C-EnterNet Annual Report 2005-2006

...National Integrated Enteric Pathogen Surveillance Program

Table of Contents

• Acknowledgements
• Executive Summary

1. Introduction
   1.1 Background
   1.2 Status of C-Enternet - July 2006
   1.3 Scope of Report
2. Human Case Summary
   2.1 Overview of Human Cases
   2.2 Outbreak-related Cases
   2.3 Travel-related Cases
   2.4 Endemic Cases
3. Campylobacter
   3.1 Human Cases
   3.2 Exposure Surveillance
   3.3 Source Attribution
4. Salmonella
   4.1 Human Cases
   4.2 Exposure Surveillance
   4.3 Source Attribution
5. Pathogenic E. coli
   5.1 Human Cases.....
   5.2 Exposure Surveillance
   5.3 Source Attribution
6. Yersinia
   6.1 Human Cases
   6.2 Exposure Surveillance
   6.3 Source Attribution
7. Listeria
   7.1 Human Cases
   7.2 Exposure Surveillance
   7.3 Source Attribution
8. Parasites
   8.1 Giardiasis
   8.2 Cryptosporidiosis
   8.3 Cyclosporiasis
   8.4 Amoebiasis
9. Getting Closer to Source Attribution
   9.1 Risks Related To Food, Animals And Water
   9.2 Episodic Activities To Inform Source Attribution In Year 1
   9.2.1 Food Consumption Survey
   9.2.2 Drinking Water Consumption Survey
   9.2.3 Private Well Water as a Risk
   9.2.4 Pathogen Enumeration Study on Retail Meat
   9.2.5 Pathogen Levels During Refrigerated Storage: A Short Study
   9.3 Source Attribution Activities
10. Moving Forward

• Appendix A: Profile of Sentinel Site 1 - The Region of Waterloo, Ontario
• Appendix B: Sampling Methodology
  
  • Exposure Source and Pathogen Focus
  • Sampling Timeline
  • Human Cases
  • Retail Food
  • Agriculture
    • Swine Sampling
    • Dairy Sampling
  • Water
• Appendix C: Questionnaire Results
• Abbreviations Used
• Glossary
• List of Figures
• List of Tables

---

Executive Summary

C-EnterNet is a multi-partner initiative facilitated by the Public Health Agency of Canada and funded by Agriculture and Agri-Food Canada through the Agricultural Policy Framework initiative. It is designed to support activities that will reduce the burden of enteric (gastrointestinal) disease, by comprehensive sentinel site surveillance implemented through local public health units. Its core objectives are:

• To detect changes in trends in human enteric disease and in levels of pathogen exposure from food, animal and water sources in a defined population;
• To conduct source attribution (determine the proportion of human cases that are due to exposure via water, food and animals); and
• To improve the analysis, interpretation and reporting of laboratory and epidemiological data for public health, water and agri-food purposes.

C-EnterNet is designed to cover 5 or 6 sentinel sites across Canada with continuous and episodic surveillance activities for each of four components: humans, food, water, and food animals. Most of the planned surveillance activities have been implemented in the program's first (pilot) sentinel site - the Region of Waterloo, Ontario (Sentinel Site 1) - over the course of 2005. More specifically, C-EnterNet was able to generate site-specific incidences for reportable human enteric diseases and risk factor data for endemic cases. In addition, C-EnterNet began to collect pathogen data from most of the a priori important sources of exposure within the sentinel site's geographical area, including untreated surface water, stored and fresh manure from dairy and swine farms, and retail raw meat. This inaugural report provides results of the first 12 months of operation, from mid-2005 to mid-2006.

A total of 417 cases of 9 bacterial and parasitic enteric diseases were reported to the local public health authorities: 68% (282) of the cases were related to endemic causes, 21% (88) to travel, and 11% (47) to outbreaks. The three most frequently reported diseases were salmonellosis, campylobacteriosis and giardiasis, accounting for more than 80% of the endemic and the travelrelated cases.

There were a total of 129 (26.7/100,000 person-years) reported cases of Campylobacter infection. Of these 129 cases, 19% (24) were travel-related and 81% (105) were classified as endemic (21.8/100,000 person-years). Most cases (69%) were reported between June and September 2005, and the majority (94%) were C. jejuni infections. For animal exposures, Campylobacter cases had higher exposure than the other cases for household pet contact with dogs (36.5%). Eating in a restaurant (37.8%) was high compared to the cases due to other pathogens. In parallel, Campylobacter was successfully isolated from swine (71% of pooled samples) and dairy farms (26% of pooled samples), raw chicken meat (37%) and untreated surface water (7%), but not from raw retail pork and beef. Swine and dairy cattle appeared to be reservoirs of Campylobacter, predominantly C. coli in the case of pigs. Raw chicken meat was a potential source of transmission of foodborne C. coli and C. jejuni in households. The level of contamination in chicken meat was usually low (< 1 MPN/g), but sporadically could be higher; thus, retail chicken may be an important source of both C. coli and C. jejuni infections in humans if food is insufficiently cooked or improperly handled during preparation. Untreated surface water cannot be ignored as a potential exposure route for Campylobacter, although further subtyping data will be needed to clarify this point.

There were a total of 151 (31.3/100,000 person-years) reported cases of salmonellosis. Of these 151 cases, 22.5% (34) were related to travel, 26.5% (40) to outbreaks and 51% (77) were classified as endemic cases (16/100,000 person-years). Most endemic cases (61%) were reported from June to September 2005. Twenty-one serotypes were identified and the top three were Typhimurium (15), Heidelberg (10) and Enteritidis (6), which comprised 53% of serotyped isolates. For most animal exposures (both on-farm and while visiting farm animal areas), Salmonella cases had lower exposure than that reported from all infections except for household pet contact (53.6%), which was slightly elevated, and pet cats (18.8%). While eating undercooked food was not frequently reported (9.0%), it is interesting to note that those cases cited involved eating chicken dishes and omelettes. Salmonella was isolated from the swine (36% of pooled samples) and dairy (13% of pooled samples) farms, raw pork (2%), chicken (27%), beef (1%) and untreated surface water (13%). The number and types of serotypes identified varied with the sample source, making it difficult to attribute specific sources to human infections.

There were a total of 31 (6.4/100,000 person-years) reported cases of E. coli O157:H7. Of these 31 cases, none were travel-related, 7 were outbreak-related and 24 were classified as endemic (5.0/100,000 person-years). For E. coli O157:H7 cases, the proportion of cases that indicated a private well as a main water source was over twice as high as cases due to other pathogens. Higher proportions were also reported for other potential types of exposure, including: had eaten meat from a butcher shop, had swum in a pool, lived on a farm or in a rural setting, had visited farm animal areas, had shopped at a butcher shop, had eaten undercooked food (barbecued roast beef and steak tartar), had eaten meat from a private kill and had drunk unpasteurized milk. A low proportion had contact with a household pet. For on-farm animal exposures, cases of E. coli O157:H7 reported higher exposure to cattle, poultry, horses and cats than cases of other diseases. Verotoxigenic E. coli (VTEC) was detected in one pork sample and in no chicken or beef samples. Three swine manure and one dairy manure sample tested positive for E. coli O157. However, further testing determined that these were not the H7 strain. Verotoxigenic E. coli was also detected by molecular analyses in untreated surface water samples throughout the year. Swine and dairy cattle may be a reservoir of non-pathogenic E. coli O157, whereas untreated surface water is a possible source of VTEC. In addition, raw meat was rarely contaminated with pathogenic E. coli.

Eleven people (2.3/100,000 person-years) were reported with Yersinia enterocolitica infection. All cases were classified as endemic; over one-half were reported between July and September 2005. Seventeen of the pork chops tested were contaminated with Yersinia. However, following further subtyping, all were determined to be of non-pathogenic strains. Of the 117 swine manure samples tested, Yersinia was detected in 6%. Presently, subtyping results are available for 3 of the positive samples; all are of pathogenic strains.

No human listeriosis cases were reported. In raw retail meats, 7%, 23% and 28% of the pork, beef and chicken samples, respectively, were contaminated. Of the swine and dairy manure samples, 61% and 55% respectively, were positive for this bacteria. Following subtyping of the manure isolates, a much smaller proportion was confirmed to be L. monocytogenes. Retail subtyping results are pending.

There were a total of 54 (11.2/100,000 person-years) reported cases of giardiasis. Of these 54 cases, 31.5% (17) were travel-related and 69% (37) were classified as endemic (7.7/100,000 person-years). The giardiasis cases reported higher exposures for: swimming in a lake (33.3%), swimming in a pool (26.7%), visiting a farm animal area (25.8%), and drinking untreated water (24.1%). Of the meat samples tested using microscopy techniques, Giardia was detected on only one pork sample. Of the dairy and swine manure samples, 54% and 50%, respectively, tested positive for Giardia. Initial subtyping work has indicated that zoonotic assemblages are present (results pending). Giardia was detected in 100% of the untreated surface water samples.

A total of 12 (2.5/100,000 person-years) cases of cryptosporidiosis were reported. Of these 12 cases, 2 were travel-related and 10 were classified as endemic (2.1/100,000 person-years). High proportions were reported for swimming and visiting a farm animal area. No Cryptosporidium was detected in the retail raw meat samples, although it was found in pooled swine manure samples and in pooled dairy manure samples. Further subtyping on 17 of the swine manure samples indicated that they were C. parvum, the bovine genotype, which is considered to be zoonotic. Cryptosporidium was detected in 30 of 32 of the untreated surface water samples collected on Sentinel Site 1 (Grand River) during the first surveillance year.

Three cases of cyclosporiasis were reported: 1 was travel-related and 2 were classified as endemic. As Cyclospora is not considered to be endemic to Canada, no active surveillance for this parasite was performed at the food, agriculture and water sources.

There were a total of 20 (4.1/100,000 person-years) reported cases of amoebiasis. Of these 20 cases, 7 were travel-related and 13 were classified as endemic (2.7/100,000 person-years). As most Entamoeba infections in Canada are related to travel, immigration or person-to-person transmission, it was not assessed in the various exposure sources (food, agriculture and water samples).

In addition to the routine surveillance for enteric pathogens in the agri-food and water components, a number of episodic activities were performed that have direct relevance to attributing the level of risk and exposure to pathogens.

A food consumption survey (n = 2332) with a focus on the food safety perspective was conducted. This survey provided baseline levels of food and water consumption and food handling in the general population, as well as a comparison for risk factors from the case questionnaires.

A quantitative evaluation of pathogen load with enumeration was conducted on retail meat samples. In general, the bacterial pathogen loads were found to be below detectable levels on the majority of positive samples (Campylobacter, Salmonella, Listeria and Yersinia). For instance, the majority of chicken samples (85%) were found to have Salmonella levels below the enumeration detection limit, although 2 samples (6%) were found to have high levels of Salmonella contamination (> 1100 MPN per gram). Thus, prevalence and concentration data are available for risk modelling.

A study was performed to quantify the influence of refrigerated storage on pathogen levels on raw chicken, for five-day and eight-day storage periods. Results exhibited no statistically significant differences for Salmonella and Campylobacter between the time of sample collection and the end of the two storage periods. Statistically significant differences were observed for Listeria, with lower counts at Day 0 than after storage, with the maximum mean difference being less than 0.6 log.

Additional work related to source attribution has been initiated. A multi-partner collaborative group is working on a conceptual framework for source attribution. Work has begun to analyze a large dataset of Canadian foodborne outbreaks for source attribution. Work to apply a quantitative method of source attribution to historical Canadian Salmonella data has been initiated. Additionally, linkages with other groups interested in source attribution and the burden of enteric diseases, both in Canada and internationally, have been established. Additional ways to analyze the C-EnterNet's rich data will be incorporated as the program matures (additional years, complete subtyping, etc.).

The results reported here will serve as a benchmark for the ongoing monitoring of trends in pathogen movement, behaviour, prevalence and impact in the first sentinel site and subsequent sites across Canada. As the surveillance system progresses, with the addition of more sentinel sites and the complete implementation of all the planned surveillance activities, more exhaustive and reliable information based on laboratory findings and epidemiological data will show trends in enteric disease occurrence and in exposure sources to inform and strengthen source attribution in Canada.

This report summarizes the findings from Canada's new integrated foodborne and waterborne disease surveillance system, which focuses on four major components. The results from this surveillance system show that the epidemiology of foodborne disease in a Canadian-based community is consistent with that observed for other developed countries; this is expected, given the increasing globalization of food distribution, technologies and practices. The full implementation of this surveillance system will enable Canada to monitor the effectiveness of interventions as they apply to our food and water systems, ensuring our ability to maintain Canada's safe food and water supply in the face of new and emerging challenges.