E. coli Case Update
This is the executive summary of a pending report by the NC Division of Public Health regarding this outbreak. The full report will be available in January 2005.
An outbreak of enterohemorrhagic (bloody diarrhea) illnesses caused by E. coli O157:H7 associated with visits to the North Carolina State Fair occurred in North Carolina during October and November, 2004. After receiving reports of hemolytic uremic syndrome or HUS, a severe complication of E. coli O157:H7 disease, among State Fair visitors in late October, the North Carolina Division of Public Health recognized the potential for a large outbreak with potentially hundreds of thousands of fair visitors exposed. We gathered resources from within state government as well as the Centers for Disease Control and Prevention (CDC) to investigate the outbreak. The Public Health Command Center was opened for command, control and communications management of a statewide outbreak. Early goals included rapidly collecting reports of patients to gauge the size of the outbreak, and identifying common links between patients.
We received reports of over 180 patients during initial case finding. Over half of all case-patients were 5 years old or younger; two thirds were younger than 18 years. We identified 15 children with HUS. HUS has a significant potential for mortality and can leave survivors with lasting kidney damage. HUS typically develops in approximately 8% of all E. coli O157:H7 cases. The average age of HUS cases in this outbreak was 3 years, with a range of 1 to 13 years. Many case-patients reported fair visits and contact to animals in petting zoos.
By early November, a downward trend in new cases provided some reassurance. Recognizing the ongoing risk for spread within households or in child care settings, and by food service workers, public health officials maintained vigilance and reinforced isolation and hygiene policies. While some secondary cases did occur, there was not a large secondary outbreak.
Partway through the investigation, we refined the case-definition to reflect recognition that most cases had attended the State Fair. Using a refined case definition which included State Fair attendance, we classified 108 reported cases as outbreak-related. To more clearly identify the nature of the exposures, we designed a case-control study to compare cases to other fair attendees who did not become ill. Investigators also conducted microbiological testing at various locations at the fairgrounds to see if E. coli O157:H7 bacteria were present.
Our leading hypothesis was that contact between animals and case-patients at petting zoos at the fair had led to infection. However, we were careful to avoid biasing the study toward petting zoo exposures. We designed the questionnaire to include questions about the fair’s 2 petting zoos as well as 8 other animal exhibit areas. We also included questions about foods and beverages such as undercooked hamburger or fresh apple cider which have well-known associations with E. coli O157:H7 outbreaks. Questions about other food and beverage items purchased at the fair were also included. We asked about household exposures that might have led to illness as well. Between November 14th and 21st interviewers enrolled 45 cases and 188 controls–persons chosen randomly from a list of persons who purchased tickets to the fair online who did not become ill after attending the fair. Interviewers selected controls to match the age distribution of case-patients.
Information from the questionnaires was coded and analyzed using SAS statistical software. Preliminary analysis showed that 4 animal exhibits at the State Fair were associated with illness. These exhibits included the Crossroads Farm Petting Zoo where direct contact with sheep and goats occurred, the Children’s Barnyard, the Graham Building, and the Poultry Tent. Many activities in the Crossroads Farm Petting Zoo were associated with illness in the preliminary analysis, while only a few activities were associated with illness in the other 3 sites. Among children younger than 3 who visited Crossroads Farm Petting Zoo, case-patients were seven and a half times more likely to have reported contact with manure. Children who became ill were over five times more likely to fall or sit on the ground than children who did not fall or sit on the ground in the petting zoo.
The Crossroads Farm Petting Zoo had implemented guidelines from a national group of public health veterinarians to encourage hand hygiene to protect visitors from illness. Signs and hand sanitizing stations were present. Yet the information from our interviews indicates that hand-hygiene did not protect against infection. Exposures from direct contact between petting zoo visitors and animals or manure might have already led to infection before hand-sanitizer use, or contamination may have occurred in areas that hand-hygiene may not have removed. A very low number of E. coli O157:H7 organisms can cause infection.
Lab results from case-patients and environmental samples supported the case-control study’s findings. Microbiological cultures from case-patients with E. coli O157:H7 were further tested by a DNA “fingerprinting” method called pulsed-field gel electrophoresis or PFGE. PFGE testing can give additional information about unique strains of E. coli O157:H7. Cultures from 33 case-patients had the same PFGE patterns. Samples obtained from 4 fairground areas grew E. coli O157:H7. These included the Crossroads Farm Petting Zoo site, the Graham Building, the Sheep and Goat exhibit, and the State Fair Ark. Investigators obtained more samples from the Crossroads Farm Petting Zoo site because a high number of the initial samples from this area were positive for E. coli O157:H7. Nineteen of the 30 samples obtained during this second sampling grew E. coli O157:H7. Most positive samples were from areas where people could have direct contact with sheep and goats. All 19 positive cultures from the Crossroads Farm Petting Zoo site had the same PFGE patterns seen from the 33 case-patients’ cultures.
Findings from clinical specimens and environmental lab samples, along with the case-control study findings, support that most case-patients in this outbreak became ill after exposure at the Crossroads Farm Petting Zoo Other case-patients likely became ill after exposures elsewhere at the fair.
In light of this investigation, we recommend restricting direct contact with animals, reducing fecal contamination, and reducing crowding in petting zoos in addition to existing recommendations to prevent future E. coli O157:H7 infections (see http://www.nasphv.org/83416/84501.html). These recommendations are particularly pertinent for young children and others with reduced immunity to infection. We believe these recommendations are prudent to implement to protect the health of North Carolinians who visit petting zoos and our state’s fairs.