Not much E. coli on Romaine at retail

A total of 720 whole, romaine lettuce heads were purchased from retail locations in the Upper Midwest and assessed for coliform and Escherichia coli contamination and for the presence of E. coli O157:H7.

Romaine-Lettuce-PhotosDuring a 16-month period (August 2010 through December 2011), coliform and E. coli counts were enumerated on Petrifilm, and the presence of E. coli O157:H7 and the virulence gene eae was evaluated by real-time PCR (qPCR). Over half (400 of 720) of the lettuce samples were processed with an immunomagnetic separation step before the qPCR assay. All retail lettuce samples were negative for E. coli O157:H7 when tested with the R.A.P.I.D. LT qPCR targeting a region of the O-antigen, and only two (0.28%) were positive for the eae gene when tested with LightCycler qPCR.

On Petrifilm, coliform counts of most lettuce samples (96.4%) were between <101 and 103 CFU/g, and E. coli counts for nearly all lettuce samples (98.2%) were <101 CFU/g. No seasonal trend in coliform and E. coli counts was observed throughout the examination period nor was a difference in coliform counts observed between packaged and nonpackaged lettuce heads.

These results contribute to the limited recorded data and understanding of microbial contamination of whole romaine lettuce heads purchased from retail locations, specifically revealing the absence of E. coli O157:H7 and low levels of contamination with coliforms and other E. coli strains.

 Occurrence of coliform and Escherichia coli contamination and absence of Escherichia coli O157:H7 on romaine lettuce from retail stores in the Upper Midwest

Journal of Food Protection®, Number 9, September 2015, pp. 1624-1769

Greve, Josephine D.; Zietlow, Mark S.; Miller, Kevin M.; Ellingson, Jay L. E.

Leave vegetation alone: Clearing habitat surrounding farm fields fails to reduce pathogens

The effort to improve food safety by clearing wild vegetation surrounding crops is not helping, and in some cases may even backfire, according to a new study led by researchers at the University of California, Berkeley.

cow.poop.spinachThe findings, to be reported Monday, Aug. 10, in the journal Proceedings of the National Academy of Sciences, call into question the effectiveness of removing non-crop vegetation as a way to reduce field contamination of fresh produce by disease-causing pathogens. This practice led to extensive loss of habitat in a region that is globally important for food production and natural resources.

The practice was implemented largely in response to a 2006 outbreak of pathogenic E. coli in packaged spinach that killed three people and sickened hundreds of others in the United States. That outbreak was traced to a farm in California’s Central Coast, a region that supplies more than 70 percent of the country’s salad vegetables. The disease-causing E. coli strain was found throughout the farm environment — including in the feces of nearby cattle and wild pigs — but the cause of the outbreak has never been officially determined.

“Wildlife took much of the blame for that outbreak, even though rates of E. coli in wildlife are generally very low,” said study lead author Daniel Karp, a NatureNet postdoctoral research fellow in UC Berkeley’s Department of Environmental Science, Policy and Management and The Nature Conservancy. “Now, growers are pressured by buyers to implement practices meant to discourage wildlife from approaching fields of produce. This includes clearing bushes, plants and trees that might serve as habitat or food sources for wild animals. Our study found that this practice has not led to the reductions in E. coli and Salmonella that people were hoping for.”

Instead, the study authors noted that the presence of diverse habitats bordering food crops can actually provide a number of agricultural benefits.

“There is strong evidence that natural habitats surrounding crop fields encourage wild bee populations and help the production of pollinated food crops,” said study senior author Claire Kremen, a UC Berkeley professor of environmental science, policy and management. “There have also been studies that suggest that a landscape with diverse plant life can filter out agrichemical runoff and even bacteria. Changing this dynamic shouldn’t be taken lightly.”

The researchers analyzed about 250,000 tests of produce, irrigation waters and rodents conducted by industry and academics from 2007 through 2013. The tests were conducted on samples from 295 farms in the United States, Mexico and Chile, and targeted the presence of pathogenic E. coli, Salmonella and generic strains of E. coli. The researchers combined the test data with a fine-scale land-use map to identify characteristics of the landscape surrounding the agricultural fields.

The researchers found that the removal of riparian or other vegetation did not result in lower detection of pathogens in produce, water or rodents. Overall, the prevalence of pathogenic E. coli in leafy green vegetables had increased since the outbreak, even as growers removed non-crop flora. In fact, the growers who removed the most vegetation experienced the greatest increase in pathogenic E. coli and Salmonella in their vegetables over time.

lettuce“Clearing surrounding vegetation is a costly, labor-intensive practice that threatens wildlife habitat,” said Karp. “Since it does not improve food safety, there is no reason to continue this practice.”

The study did find, however, that the likelihood of detecting pathogenic E. coli was greater when fields were within 1.5 kilometers of grazeable land than when they were farther away.

“It is unclear whether it was the cattle or wildlife grazing on those lands that were responsible for the elevated pathogen levels, but there are a number of ways that farming and ranching can co-exist in a diversified system,” said Karp.

Some suggestions include:

  • Leaving strips of vegetation between the grazed areas and fresh produce areas
  • Fencing off upstream waterways from cattle to prevent waste from going downstream
  • Planting crops that are usually cooked before being eaten – such as corn, artichokes and wheat – between fresh produce fields and grazeable lands

After the 2006 E. coli outbreak in spinach, California’s agricultural industry implemented a voluntary certification program called the Leafy Green Products Handler Marketing Agreement. At the federal level, in 2011 President Obama signed the Food Safety Modernization Act, considered one of the most sweeping reforms in farming practices in decades. Both efforts shift the focus to preventing rather than responding to outbreaks.

Notably, neither the federal law nor the state program calls for the removal of wildlife habitat surrounding crops, but private buyers, keen on retaining consumer confidence in their products, may still require growers to take steps that go beyond government regulations.

“The real worry for me is that federal law will be interpreted as the floor rather than the ceiling of what farmers should do,” said Karp. “There is this misguided idea that agricultural fields should be a sanitized, sterilized environment, like a hospital, but nature doesn’t work that way.”

Other co-authors of the study are Sasha Gennet, senior scientist at The Nature Conservancy; Christopher Kilonzo, Melissa Partyka and Edward Atwill at UC Davis; and Nicolas Chaumont at Stanford University.


A farming landscape can be co-managed for both produce safety and nature conservation. Promising practices include: (1) planting low-risk crops between leafy-green vegetables and pathogen sources, such as grazeable lands, (2) buffering farm fields with non-crop vegetation to filter pathogens from runoff (3) fencing upstream waterways from cattle and wildlife (4) attracting livestock away from upstream waterways with water troughs, supplement and feed (5) vaccinating cattle against foodborne pathogens (6) creating secondary treatment wetlands near feedlots and high-intensity grazing operations (7) reducing agro-chemical applications to bolster bacteria that depredate and compete with E. coli (8) exposing compost heaps to high temperatures through regular turning to enhance soil fertility without compromising food safety, and (9) maintaining diverse wildlife communities with fewer competent disease hosts.

Illustration by Mattias Lanas and Joseph Burg

E. coli O157 gets a boost from downy mildew in lettuce

Escherichia coli O157:H7 is more likely to contaminate lettuce when downy mildew is already present, according to U.S. Department of Agriculture (USDA) scientists.

downeymildew150723Downy mildew, a lettuce disease caused by the fungus-like water mold Bremia lactucae, is one of the biggest problems that lettuce growers must deal with.

But microbiologist Maria Brandl, with the USDA Agricultural Research Service’s (ARS) Produce Safety and Microbiology Research Unit in Albany, California, has been investigating why so many E. coli O157:H7 outbreaks can be traced back to lettuce fields when E. coli O157:H7 sources are as diverse as undercooked beef, sprouts, raw dairy, shelled walnuts, fruits and vegetables. ARS is USDA’s chief in-house research agency.

Lettuce leaves are actually a harsh place for microbes to survive. But the epidemiological evidence is indisputable about how often lettuce is the source of E. coli O157:H7 contamination.

In earlier research, Brandl found that E. coli O157:H7 preferred cut, injured and younger leaves to undamaged and older ones. Then, she collaborated with ARS geneticist and lettuce breeder Ivan Simko from the Crop Improvement and Protection Research Unit in Salinas, California.

They found that under warm temperature and on wet leaves, E. coli O157:H7 multiplied 1,000-fold more in downy mildew lesions than on healthy lettuce leaf tissue. Even on dry lettuce leaves, where most bacteria struggle to survive, E. coli O157:H7 persisted in greater numbers when downy mildew disease was present.

The researchers also found that E. coli O157:H7 did not grow as well in downy mildew lesions on the lettuce line RH08-0464, bred by Simko and a colleague to be less susceptible to the lettuce disease, as the bacteria did on Triple Threat, a commercial variety that is highly susceptible to downy mildew.

The exact factors that caused less growth of E. coli O157:H7 in the more resistant line still need to be carefully explored. But if a genetic hurdle to E. coli O157:H7 colonization could be bred into commercial lettuce varieties along with downy mildew resistance, it would add a new defensive line to contamination of lettuce, helping farmers to improve the microbial safety of their crop as well as control their number-one plant disease problem.

Salmonella sticks to older lettuce better

Salmonella can bind to the leaves of salad crops including lettuce and survive for commercially relevant periods. Previously studies have shown that younger leaves are more susceptible to colonization than older leaves and that colonization levels are dependent on both the bacterial serovar and the lettuce cultivar.

lettuceIn this study, we investigated the ability of two Lactuca sativa cultivars (Saladin and Iceberg) and an accession of wild lettuce (L. serriola) to support attachment of Salmonella enterica serovar Senftenberg, to the 1st and 5–6th true-leaves and the associations between cultivar-dependent variation in plant leaf surface characteristics and bacterial attachment. Attachment levels were higher on older leaves than on the younger ones and these differences were associated with leaf vein and stomatal densities, leaf surface hydrophobicity and leaf surface soluble protein concentrations. Vein density and leaf surface hydrophobicity were also associated with cultivar-specific differences in Salmonella attachment, although the latter was only observed in the older leaves and was also associated with level of epicuticular wax.

 Older leaves of lettuce (Lactuca spp.) support higher levels of Salmonella enterica ser. Senftenberg attachment and show greater variation between plant accessions than do younger leaves.

Microbiology Letters [ahead of print]

Paul J. Hunter , Robert K. Shaw , Cedric N. Berger , Gad Frankel , David Pink , Paul Hand

Microbiological risk? Lettuce from British Columbia

As I sit and overlook Vancouver Harbour, it’s timely that hockey buddy Kevin Allen had a paper published about lettuce: because who doesn’t think of locally grown lettuce in Jan.

lettuceIncreased consumer demand for fresh leafy produce has been paralleled by an increase in outbreaks and illness associated with these foods.

Presently, data on the microbiological quality and safety of produce harvested in the Lower Mainland of British Columbia is lacking. Therefore, fresh green, red, and romaine lettuce samples (n = 68) were obtained from five regional farmers’ markets in late summer of 2012 and subsequently analyzed to determine total numbers of aerobic bacteria, coliforms, and Escherichia coli. Additionally, enrichment procedures were used to detect low concentrations of E. coli.

Obtained E. coli isolates were subjected to multiplex PCRs to determine phylogenetic groupings and the presence of virulence genes (eaeA, hlyA, stx 1, and stx 2). All E. coli were tested for resistance to 15 antibiotics using a disk diffusion assay.

Lettuce samples yielded mean aerobic colony counts of 6.3 log CFU/g. Coliforms were detected in 72% of samples, with a median concentration of 1.9 log CFU/g. Of samples, 13% were found to harbor E. coli, with a median level of 0.7 log CFU/g. Antibiogram typing of all E. coli (n = 33) revealed that 97% possessed resistance to one or more antimicrobials, with resistance to amikacin (58%), trimethoprim (48%), and trimethoprim-sulfamethoxazole (45%) being the most common.

Phylogroup typing showed that 79% of these isolates belonged to group B1, with the remaining assigned to groups A (9%) or D (12%); no virulence genes were detected. Considering that phylogroup indicators suggestive of fecal contamination (groups A and D E. coli) were recovered in lettuce samples presented at retail, further work is required to explore at what point along the food chain contamination occurs.

Also, this study shows the presence of multidrug-resistant E. coli in fresh vegetables. Summed, these data provide important information on the microbiological quality of leafy vegetables grown in British Columbia through the detection and characterization of frequently used indicator organisms.


Microbiological survey of locally grown lettuce sold at farmers’ markets in Vancouver, British Columbia


Journal of Food Protection®, Number 1, January 2015, pp. 4-234, pp. 203-208(6)

Wood, Jayde L.; Chen, Jessica C.; Friesen, Elsie; Delaquis, Pascal; Allen, Kevin J.

Biggest PR screw-up in NZ for 2014? Bad lettuce

The handling of a food poisoning scare involving carrots and lettuce has been deemed the biggest public relations challenge this year by a Wellington PR firm.

lettuce.skull.e.coli.O145The handling of the Yersinia pseudotuberculosis issue by the Ministry for Primary Industries beat the closure of regional flight routes by Air New Zealand and Roger Sutton’s resignation by the State Services Commission to make the top of the list.

“In a year of dirty politics, what really concerned New Zealanders most was dirty lettuce and carrots,” BlacklandPR director Mark Blackham said.

“Everyone had these vegetables in our fridges, yet no one in authority could say for some time whether they were a health threat.

Millions of people were affected and little information is a recipe for fear, rumours and anger.”

Epidemiology, just trying to do this jigsaw puzzle: Outbreak of E. coli O157:H7 associated with lettuce served at fast food chains in the Maritimes and Ontario, Canada, Dec 2012

Background: Identification and control of multi-jurisdictional foodborne illness outbreaks can be complex because of their multidisciplinary nature and the number of investigative partners involved.

spongebob.oil.colbert.may3.10Objective: To describe the multi-jurisdictional outbreak response to an E. coli O157:H7 outbreak in Canada that highlights the importance of early notification and collaboration and the value of centralized interviewing.

Methods: Investigators from local, provincial and federal jurisdictions, using a national outbreak response protocol to clarify roles and responsibilities and facilitate collaboration, conducted a rapid investigation that included centralized re-interview of cases, descriptive methods, binomial probability, and traceback findings to identify the source of the outbreak.

Results: There were 31 laboratory confirmed cases identified in New Brunswick, Nova Scotia, and Ontario. Thirteen cases (42%) were hospitalized and one case (3%) developed hemolytic uremic syndrome; there were no deaths. Due to early notification a coordinated investigation was initiated before laboratory subtyping was available. Re-interview of cases identified 10 cases who had not initially reported exposure to the source of the outbreak. Less than one week after the Outbreak Investigation Coordinating Committee was formed, consumption of shredded lettuce from a fast food chain was identified as the likely source of the illnesses and the implicated importer/processor initiated a precautionary recall the same day.

Conclusion: This outbreak investigation highlights the importance of early notification, prompt re-interviewing and collaboration to rapidly identify the source of an outbreak.

Canada Communicable Disease Report CCDR

Tataryn J, Morton V, Cutler J, McDonald L, Whitfield Y, Billard B, Gad RR and Hexemer A

Tracking an Escherichia coli O157:H7–contaminated batch of leafy greens through a pilot-scale fresh-cut processing line

Cross-contamination of fresh-cut leafy greens with residual Escherichia coli O157:H7–contaminated product during commercial processing was likely a contributing factor in several recent multistate outbreaks.

lettuceConsequently, radicchio was used as a visual marker to track the spread of the contaminated product to iceberg lettuce in a pilot-scale processing line that included a commercial shredder, step conveyor, flume tank, shaker table, and centrifugal dryer. Uninoculated iceberg lettuce (45 kg) was processed, followed by 9.1 kg of radicchio (dip inoculated to contain a four-strain, green fluorescent protein–labeled nontoxigenic E. coli O157:H7 cocktail at 106 CFU/g) and 907 kg (2,000 lb) of uninoculated iceberg lettuce. After collecting the lettuce and radicchio in about 40 bags (∼22.7 kg per bag) along with water and equipment surface samples, all visible shreds of radicchio were retrieved from the bags of shredded product, the equipment, and the floor. E. coli O157:H7 populations were quantified in the lettuce, water, and equipment samples by direct plating with or without prior membrane filtration on Trypticase soy agar containing 0.6% yeast extract and 100 ppm of ampicillin. Based on triplicate experiments, the weight of radicchio in the shredded lettuce averaged 614.9 g (93.6%), 6.9 g (1.3%), 5.0 g (0.8%), and 2.8 g (0.5%) for bags 1 to 10, 11 to 20, 21 to 30, and 31 to 40, respectively, with mean E. coli O157:H7 populations of 1.7, 1.2, 1.1, and 1.1 log CFU/g in radicchio-free lettuce. After processing, more radicchio remained on the conveyor (9.8 g; P < 0.05), compared with the shredder (8.3 g), flume tank (3.5 g), and shaker table (0.1 g), with similar E. coli O157:H7 populations (P > 0.05) recovered from all equipment surfaces after processing.

These findings clearly demonstrate both the potential for the continuous spread of contaminated lettuce to multiple batches of product during processing and the need for improved equipment designs that minimize the buildup of residual product during processing.

Journal of Food Protection®, Number 9, September 2014, pp. 1448-1648, pp. 1487-1494(8)

Buchholz, Annemarie L.1; Davidson, Gordon R.1; Marks, Bradley P.2; Todd, Ewen C.D.3; Ryser, Elliot T.

Barriers to trace-back in a salad-associated EHEC outbreak, Sweden, June 2013

In June-July 2013, six counties notified the Swedish Institute for Communicable Disease Control of enterohaemorrhagic E.coli (EHEC) infections among attendees at a hotel in Dalarna, Sweden. An outbreak control team investigated to identify the source and implement control measures.

lettuce.skull.e.coli.O145We included individuals who attended the hotel between June 19th-25th in a cohort. We asked them about animal contact, swimming, and consumption of food items during this time using a questionnaire. A confirmed case was an EHEC O157:H7 outbreak strain positive individual who developed abdominal pain or diarrhoea between June 20th-July 2nd. We described the outbreak in time, place and person, calculated risk ratios (RR) and 95% confidence intervals (CI). We investigated the kitchen, tested and traced back implicated food items.

172 individuals responded. We identified 19 confirmed cases (Median age: 17 years, 64% female) with symptom onset between June 22nd-27th. Eating green salad on June 20th was associated with illness (RR:3.7;CI:1.3–11). The kitchen mixed green salads without records and destroyed leftovers immediately. Hence we could not conduct trace-back or obtain microbiological confirmation.

Green salad contaminated before entering the kitchen was the likely outbreak source. We recommended early collaboration with food agencies and better restaurant records to facilitate future investigations.

PLOS Currents Outbreaks

Michael Edelstein, Camilla Sundborger, Maria-Pia Hergens, Sofie Ivarsson, Rikard Dryselius, Mona Insulander, Cecilia Jernberg, Yvan Hutin, Anders Wallensten

Leafy green cone of silence: lawsuits dismissed against Tanimura & Antle

With different interpretations offered by opposing legal teams, two E. coli-related lawsuits filed by Seattle lawyer Bill Marler against Salinas-based Tanimura & Antle have been dismissed.

lettuce.skull.noroMarler told The Packer both cases had merit and were dismissed for reasons not related to the legitimacy of the cases.

Wesley Van Camp, vice president legal and general counsel for Tanimura & Antle, said the cases illustrated the importance of putting up a rigorous legal fight if there is no clear-cut connection between foodborne illnesses and specific fresh produce. She said the government must investigate to find a link between foodborne illness and specific produce items.’

In Pueblo, Colo., Tanimura & Antle had been named as a third-party defendant in an E. coli food contamination lawsuit, Liebnow vs. Boston Enterprises. The trial, which had been set for June 3, ended when the plaintiff dismissed her lawsuit against Tanimura & Antle, according to a release from the company.

Tanimura & Antle’s legal counsel was present at a mediation of the Colorado case and insisted upon dismissal of all claims against the company and defendants with prejudice, Van Camp said. The company was not a party to the settlement and no confidentiality provisions applied to Tanimura & Antle, she said.

Tanimura & Antle said in the release that a report filed by David Acheson, former associate commissioner for foods at the Food and Drug Administration, was pivotal in the case. Though a state epidemiologist had earlier named Tanimura & Antle as the lettuce supplier in the case, Acheson said in his lengthy deposition that public health officials did not conduct a full traceback of romaine lettuce and therefore could not rule out other sources of exposure. Acheson said FDA officials indicated the agency did not believe there was a compelling case to undertake the traceback on romaine lettuce.

Marler said May 3 that the Colorado case was ultimately dismissed by agreement of both parties through mediation. He said the law firm presented expert testimony that refuted Acheson’s conclusions. Timothy Jones, medical doctor and expert witness for Marler’s plaintiff, said in a report that the 2009 E. coli outbreak that caused ten cases of illness — including then 10-year old Emily Liebnow of Colorado Pueblo, Colo. Victims of the outbreak were identified in six states, including Colorado, Connecticut, Iowa, Minnesota, Missouri and North Carolina, he said. Jones said that evidence makes it “extremely unlikely” that anything other than a widely distributed food product was the source of multiple cases of illnesses caused by the E. coli O157:H7 strain.

spongebob.oil_.colbert.may3_.10-300x234“It is far more likely than not that Emily Liebnow’s infection resulted from consumption of contaminated Romaine lettuce from Tanimura and Antle, served at Giacomo’s restaurant on Sept. 6, 2009,” Jones said in this report.

In California, Marler also filed a motion to dismiss the lawsuit brought by Oroville Bernacki on behalf of his deceased wife Gail Bernacki in the U.S. District Court in Northern California against Tanimura & Antle. Tanimura & Antle’s attorney Gregory Rockwell said the plaintiff would not be able to prove that the E. coli infection resulted from the consumption of the defendant’s product.

Disputing that point, Marler said that the evidence presented in the lawsuit showed a genetic match between the woman’s illness and Tanimura & Antle product in the same time frame. Marler said the case had merit but was dismissed because the elderly plaintiff, a Canadian resident, was in declining health and was not mentally or physically up to litigation.