April 29, 1998 - SPECIAL REPORT: Despite Industry Efforts, E. Coli O157:H7 Remains Serious Threat

by Dr. Simon Shane

[Editor's Note: The following report was written by Dr. Simon Shane, a professor in the Department of Epidemiology at Louisiana State University, Baton Rouge, and a regular contributor to Poultry magazine. It contains an extensive overview of the current scientific consensus on identification, management, and control of the E. coli O157:H7 pathogen.]

The impact of the Hudson Foods recall of E. coli O157:H7-contaminated ground beef in the fall of 1997 is still reverberating throughout the meat industry. Although several initiatives to regarding re-work and grinding procedures have been put into place to prevent a recurrence, industry and regulatory professionals are hampered by a lack of knowledge concerning the epidemiology of this deadly pathogen.

What do we know about E. coli O157:H7? We know the organism is widespread, that it is virulent in very small doses, and that infection -- especially among immuno-compromised patients or children -- can result in painful, costly, and often fatal illness.

That's the grim reality that continues to underline research on the pathogen. However, as more information is gained, the battle to control E. coli O157:H7 begins to look more promising.

Here is a review of current scientific knowledge.

First, we know that the organism is present in the rumen of commercial dairy cows, calves and feedlot steers. It is presumed that the bacteria enters the food chain during evisceration -- hence, the avoidance of fecal contamination and the introduction of surface sterilization. But at present, there is no evidence that infected cattle show bacteremia at the time of slaughter. Even if this situation were to occur, control at the primary processing level would be difficult without serologic assay of individual animals before slaughter. With ongoing industry consolidation, it is difficult to conduct trace-back studies to identify causal factors. In one outbreak, ground beef implicated in a food-borne E. coli O157:H7 infection originated from more than 500 animals derived in three countries plus six states. Five plants were associated with the implicated batch. The strains of E. coli O157:H7 that produce the shiga-like toxins can survive in an acidic environment and are tolerant to a pH as low as 2.0. These organisms can survive a wide range of temperatures, from 12 degrees C to 41 degrees C.

An important distinguishing characteristic is that these strains do not ferment sorbitol, a factor that is used in primary isolation and identification. The standard method to screen for E. coli O157:H7 in fecal specimens comprises culture on sorbitol MacConkey agar for 24 hours at 38 degrees C. A clear colony (i.e., non-sorbitol fermenting) E. coli, presumably O157:H7, is then subjected to serologic screening using a latex-conjugated antibody. This two-phase procedure lacks sensitivity and it is evident that a number of E. coli isolates are not identified.

New proprietary media are available, which enhance isolation and can differentiate E. coli O157 from Proteus, and E. hermanii. Commercial immuno-based diagnostic systems are available, including the Reveal(r) assay from Neogen and the EHEC-Tek assay produced by Organon Teknika. Both systems require enrichment incubation followed by ELISA detection.

A number of molecular epidemiologic techniques have been used by public health authorities to investigate E. coli O157:H7 outbreaks. These include plasmid typing, phage lambda restricted fragment length polymorphism and multilocus enzyme electrophoresis. Multiplex polymerase chain reaction systems are useful to detect E. coli O157:H7. Primers detect the genes responsible for production of the two toxins and the 60 MDa plasmid, which is characteristic of the strains implicated in food-borne outbreaks. Sensitivity of these assays ranges from as few as 1 to 1,000 organisms. How prevalent is E. coli O157:H7? Studies have shown that up to 20 percent of samples derived from calf barns in Canada during a 1993 study yielded toxigenic E. coli O157:H7. In Scotland, 13 percent of patients in an E. coli O157 outbreak there were associated with a farm, compared with 87 percent of the affected individuals who were urban residents. Although public concern for E. coli O157:H7 dates from the 1993 outbreaks among children who ate undercooked hamburgers at West Coast fastfood restaurants, it is evident that the infection was present in this country as early as 1975. Serological evidence from Holland suggests that the pathogen was present at low levels in the human population in that country in 1974, with England having identified isolates between 1979 and 1982. The incidence of outbreaks and the intensity of surveillance have increased sharply in all industrialized countries since the beginning of 1993. The Centers for Disease Control in Atlanta investigated approximately 35 outbreaks in 1994, involving almost 600 cases.

As with most food-borne infections, public health authorities generally become aware of E. coli outbreaks that are associated with large-scale catering, commercial or institutional kitchens. Cases involving households are often not diagnosed unless complications lead to hospitalization.

Four forms of E. coli O157:H7 infection are recognized in human patients:

* Gastroenteritis, characterized by diarrhea, is the most common and least-documented clinical manifestation, often remaining undiagnosed.

* Thrombotic thrombocytic purpura (TTP) occurs in both children and the elderly and is characterized by discoloration of the skin. * Hemorrhagic colitis is characteristic of E. coli O157:H7 infection and can result in hospitalization, especially in the immuno-suppressed and young patients. * Hemolytic uremic syndrome (HUS) is an infrequent but life-threatening sequel of E. coli O157:H7 infection, especially in children. HUS may cause renal failure and death. The shiga-like toxins produced by E. coli O157:H7 are comprised of a binding protein specific for host-cell receptors lining blood vessels and the kidney tissue. The active component of the toxin acts to inhibit needed protein synthesis.

A 1989 E. coli O157 infection in Scotland parallels U.S. episodes, where infants and young children were infected at the rate of 100 per 100,000 population, compared with a rate of less than 5 per 100,000 for the 20- to 40-year age group. From 1982 to 1990, 7 of 12 outbreaks in the United States investigated by CDC occurred in northern states. Of the average of 56 cases per outbreak, 18 percent were hospitalized and 4 percent developed HUS or TTP complications. Fully one-half of these patients died, despite intensive treatment.

Even when death is averted, hospitalization of a single case of HUS costs from $50,000 to $100,000 per patient, without taking into account future complications arising from lingering complications of kidney damage.

Investigation of routes of E. coli infection confirm that meat, contaminated water, non-pasteurized milk or juice, and vegetables are all sources of infection. Secondary infection occurs, especially in hospitals, childcare centers and nursing homes. Ground beef was implicated as a source of six out of the eight outbreaks investigated by the CDC during the early 1990's.

The infective dose of E. coli O157:H7 may range from as low as 200 to 700 organisms. Given levels of carcass contamination and the relative conversion factors, a quarter-pound hamburger patty may have as many as 1,000 infective doses. Hamburger implicated in an outbreak of E. coli infection in Connecticut in 1993 yielded 10,000 E. coli O157 organisms per patty, compared with the absence of any E. coli in control patties. In 69 outbreaks investigated by the CDC from 1982 to 1994 involving 2,300 individuals, ground beef was positively identified as the source of infection in one-half of the patients and one-third of the outbreaks. No vector was identified in almost one-third of the outbreaks. Cases that have been attributed to swimming in contaminated rivers or dams, or to consumption of unpasteurized milk, unwashed vegetables, and non-sterilized apple cider, share a common contamination source: bovine fecal material. This confirms the importance of the intestinal tract of ruminants as a reservoir of infection.

Prevention of meat-source infection relies on a coordinated approach to prevention at the pre-harvest, slaughter and processing stages. Good production practices include maintaining the health of herds, strict adherence to regulations prohibiting administration of specified antibiotics, and implementing adequate hygiene and decontamination.

Slaughter procedures concentrate on avoiding contamination with fecal or rumen material and the adoption of Hazard Analysis and Critical Control Point systems with appropriate screening for E. coli O157:H7. Further processing should be carried out with good manufacturing practices, incorporating HACCP systems and close and careful monitoring.

Refrigeration is essential to limit proliferation of Stx-producing E. coli and other food-borne pathogens. Ultimately, irradiation may be the principal control method since it is economically and technically effective in eliminating E. coli. Strict personal hygiene and screening of workers to establish freedom from food-borne infections is strongly recommended. Institutional and commercial kitchens have now adopted a policy of thorough cooking to eliminate E. coli. A minimum temperature of 155 degrees F at the core of a patty should be attained for at least 30 seconds to ensure destruction of E. coli and other bacterial non-sporeforming pathogens. Proper cooking is vital. A Connecticut study in 1993 showed that the attack rate among patients increased from 3 percent in respondents confirming that they had consumed "well-done" hamburgers to 43 percent in those who reported eating rare meat. Cross-contamination is possible between raw meat and non-cooked foods, including salads, through work surfaces, hands and implements. In a recent control study conducted in a western state, it was demonstrated that risk factors for infection among consumers included food preparation workers not washing their hands with soap and water after handling raw ground beef, not washing work surfaces with soap and water after contact with raw ground beef, and placing cooked meat onto unwashed plates contaminated with raw ground beef.

Studies in Europe have confirmed that other E. coli serotypes have the ability to produce shiga-like toxins. E. coli O113:H21, O26:H11 and the non-motile O111 and O145 serotypes have been recently implicated in food-borne outbreaks of hemorrhagic colitis.

Additional research is required on the epidemiology of Stx-producing E. coli in order to develop appropriate preventive recommendations. Given current knowledge, it is evident that control programs based on HACCP alone will not guarantee freedom from infection. Appropriate handling and preparation are necessary to reduce the probability of food-borne disease due to contaminated meat, dairy and vegetable products.

This article reprinted with permission from Meat Marketing & Technology.

Return to 1998 News Archives