From farm to food to outbreak

As dumpsters filled with bags of spinach last week, Americans were forced to forgo eating one of the most healthful green leafy vegetables in their diet.

Questions linger about the contamination source and farming and irrigation practices. But in this E. coli O157:H7 outbreak, at least word got out quickly.

Less than a week elapsed between Sept. 8, when Wisconsin notified the Centers for Disease Control and Prevention of an uptick in E. coli cases, and Sept. 14, when the Food and Drug Administration told stores, restaurants and consumers to chuck their spinach. “This is an example of how fast it can work,” says Dr. Patricia Griffin, acting chief of enteric diseases at the CDC.

It wasn’t that fast in 1993, when an E. coli outbreak traced to tainted meat served at Jack in the Box restaurants killed four children, sickened 700 people and cost the chain $160 million.

“It took 39 days from the first reported case to recognizing that there was an ongoing outbreak,” says Bala Swaminathan, acting senior advisor for laboratory science at CDC’s division of food-borne safety. “Quite a few people became ill before the public health system recognized there was a problem.”

The 1993 outbreak, caused by contaminated hamburger sold in California, Washington, Idaho and Nevada, marked a turning point in how quickly food-borne illnesses are tracked and contaminated products are removed from the marketplace.

Change was necessary because of the way food is distributed in the United States. In the old days of food poisoning, the source usually could be traced back to a local event: potato salad at a family picnic or bad chicken at a church supper. Today, with massive amounts of food going from farms to food processing centers and then mixing with food from other farms before heading to tables in all 50 states, outbreaks are far less likely to be local, and thus, far more difficult to notice.

The change that came out of the 1993 tragedy is called PulseNet, a partnership between the CDC and state health departments. Stool samples are sent to state public health departments by doctors. The health departments then use DNA analysis to subtype pathogenic bacteria present in the stools. The resulting DNA fingerprints can identify not only species but also strains of such bacteria as E. coli, salmonella, shigella, listeria and Campylobacter.

Then the information gets sent to a central CDC database. This can quickly let public health officials know if cases strewn far and wide across the country carry identical fingerprints.

Epidemiologists also weigh in, fanning out within affected states and using food questionnaires to find out what people had eaten before getting sick (in the current outbreak, identifying pre-washed, bagged spinach).

The questionnaire “looks like an exhaustive, ridiculous list, asking about everything under the sun,” says Dr. Jeffrey Griffiths, professor of public health and family medicine at Tufts University School of Medicine. That’s because many foods can be contaminated.

Meat, poultry and produce that grows close to the ground (such as spinach, lettuce and cilantro) have all been implicated in E. coli outbreaks. It was even found inside watermelons, Griffiths says, after sellers injected the fruit with water to make it heavier and increase its price.

The combination of on-the-ground sleuthing and DNA fingerprinting has helped in other outbreaks of E. coli. In 1996, cases in Connecticut and New York were linked to unpasteurized apple cider.

In 2002, an outbreak that sickened 28 people in six states was linked to hamburger, resulting in a nationwide recall of 18.6 million pounds of fresh and frozen ground beef.

But this is still not a perfect system, Swaminathan says. It typically takes three to four days for a patient’s symptoms to kick in, a day or two before the patient seeks medical care, then up to three days for results from a stool sample to reach the physician. And frustrating slowdowns can come from something as simple as how long it takes for positive results to make their way from a physician’s office to state labs and then into the PulseNet database.

In some states, reporting can happen the same day the physician gets lab results. In others, it can take up to a week. And all this time, Americans -- oblivious to any problem -- continue to eat the contaminated food.

“The technology is not the limiting factor,” Swaminathan says. “We can do the DNA fingerprinting in a day. It’s limited by the resources of the state public health departments.”

*

susan.brink@latimes.com