Washing is futile: Leafy green cone of silence

Lettuce is overrated.

lettuce-skullThat’s my response to people who ask about the proportionally high rates of foodborne illness in lettuce and other leafy greens eaten raw.

I like spinach – in a lasagna or stir-fry – but not raw.

Raw is risky.

There’s a bunch of new findings on foodborne pathogens and leafy greens which are summarized below.

In the sphere of public conversation, it is notable the Leafy Greens Marketing Agreement, the group formed after the 2006 E. coli-in-spinach outbreak that killed four and sickened at least 200 in the U.S. – has been once again silent on any research or outbreaks that associate risk with greens.

The leafy green cone of silence.

Investigations by University of Leicester microbiologists have revealed that just a small amount of damage to salad leaves can massively stimulate the presence of the food Salmonella in ready-prepared salad leaves.  

The scientists have discovered that juices released from damaged leaves also had the effect of enhancing the virulence of the pathogen, potentially increasing its ability to cause infection in the consumer.  

The research is led by Dr Primrose Freestone of the University’s Department of Infection, Immunity and Inflammation and PhD student Giannis Koukkidis, who has been funded by a Biotechnology and Biological Sciences Research Council (BBSRC) i-case Studentship.  

Their research investigates novel methods of preventing food poisoning pathogens from attaching to the surface of salad leaves to help producers improve food safety for consumers.

This latest study, published in Applied and Environmental Microbiology, found that juices from damaged leaves in bagged spinach and mixed salad increased Salmonella pathogen growth 2400-fold over a control group and also enhanced their adherence to surfaces and overall virulence, or capacity to cause disease.  

Dr Freestone said: “Salad leaves are cut during harvesting and we found that even microliters of the juices (less than 1/200th of a teaspoon) which leach from the cut-ends of the leaves enabled Salmonella to grow in water, even when it was refrigerated. These juices also helped the Salmonella to attach itself to the salad leaves so strongly that vigorous washing could not remove the bacteria, and even enabled the pathogen to attach to the salad bag container.  

“This strongly emphasizes the need for salad leaf growers to maintain high food safety standards as even a few Salmonella cells in a salad bag at the time of purchase could become many thousands by the time a bag of salad leaves reaches its use by date, even if kept refrigerated. Even small traces of juices released from damaged leaves can make the pathogen grow better and become more able to cause disease.  

 “It also serves as a reminder to consume a bagged salad as soon as possible after it is opened. We found that once opened, the bacteria naturally present on the leaves also grew much faster even when kept cold in the fridge.  

“This research did not look for evidence of Salmonella in bagged salads. Instead, it examined how Salmonella grows on salad leaves when they are damaged.”  

Leafy green and other salad vegetables are an important part of a healthy diet, providing vitamins, minerals, and dietary fiber. Ready to eat prepared salads are particularly popular, are widely consumed and so of significant economic importance. Over recent years there has however been a number of outbreaks associated with fresh salad produce contaminated with Salmonella and E. coli both in the USA and Europe. 

lettuce-harvestThis has triggered considerable interest in effective strategies for controls and interventions measures both in UK industry, the EU and key research funding bodies. 

Despite a number of published reports on improving the microbiological safety of salad leaf production, very few studies have investigated the behavior of Salmonella once the leaves have been bagged.  

Giannis said: “Anything which enhances adherence of foodborne pathogens to leaf surfaces also increases their persistence and ability to resist removal, such as during salad washing procedures.  Even more worrying for those who might eat a Salmonella contaminated salad was the finding that proteins required for the virulence (capacity to cause infection) of the bacteria were increased when the Salmonella came into contact with the salad leaf juices. “Preventing enteric pathogen contamination of fresh salad produce would not only reassure consumers but will also benefit the economy due to fewer days lost through food poisoning. We are now working hard to find ways of preventing salad-based infections.”  

No comment from the LGMA.

While this research may make it seem like pre-packaged salads pose a scary risk, the researchers themselves were quick to say they still eat bagged salads. But they make sure to look for packages that have appropriate use-by dates and crisp-looking leaves. They stay away from salads that have mushy, slimy-looking greens, or bags with accumulated salad juice at the bottom. And they make sure to eat the greens within one day of purchase.

“Our project does not indicate any increased risk to eating leafy salads, but it does provide a better understanding of the factors contributing to food poisoning risks,” said Freestone. 

If you feel like it, you can wash greens that have already been pre-washed by manufacturers just before eating, but Freestone says this doesn’t have much of an effect on the salmonella bacteria that may already be attached or internalized by the leaves. 

Her study was published in the journal Applied and Environmental Microbiology.

No comment from LGMA.

Assessment of irrigation water quality and microbiological safety of leafy greens in different production systems

Journal of Food Safety, 2 November 2016, DOI: 10.1111/jfs.12324


Foodborne disease outbreaks associated with fresh produce irrigated with contaminated water are a constant threat to consumer health. In this study, the impact of irrigation water on product safety from different food production systems (commercial to small-scale faming and homestead gardens) was assessed.

spongebob-oil-colbert-may3-10Hygiene indicators (total coliforms, Escherichia coli), and selected foodborne pathogens (Salmonella spp., Listeria monocytogenes, and Escherichia coli O157:H7) of water and leafy green vegetables were analyzed. Microbiological parameters of all irrigation water (except borehole) exceeded maximum limits set by the Department of Water Affairs for safe irrigation water. Microbial parameters for leafy greens ranged from 2.94 to 4.31 log CFU/g (aerobic plate counts) and 1 to 5.27 log MPN/100g (total coliforms and E. coli). Salmonella and E. coli O157:H7 were not detected in all samples tested but L. monocytogenes was present in irrigation water (commercial and small-scale farm, and homestead gardens).

This study highlights the potential riskiness of using polluted water for crop production in different agricultural settings.

No comment from LGMA.

Adaptive response of Listeria monocytogenes to heat, salinity and low pH, after habituation on cherry tomatoes and lettuce leaves


Sofia V. Poimenidou, Danai-Natalia Chatzithoma, George-John Nychas, Panagiotis N. Skandamis


Pathogens found on fresh produce may encounter low temperatures, high acidity and limited nutrient availability. The aim of this study was to evaluate the effect of habituation of Listeria monocytogenes on cherry tomatoes or lettuce leaves on its subsequent response to inhibitory levels of acid, osmotic and heat stress.

Habituation was performed by inoculating lettuce coupons, whole cherry tomatoes or tryptic soy broth (TSB) with a three-strains composite of L. monocytogenes, which were further incubated at 5°C for 24 hours or 5 days. Additionally, cells grown overnight in TSB supplemented with 0.6% yeast extract (TSBYE) at 30°C were used as control cells. Following habituation, L. monocytogenes cells were harvested and exposed to: (i) pH 3.5 adjusted with lactic acid, acetic acid or hydrochloric acid (HCl), and pH 1.5 (HCl) for 6 h; (ii) 20% NaCl and (iii) 60°C for 150 s.

Results showed that tomato-habituated L. monocytogenes cells were more tolerant (P < 0.05) to acid or osmotic stress than those habituated on lettuce, and habituation on both foods resulted in more stress resistant cells than prior growth in TSB. On the contrary, the highest resistance to heat stress (P < 0.05) was exhibited by the lettuce-habituated L. monocytogenes cells followed by TSB-grown cells at 5°C for 24 h, whereas tomato-habituated cells were highly sensitized. Prolonged starvation on fresh produce (5 days vs. 24 h) increased resistance to osmotic and acid stress, but reduced thermotolerance, regardless of the pre-exposure environment (i.e., tomatoes, lettuce or TSB).

lettuce-woolies-sep_-12-300x225These results indicate that L. monocytogenes cells habituated on fresh produce at low temperatures might acquire resistance to subsequent antimicrobial treatments raising important food safety implications.

No comment from LGMA.

Efficacy of post-harvest rinsing and bleach disinfection of E. coli O157:H7 on spinach leaf surfaces

Food Microbiology, April 2017, vol 62, pg 212-220, http://dx.doi.org/10.1016/j.fm.2016.10.019

Attachment and detachment kinetics of Escherichia coli O157:H7 from baby spinach leaf epicuticle layers were investigated using a parallel plate flow chamber. Mass transfer rate coefficients were used to determine the impact of water chemistry and common bleach disinfection rinses on the removal and inactivation of the pathogen. Attachment mass transfer rate coefficients generally increased with ionic strength. Detachment mass transfer rate coefficients were nearly the same in KCl and AGW rinses; however, the detachment phase lasted longer in KCl than AGW (18 ± 4 min and 4 ± 2 min, respectively), indicating that the ions present during attachment play a significant role in the cells’ ability to remain attached. Specifically, increasing bleach rinse concentration by two orders of magnitude was found to increase the detachment mass transfer rate coefficient by 20 times (from 5.7 ± 0.7 × 10−11 m/s to 112.1 ± 26.8 × 10−11 m/s for 10 ppb and 1000 ppb, respectively), and up to 88 ± 4% of attached cells remained alive.

lettuceThe spinach leaf texture was incorporated within a COMSOL model of disinfectant concentration gradients, which revealed nearly 15% of the leaf surface is exposed to almost 1000 times lower concentration than the bulk rinse solution.

No comment from LGMA.

Development of growth and survival models for Salmonella and Listeria monocytogenes during non-isothermal time-temperature profiles in leafy greens

Food Control, Vol 71, Part A, Jan 2017, Pg 32-41



Leafy greens contaminated with Salmonella enterica have been linked to large number of illnesses in many countries in recent years. Listeria monocytogenes is also a pathogen of concern for leafy greens because of its prevalence in the growing and processing environment and its ability to grow at refrigeration temperatures. Experimental data for the growth and survival of S. enterica and L. monocytogenes under different conditions and storage temperatures were retrieved from published studies. Predictive models were developed using the three-phase linear model as a primary growth model and square-root model to calculate specific growth rate (ln CFU g−1 h−1) at different temperatures (°C). The square-root model for S. enterica was calculated as μ = (0.020(Temperature+0.57))2. The square-root model for L. monocytogenes was fitted as μ = (0.023(Temperature-0.60))2. The growth-survival model for S. enterica and growth model for L. monocytogenes were validated using several dynamic time-temperature profiles during the production and supply chain of leafy greens. The models from this study will be useful for future microbial risk assessments and predictions of behavior of S. enterica and L. monocytogenes in the leafy greens production and supply chain.

No comment from LGMA.


Is there a relation between the microscopic leaf morphology and the association of Salmonella and Escherichia coli O157:H7 with iceberg lettuce leaves?

Journal of Food Protection, Number 10, October 2016, pp. 1656-1662, pp. 1784-1788(5)

I Van der Linden, M  Eriksson, M Uyttendaele, F Devlieghere


To prevent contamination of fresh produce with enteric pathogens, more insight into mechanisms that may influence the association of these pathogens with fresh produce is needed.

lettuce-washIn this study, Escherichia coli O157:H7 and Salmonella were chosen as model pathogens, and fresh cut iceberg lettuce was chosen as a model fresh produce type. The morphological structure of iceberg lettuce leaves (stomatal density and length of cell margins per leaf area) was quantified by means of leaf peels and light microscopy of leaves at different stages of development (outer, middle, and inner leaves of the crop) on both leaf sides (abaxial and adxial) and in three leaf regions (top, center, and bottom). The morphology of the top region of the leaves was distinctly different from that of the center and base, with a significantly higher stomatal density (up to five times more stomata), different cell shape, and longer cell margins (two to three times longer). Morphological differences between the same regions of the leaves at different stages of development were smaller or nonsignificant. An attachment assay with two attenuated E. coli O157:H7 strains (84-24h11-GFP and BRMSID 188 GFP) and two Salmonella strains (serovars Thompson and Typhimurium) was performed on different regions of the middle leaves. Our results confirmed earlier reports that these pathogens have a higher affinity for the base of the lettuce leaf than the top. Differences of up to 2.12 log CFU/g were seen (E. coli O157:H7 86-24h11GFP). Intermediate attachment occurred in the central region.

The higher incidence of preferential bacterial attachment sites such as stomata and cell margins or grooves could not explain the differences observed in the association of the tested pathogens with different regions of iceberg lettuce leaves.

No comment from LGMA.

The N.Y Times reports the one place the one place the Salinas Valley’s bounty of antioxidants does not often appear is on the tables of the migrant workers who harvest it.

sponge-bob-handwashingMore than a third of the children in the Salinas City Elementary School District are homeless; overall diabetes rates are rising and projected to soar; and 85 percent of farmworkers in the valley are overweight or obese, partly because unhealthy food is less costly, said Marc B. Schenker, a professor at the University of California, Davis, who studies the health of farmworkers.

No comment from LGMA.

Sponge Bob wants answers.

LGMA silent on Listeria outbreak

About four times a day I’ll get a tweet from the Leafy Green Marketing Agreement – the folks who set themselves up after the spinach outbreak of E. coli in 2006 that killed four and sickened 200 – blowing themselves about how great they are, and how their products are so safe.

spongebob.oil.colbert.may3.10If you want that kind of PR, then you have to be to take the hits as well.

LGMA never talks about an outbreak linked to leafy greens (publicly).

To me, they’ve succeeded best at lowering the leafy greens cone of silence and intimidating public health types into delaying reports of outbreaks.

LGMA says essentially that epidemiology doesn’t matter, and product must be shown to have the same outbreak strain as someone who is sick.

lettuce.skull.e.coli.O145That happened with spinach in 2006, and it has happened again with Listeria in 2016 – 2 dead, 19 sick, Canada and the U.S., all linked to Dole pre-packed salads.

Sure, it was probably the plant in Ohio that processed the stuff that was the source of the Listeria (and when I think of Ohio, I think salad).

But where’s the tweet, LGMA?


‘Lab confirmation should never be a requisite to implicating food in an outbreak’ Epidemiology is getting hammered to protect interests

From Australia to Wisconsin to cucumbers, epidemiology is getting hammered with a bunch of people defending their turf by saying, there’s no definitive proof.

spongebob.oil.colbert.may3.10The hepatitis A outbreak in Australia linked to Chinese frozen berries that has now sickened 19 has prompted many an Aussie to proclaim, I’ll never use frozen berries, without acknowledging they live in a sub-tropical climate, not Canada. And that preservation technologies work, when used properly.

The raw milk dairy farmers in Pepin County, Wisconsin, are denying charges that their raw milk served at a potluck dinner last September sickened at least 39 students and coaches affiliated with the Durand High School football team with Campylobacter.

Information collected from 65 students, coaches and parents indicated that consumption of milk was the only exposure statistically associated with the illness, DHS spokeswoman Jennifer Miller said.

For instance, the report said that some of the 38 people sickened at the dinner did not eat chicken, that 32 drank unpasteurized milk, and that six others drank milk consumed from a store-bought jug that could have contained pasteurized or unpasteurized milk.

Also, DATCP staff collected cow manure specimens from the Reeds’ cows and genetic fingerprinting proved that the bacteria that caused the illness at the dinner was the same bacteria strain found on the Reeds’ farm, Miller said.

But it’s the Salmonella-in-cucumbers outbreak – publicly unknown until a week ago — that reveals how deep interests influence.

An 18-week Salmonella outbreak linked to fresh cucumbers in 2014 sickened at least 275 people across 29 U.S. states, and killed one man.

Several of us in the food safety world – and probably the cucumber-consuming world – were left wondering, why didn’t we know about this?

Here’s the explanation from Dr. Bob (Bob Whitaker, chief science officer at the U.S. Produce Marketing Association; if anyone calls me Dr. Doug, I glare, and say, that’s Dr. Evil to you; I didn’t spent seven years in Evil University to be Mr. Evil):

epidemiology.WATER PUMP3_Page_4.storyThe Centers for Disease Control (CDC) issued their Morbidity and Mortality Weekly Report (MMWR) today. Anything with “morbidity” and “mortality” in the title is probably not going to rank highly on your general reading list and this weekly report was fairly typical of the reports published by CDC; highly technical and detailed, very informative, but probably targeted to a fairly specific audience.

So mere mortals are too dumb to understand?

Unfortunately, the process of identifying the cause of an illness outbreak and then further identifying how the people became ill and where that product originated from involves, as a first step, time consuming reporting of patient sample microbiology from local communities to states and perhaps up to the federal level. Often the process of just collecting this type of information and recognizing that there may be an illness outbreak exceeds the shelf life of our perishable products. Once an outbreak has been recognized, the epidemiological traceback or the process of determining what people might have come in contact with to make them ill can begin.

Epidemiology involves patient surveys, case control studies and correlation coefficients and can be time consuming and may yield multiple potential contamination vehicles. Finally, once potential food vehicles are identified, a traceback can be conducted to learn where patients might have obtained or consumed a specific food and investigators can work back to where the food was ultimately grown, harvested, packed or processed. Once production locations are identified, investigators can take microbial samples in an attempt to match any strains from the production environment to those isolated clinically from patients or in some limited cases from the food itself. Often by the time all of these vitally necessary steps have been completed, the crop is long gone and direct proof for how the contamination occurred or the source of the contamination is impossible to ascertain. Such was the case described by this Salmonella outbreak related to cucumbers.

CDC also disclosed that they employed consultations with industry experts early on in their investigation to gather information about industry practices, crop production cycles in the suspected region and product distribution. This is a positive step and is the result of a great deal of effort by the industry and CDC to determine how to engage industry to better inform investigations.

More accurate and rapid epidemiological investigations will ultimately help our industry determine what went wrong when these unfortunate illness outbreaks occur and are assigned to a produce item. This will help us direct research efforts aimed at identifying mitigation steps that can reduce the risk of further occurrences and assist operators in building improved risk and science-based food safety programs.

So why wasn’t the public informed there were a bunch of sick people?

Sounds more like the produce cone of silence.

Kirk Smith, epidemiology supervisor for the Minnesota Department of Health, told the Washington Post four years ago it’s rare for scientists investigating foodborne illness outbreaks to test the exact food suspected of carrying pathogens. By the time symptoms occur and a foodborne illness is reported and confirmed, the product in question has likely been consumed or has exceeded its shelf-life and been thrown away.

Instead, scientists, like detectives, interview victims, collect data, analyze patterns and match food “fingerprints” to determine the likely source of an outbreak.

“The majority of outbreaks, we don’t have the food to test,” Smith said. “Laboratory confirmation of the food should never be a requisite to implicating a food item as the vehicle of an outbreak.

“Epidemiology is actually a much faster and more powerful tool than is laboratory confirmation.”

As we have written, often during an outbreak of foodborne illness there are health officials who have data indicating that there is a risk, prior to the public. During the lag period between the first public health signal and some release of public information there are decision makers who are weighing evidence with the impacts of going public.

There is no indication in the literature that consumers benefit from paternalistic protection decisions to guard against information overload.

Leafy greens cone of silence; Phoenix E. coli victim sues Trader Joe’s, salad maker

A Phoenix man is suing Trader Joe’s and Glass Onion Catering and Gourmet Foods, the manufacturer of ready-to-eat salads that were linked to an outbreak of a deadly strain of E. coli last year that sickened 33 people in four Western states.

spongebob.oil.colbert.may3.10Steven Rabinowitz was the only Arizona resident who became seriously ill after eating the salads, which were sold at Trader Joe’s stores in the West. Seven others who were sickened also have sued.

The outbreak has been traced back to romaine lettuce produced in a single field, grown directly across from a cattle operation in Modesto, Calif. The strain of E. coli linked to the outbreak is most commonly found in cattle manure.

State public-health investigators from California could not determine a root cause of the outbreak, but they believe contaminants from the cattle field may have blown on to the lettuce grown by Ratto Bros., according to the investigative report. The grower has improved and adopted new procedures to prevent future contamination, the report says.

The case highlights the responsibility of everyone involved in the farm-to-table continuum — including growers, manufacturers and sellers — because consumers have no control over how a pre-packaged product is made, said Bill Marler, a Seattle attorney representing Rabinowitz and the other victims in the outbreak.

“There’s no consumer intervention here at all. You’re not going to buy a pre-made salad, take it out and wash it, and put it back in again,” said Marler, a national expert who has represented food-poisoning victims for more than 20 years. “Frankly, that’s why we sued. … This really requires more responsibility on the part of the entity that’s selling you the product.”