Laboratory testing, PFGE, and epidemiologic investigations

Health care providers may in some instances order testing of an ill person’s blood or stool to help determine the cause of illness. In many circumstances a positive result in such a test must be reported to the health authorities pursuant to statute or regulation. Many states require reporting of positive tests for a number of pathogens, including E. coli O157:H7, Salmonella, Shigella, Listeria, hepatitis A, Campylobacter, and others. It is the report of such positive results that often triggers health department investigations of outbreaks.

Perhaps the most important scientific advance in tracing food poisoning is in pulsed field gel electrophoresis (PFGE), or DNA “fingerprinting.”
When a sample of bacteria, such as E. coli or Salmonella, is taken from a stool sample or a meat product, it can be cultured to obtain and identify the bacterial isolate. Bacterial isolates can be further broken down into their various component parts, creating a DNA “fingerprint.” PFGE operates by causing alternating electric fields to run the DNA through a flat gel matrix of agarose, a polysaccharide obtained from agar. The pattern of bands of the DNA fragments – or “fingerprints”- in the gel after the exposure to the electrical current is unique for each strain and sub-type of bacteria. By performing this procedure, scientists can identify hundreds of strains of E. coli as well as other pathogenic bacteria. The pattern of bacteria isolated from contaminated food can be compared and matched to the PFGE pattern isolated from infected persons.

When PFGE patterns match, they, along with solid epidemiological work, provide reliable evidence that the contaminated product was the likely source of the person’s illness. This is particularly true where the PFGE pattern has not been reported elsewhere. For example, suppose two unrelated persons both test positive for a genetically identical, unique strain of E. coli O157:H7 in a given town within a matter of days. If the subsequent inquiry into these two illnesses reveals no other common exposures between the two people other than a hamburger from the same restaurant on the same day, it is nearly impossible to find a credible, alternate explanation for their illnesses.

In 1993, a large E. coli outbreak occurred in the western United States. Scientists at the CDC performed DNA "fingerprinting" by PFGE and determined that the strain of E. coli O157:H7 found in patients had the same pattern as the strain isolated from hamburger patties served at a large chain of regional fast food restaurants. Had this source been identified sooner, it might have prevented hundreds of illnesses. As a result, the CDC developed standardized PFGE methods and, in collaboration with the Association of Public Health Laboratories, created PulseNet so that scientists at public health laboratories throughout the country can rapidly compare the PFGE patterns of bacteria isolated from ill persons and determine whether they are similar, thus indicating an outbreak linked to exposure to a common source of bacteria. PulseNet is an early warning system for outbreaks of foodborne disease, a national network of public health laboratories that performs DNA “fingerprinting” on bacteria that may be foodborne. The network identifies and labels each pattern and permits rapid comparison of these patterns through an electronic database at the CDC to identify related strains. At present, PulseNet tracks four foodborne disease-causing bacteria: Campylobacter jejuni, E. coli O157:H7, Listeria monocytogenes, nontyphoidal Salmonella, Shigella, and Yersinia pestis.

With an isolated illness, the lack of a positive stool culture may be problematic for a claimant. In the context of most outbreaks, however, circumstantial evidence may compensate. One such example is where one member of a dining party does not get tested, and others do. If three of four persons who ate together fall ill with the same documented pathogen, and the fourth demonstrates the same symptoms in the same time frame, liability can easily be established without the positive stool culture. In addition, in food poisoning cases there frequently is no food to test because, not surprisingly, it was eaten or discarded. But if leftover food, or uncooked portions, tests positive for the given bacteria or virus, we have powerful evidence that the food is the likely cause of the illness.