion and with pigs, it would have been obvious.
    In the 2006 spinach contamination, there was lab confirmed surface water and bovine contamination within, as I recall, 2 miles of the contaminated field.
    Clearly, the SOURCE of the contamination was the infected cows. In turn, it is clear to me that infected cows are infecting other hosts which spread the contagion. Thus, using a HACCP style approach, the hazard must be contained at the CAFOs.
    We need to address the source of the contagion before these very dangerous E. coli. become widely established in nature. It they do, the impact on public health will be cataclysmic due to the importance of raw vegetables to human health.

  • John Munsell

    The article above states that FDA investigators are focusing on an undisclosed farm (singular, not plural) in Yuma, Arizone. Likewise, FDA has previously successfully traced back to a farm, or small number of farms, in California and Mexico where E.coli contamination of produce had occurred. It is ironic that FDA displays an ability to traceback to a solitary entity where contamination occured, in spite of the fact that FDA, by its own admission, has inspectors in produce plants as infrequently as once every several years. In stark contrast, USDA/FSIS has inspectors in every meat plant every day. Yet, USDA seldom traces back to the slaughter house of origin of E.coli contamination, preferring to terminate its investigation at the downstream further processing plant which unwittingly purchased meat which had been previously contaminated at one of its source slaughter providers. Make no mistake about this: USDA’s failure to trace back to the source is not caused by the agency’s inability to conduct tracebacks! Lack of meat tracebacks are primarily caused by the agency’s UNWILLINGNESS to conduct tracebacks to the SOURCE. FDA WANTS to determine the source, and require corrective actions to prevent recurrences. USDA is lacking in this regard, by intentional agency design. John Munsell

  • hhamil

    Thanks to FSN and Dick Raymond for the additional information. It is great to be able to have a genuine discussion though passions may be high.
    The strongest statement in the second paragraph of Atwill’s paper is “Although the definitive source of E. coli O157:H7 for the 2006 outbreak was never determined, both cattle and wildlife were high on the list of species of concern during the outbreak investigation.” There is no distinction between the likelihood of the cattle or wildlife as the source of the contamination.
    Your statement above (“The large, multistate E. coli spinach outbreak in 2006 linked to Salinas is thought to have been caused by a wildlife intrusion, possibly by a feral pig, but a mode of transmission was never found.”) shows a certainty that is not present in the study when you name wildlife as the likely source.
    Once again, there is nothing in either paper that says the contamination is “thought to have been caused by a wildlife intrusion.”
    I reiterate my earlier comment, “As feral pigs are voracious and messy eaters, it makes little sense to me that a recent intrusion by a pig into that spinach field in 2006 could have been the source of the contamination. Surely, the investigators would have asked whether there was any evidence of an intrusion and with pigs, it would have been obvious.”
    FSN and food safety advocates regularly stress the need for good science. Good science always requires accurate reporting of facts as does good journalism and is always subject to critiques like this one.
    I contend that you have inaccurately reported your unwarranted conclusion as if it were fact; therefore, I ask that you reduce the certainty in your statement so that your readers don’t misunderstand what is known.
    As FSN is quoted elsewhere (e.g., Change.org), your misunderstanding could become widespread.

  • Harry Hamil

    Thanks to FSN and Dick Raymond for the additional information. It is great to be able to have a genuine discussion though passions may be high.
    The strongest statement in the second paragraph of Atwill’s paper is “Although the definitive source of E. coli O157:H7 for the 2006 outbreak was never determined, both cattle and wildlife were high on the list of species of concern during the outbreak investigation.” There is no distinction between the likelihood of the cattle or wildlife as the source of the contamination.
    Your statement above (“The large, multistate E. coli spinach outbreak in 2006 linked to Salinas is thought to have been caused by a wildlife intrusion, possibly by a feral pig, but a mode of transmission was never found.”) shows a certainty that is not present in the study when you name wildlife as the likely source.
    Once again, there is nothing in either paper that says the contamination is “thought to have been caused by a wildlife intrusion.”
    I reiterate my earlier comment, “As feral pigs are voracious and messy eaters, it makes little sense to me that a recent intrusion by a pig into that spinach field in 2006 could have been the source of the contamination. Surely, the investigators would have asked whether there was any evidence of an intrusion and with pigs, it would have been obvious.”
    FSN and food safety advocates regularly stress the need for good science. Good science always requires accurate reporting of facts as does good journalism and is always subject to critiques like this one.
    I contend that you have inaccurately reported your unwarranted conclusion as if it were fact; therefore, I ask that you reduce the certainty in your statement so that your readers don’t misunderstand what is known.
    As FSN is quoted elsewhere (e.g., Change.org), your misunderstanding could become widespread.

  • mjayrussell

    Here is the original paper describing environmental studies during the 2006 spinach outbreak.
    Escherichia coli O157:H7 in Feral Swine near Spinach Fields and Cattle, Central California Coast
    http://www.cdc.gov/eid/content/13/12/1908.htm
    E. coli O157:H7 culture and PFGE results (from Table 1):
    26 of 77 cattle feces positive; 15 indistiguishable from the outbreak strain (note, this was a grassfed beef pasture operation, not a feedlot)
    13 of 87 feral swine feces/colonic positive, 11 indistinguishable from the outbreak strain
    3 of 79 surface water samples positive, 2 indistinguishable from the outbreak strain
    3 of 37 soil or sediment samples positive, 3 indistinguishable from the outbreak strain
    No E. coli O157:H7 isolated from irrigation water

  • mjayrussell

    Sorry about the typo – second bullet should read 13 of 87 feral swine feces/colonic feces, 8 indistinguishable from the outbreak strain. Also notable during the epidemiological investigation was the frequency of sightings/trap sucess for feral swine at that location.

  • Michele Jay-Russell

    Here is the original paper describing environmental studies during the 2006 spinach outbreak.
    Escherichia coli O157:H7 in Feral Swine near Spinach Fields and Cattle, Central California Coast
    http://www.cdc.gov/eid/content/13/12/1908.htm
    E. coli O157:H7 culture and PFGE results (from Table 1):
    26 of 77 cattle feces positive; 15 indistiguishable from the outbreak strain (note, this was a grassfed beef pasture operation, not a feedlot)
    13 of 87 feral swine feces/colonic positive, 11 indistinguishable from the outbreak strain
    3 of 79 surface water samples positive, 2 indistinguishable from the outbreak strain
    3 of 37 soil or sediment samples positive, 3 indistinguishable from the outbreak strain
    No E. coli O157:H7 isolated from irrigation water

  • Michele Jay-Russell

    Sorry about the typo – second bullet should read 13 of 87 feral swine feces/colonic feces, 8 indistinguishable from the outbreak strain. Also notable during the epidemiological investigation was the frequency of sightings/trap sucess for feral swine at that location.