5/18/2016

​CDC estimates that each year, one in six US residents becomes a victim of foodborne disease—which equates to around 48 million annual cases of foodborne illness. PulseNet is working hard to curb these statistics.

Lab Matters Spring 2016 Cover

by Nancy Maddox, MPH, writer


Google farm to table and you will get an astounding number of hits—more than 38 million. Eating local hasn’t been so hot since consumers literally had no choice. According to the management consulting firm A.T. Kearney, the US local food movement was worth an estimated $12 billion in 2014, with a robust outlook for growth.

But food, according to Forbes, is the world’s biggest industry. And $12 billion is chump change. The US Department of Agriculture (USDA) figures agriculture and related industries added $835 billion to the country’s gross domestic product in 2014. Meanwhile, the World Bank values the global food and agricultural sector at around $4.8 trillion. Let’s face it:  a lot of this food is on the move.

In the United States, more than 90% of seafood offered for sale comes from abroad, as well as half the fresh fruit on the market, 20% of fresh vegetables and 4% of meat and poultry products.

Hungry consumers assume the carefully displayed rows of pineapple, jars of exotic spices and fresh ahi tuna steaks are safe to eat, but sometimes they aren’t. Following food from farm to table is a lot harder when the farm is in Croatia and the table is in Colorado Springs.

Who polices the vital yet vast and sprawling international web of producers, exporters, importers and retailers? The answer is as complex as the question, involving food regulators, food inspectors, health authorities and, of course, laboratory scientists.

Let’s begin at the table and work our way backward.

PulseNet—The Last Line of Defense

CDC estimates that every year, one in six US residents becomes ill after ingesting tainted food. That equates to about 48 million annual cases of foodborne illness, of which about 128,000 require hospitalization and 3,000 result in death.

A particularly severe 1993 outbreak of E. coli O157:H7—traced to beef sold at Jack in the Box restaurants in Washington, California, Idaho and Nevada—killed four children, permanently injured 178 people and sickened hundreds of others, mostly under age 10. This tragedy proved a turning point:  just three years later, CDC and APHL co-founded PulseNet, the national molecular subtyping network for foodborne disease surveillance. (See related articles on Pages 4 and 12.)

PulseNet is the last line of defense against foodborne disease. The 83-laboratory network generates “DNA fingerprints” of bacteria associated with foodborne illness, so outbreak responders can link cases of illness, identify their common source and take action to eliminate the source. To a lesser extent, PulseNet also stores fingerprints of bacteria isolated directly from food, animals and the environment.

The network has proved enormously effective. According to CDC, investigations of PulseNet-identified outbreaks are responsible for the recall of more than a billion pounds of contaminated food since its inception. The enhanced accountability created by PulseNet has also spearheaded changes in the food industry to prevent food contamination in the first place.

Craig Hedberg, PhD, a professor and food safety expert at the University of Minnesota, credits PulseNet’s success to three factors: (1) the use of a standardized pulsed-field gel electrophoresis (PFGE) subtyping method that enables “a profound level of [bacterial] comparison that didn’t exist before,” (2) the digitization of PulseNet data, enabling “the mass movement of data among partners,” and (3) dissemination of the methods and equipment “to make [the network] work.” Hedberg said, “These distinct evolutions in laboratory methods and technology could then be replicated by other countries around the world, creating the framework for a truly international surveillance system.”

The first international PulseNet partner was Canada, which created its laboratory-based network in 2000 and immediately entered into a relationship with PulseNet USA. “It was the logical thing to do because we have the same food market,” said Peter Gerner-Smidt, MD, DSc, who heads CDC’s Enteric Diseases Laboratory Branch, where PulseNet USA is based.

Said Gerner-Smidt, “We quickly got direct access to their database, and vice versa. And we can do direct [bacterial] comparisons, and this has been extremely helpful.”

Soon after the cross-border partnership began, PulseNet data helped to resolve a US-Canada outbreak of Salmonella Enteritidis phage type 30 traced to raw whole almonds. Just last year, US health officials worked with the Public Health Agency of Canada to investigate an outbreak of Listeria monocytogenes that affected individuals in Canada and nine US states. The source was traced to salad greens packaged at a Dole processing plant in Springfield, OH.

PulseNet’s success has driven its expansion. PulseNet Asia Pacific was formed as an independent network in 2002, PulseNet Latin America & Caribbean in 2003, PulseNet Europe in 2003-2004, PulseNet Middle East in 2006 and PulseNet Africa in 2009.

Gerner-Smidt, who chairs the PulseNet International steering committee, said, “We are working with Mexico to see if we can create something similar to [the direct database access] we have with Canada. . . . We import a whole lot of food from Mexico.”

He said, “The biggest challenge is that the surveillance system doesn’t function the same way everywhere. Some places have access to technologies that other places don’t. Some places have other priorities.”

Currently, all the networks perform PFGE for at least one of the ten genera of microbes monitored by PulseNet USA: Salmonella, E. coli O157, non-O157 E. coli, Shigella, Shigella flexneri, Listeria, Campylobacter, Clostridium botulinum, Vibrio cholera and Vibrio parahaemolyticus. Some networks, including PulseNet USA and PulseNet Europe, also use multiple-locus variable number tandem repeat analysis for a few specific microbial serotypes that aren’t well characterized by PFGE. However, the next cutting-edge technology is unquestionably whole genome sequencing (WGS).

The “Next Big Thing”

“The whole world has been caught by the WGS craze and for good reason,” said Gerner-Smidt. “In one fell swoop,” he said, WGS can reveal a pathogen’s species, serotype, subtype, drug resistance markers and virulence factors. Even though the technology currently costs about three times as much as PFGE, the price is expected to fall. Moreover, “you get so much more than you get from PFGE, you need to put that into the calculation too,” said Gerner-Smidt.

Currently, the United States is the only country systematically uploading all microbial WGS data to a central database in more or less real time. The hope is that other nations will follow suit.

Because of its greater discriminatory power—detecting microbial differences at the base pair level—WGS enables a much more precise definition of outbreak strains. This level of detail, in turn, enables epidemiologists to more precisely separate outbreak cases from random, unrelated, background noise and hence to detect small outbreaks that might otherwise go unnoticed.

Since late 2013, CDC and the US Food and Drug Administration (FDA) have led a project to sequence all Listeria monocytogenes isolates collected from cases of human illness in the United States. “We’ve seen that in two years’ experience with [WGS for] Listeria, the number of [patient] clusters detected has increased 50%,” said Gerner-Smidt. “And the number of outbreaks we’ve been able to solve has more than doubled. And the median number of cases in the clusters that we’ve been investigating has also gone down.”

APHL, which co-manages PulseNet’s quality assurance program, is collaborating with CDC on a white paper describing the vision for implementing WGS across the various national and regional networks. “We want it to become available to as many networks as possible, as quickly as possible,” said Gerner-Smidt.

At the same time, Gerner-Smidt noted that CDC will maintain PFGE capability over the long term: “We need to make sure that countries that do not have capacity to do WGS don’t feel left out and stop participating. We know how to address that issue—countries moving to WGS will still need to do some PFGE for every outbreak they are investigating. We will keep that capacity at CDC for sure.”

Gerner-Smidt said, “APHL is a critical PulseNet partner, helping us and facilitating the work we are doing with all our partners; organizing meetings and trainings and implementing the methods that we use. . . . . We could not do what we are doing internationally without APHL.”

A 2012-2013 outbreak of Salmonella Montevideo and Salmonella Mbandaka showcases PulseNet’s ability to track microscopic pathogens across the globe. The outbreak began in New Zealand with more than a dozen cases of illness and was traced to imported Turkish tahini. The following year, identical strains of Salmonella showed up in the United States, with cases spread across the continent from California to Georgia. Investigators soon fingered the same brand of tainted tahini.

“There were actually several cases identified in Europe also,” said Gerner-Smidt, who noted that the World Health Organization eventually got involved to follow up with the Turkish manufacturer.

“The food aisle is full of billion-dollar products”

Ending outbreaks is critically important. But identifying contaminated foods before they reach the dinner table is even more valuable from a public health perspective. Thus, federal and state food safety laboratories routinely test retail foods taken from grocery shelves.

Steve Stich, division director of food safety inspection at the New York Department of Agriculture and Markets, said his agency initiates about 250 food recalls each year. Most of those, about 70%, are for imported foods. And, among those imports, the most common violations are unapproved food dyes, undeclared preservatives (e.g., sulfites), heavy metal adulteration and uneviscerated fish, a risk factor for botulinum toxin.

Since it is impossible to test more than a tiny fraction of the food sold in the US market, highly trained food inspectors fill an important role targeting high-risk items. “Of all the foods we sample, about 1 in 6 or 1 in 7 results in a recall. That’s pretty effective sampling, we think,” said Stich, who is immediate past president of the Association of Food and Drug Officials.

Along another busy US border, Matthew Schaab, PhD, FERN Chemistry manager at the Arizona Department of Health Services (ADHS) Laboratory, said it isn’t only legally imported food that poses a problem: individuals sometimes bring food into the country in their luggage to sell in ethnic markets or consume in their own households.

Last year, the ADHS laboratory tested several spice samples submitted by the state’s childhood lead poisoning prevention program during a home investigation for a lead-poisoned child. The spices had been purchased by the child’s family in India and brought home for culinary use. Analytical results revealed that the turmeric sample was contaminated with chromium and lead, the latter of which is linked to cognitive, renal and immune system impairment, especially in children.

In the wake of this finding, Schaab’s staff purchased dozens of spice samples locally, in Maricopa County, as part of the laboratory’s surveillance function and analyzed them using the Food Emergency Response Network metal screen. None had poisonous metal contamination.

Unlike PulseNet—which typically tests specimens from people who are actively ill—food testing laboratories are required to demonstrate pathogens’ viability by growing them in culture. Maria Ishida, PhD, who heads the NY Department of Agriculture and Markets’ Food Laboratory, said, “For us to have a regulatory action, we have to have the culture. . . . . We can have the [bacterial] DNA in the food, but it doesn’t mean it’s alive to cause illness.”

Or, as Schaab pointed out, “The food aisle is full of billion-dollar products, so you have to be right when you report something. You have to be able to stand behind your test results in court.”

“Our food is the safest in the world”

Moving one step closer to the farm, both the US Department of Agriculture (USDA) and FDA monitor imports before they ever reach the food aisle. The USDA inspects every single shipment of imported meats and poultry products (its regulatory focus) and tests about 4%, choosing items based on inspection results and the safety history of the food or country of origin.

FDA consistently samples between 1 and 2% of the far greater volume of other imported foods. In fiscal year 2014, for example, FDA inspectors physically examined 1.7% of food imports under its purview or 188,578 of roughly 11 million “import lines”—commodities within a shipment with identical packaging and contents. The agency screens all import lines with an electronic tool to help field inspectors home in on riskier items.

But even before the Japanese wagyu, Mexican limes, Chilean sea bass, Spanish olive oil and other fare reach US soil, federal authorities are active behind the scenes.

One of those authorities is Emilio Esteban, DVM, MBA, MPVM, PhD, executive associate for laboratory services at USDA’s Food Safety and Inspection Service (FSIS). Esteban said FSIS must approve every country that exports USDA-regulated foods to the US under its “equivalency” process: “The food safety system has to be equivalent [to the US system]. They don’t have to do everything the same way we do, but they do have to meet the same sanitation standards, and we have to approve the laboratory methods they use to document their effectiveness.”

The equivalency process includes evaluation of a country’s food safety regulations and an in-country review of key systems, such as microbiological testing programs and hazard analysis and critical control point systems—not quite on the farm, but close. Esteban said the most common problem FSIS encounters is pesticide residue that either exceeds US tolerances or contains chemicals banned in the US. Currently, 35 countries have equivalency status for at least one USDA-regulated food sector.

On the FDA side, the Systems Recognition Program functions in much the same way as the USDA equivalence process, and also includes an in-country assessment. The key difference is that systems recognition is not required for exporters to access the US market. Instead, it facilitates risk-based inspections and information-sharing.

The Food Safety Modernization Act, now five years old, requires importers to verify that foods produced abroad meet applicable US safety standards. Among other things, importers must evaluate the risk associated with a food and conduct supplier verification activities, such as on-site audits of the supplier’s facility or food sampling and testing.
The combination of import controls, targeted food inspection and testing, and aggressive foodborne disease surveillance have given the US a relatively safe food supply. Esteban, who has participated in food safety audits on six continents, averred, “I can truly say our food is the safest in the world.”

He also noted that the increasing ISO accreditation of APHL-member food laboratories, “may open a lot of possibilities for us to collaborate with them in ways we have not done before.”

Ultimately, however, the sheer volume of imported food—roughly 68.8 million metric tons in 2014, coming from more than 200 countries and territories—poses an immense challenge. So too does the continual emergence of new foodborne pathogens and other contaminants. When tainted food slips through the cracks to consumers’ plates, PulseNet remains the last line of defense to track down the unsavory items so they can be removed from commerce and experts can figure out what went wrong.

As Esteban noted, “Food is a live thing.” And it can be a long way from the farm to the table.

Top Ten US Food Import Sources in 2014

1. Canada
2. Mexico
3. China
4. India
5. France
6. Chile
7. Italy
8. Brazil
9. Australia
10. Indonesia

Source:  USDA Economic Research Service, www.ers.usda.gov/data-products/us-food-imports.aspx#52236.