1. A History of Listeriosis and Listeria Issues for Meat/Poultry Products in The U.S.A.
L. Victor Cook
Microbiology Division
Office of Public Health and Science
Food Safety and Inspection Service

2. Topics
Provide a timeline of historic listeriosis outbreaks in the U.S.
Discuss two U.S. Risk Assessments that defined relative risk for products and strategies for intervention.
Provide data from FSIS regulatory testing programs and other large-scale studies.
Discuss impact of recalls on adulterated product.

3. Comparison with Other Foodborne Pathogens in the U.S.
Bacteria cases/year deaths/year
L. monocytogenes 2,
Salmonella ,400,
E. coli O157:H ,
Foodborne disease in the US (Mead, 1999)

4. Confounding Factors for Listeriosis Investigations
Incubation period is highly variable and can exceed one month.
Limited vulnerable subpopulation requires large-scale distribution and extended time (weeks/months) for evidence of incipient outbreak to appear.
Most illness is sporadic and difficult to attribute to a source.

5. Mexican-Style Cheese Outbreak
First high profile foodborne listeriosis outbreak in U.S.
Jan-Jun 1985 – reported in MMWR, June 21, 1985
Mexican-style fresh soft cheese (e.g., queso fresco, cotija)
86 cases in southern California
29 deaths (8 neonatal, 13 stillbirths, 8 non-neonatal)
Serotype 4b was isolated from patients and products from various production dates.
Homemade Mexican-style cheese continues to be a problem- Winter outbreak in North Carolina- 12 illnesses.

6. First Listeriosis Case Associated with an FSIS-Regulated Product
December (MMWR, April 1989)
One immunocompromised Oklahoma woman with cancer died from listeriosis.
She had consumed turkey franks (i.e., frankfurters, hot dogs) from a Texas establishment.
Product was warmed in a microwave prior to consumption.
Serotype 1/2a isolated from patient, opened and unopened product.
Multiple products in patient’s refrigerator were cross-contaminated with the same strain.

7. The Development of PulseNet
1995-CDC worked with FSIS and four state health department laboratories to establish:
National network of state and federal clinical and regulatory laboratories to contribute PFGE patterns to central CDC database.
PFGE method of choice but required standardized protocols
First protocol (1996)- E. coli O157:H7 (XbaI and BlnI)
First big outbreak from raw ground beef detected by PulseNet in 1997
Salmonella- same protocol as E. coli O157:H7
PFGE Protocol successfully reduced from 4 to 2 days, which helped throughput.
Dec PulseNet testing of L. monocytogenes began just in time to help detect an ongoing outbreak. Enzymes of choice were ApaI and AscI.

8. Listeria monocytogenes PFGE patterns

9. Fall 1998 Outbreak
MMWR, Dec 25, PulseNet and epidemiology linked at least 40 illnesses, including four deaths, from 10 states (OH, NY, TN, MA, WV, MI, CT, OR, VT, GA)
Franks and deli meats from Michigan plant were implicated.
Construction events contributed to distributing L. monocytogenes in the plant environment.
Serotype 4b
Based on CDC testing of implicated products, relatively low MPN levels were found (about 100 CFU/g).

10. Detroit Free Press
Feb. 10, 1999

11. Multistate Outbreak- 2000
MMWR, Dec 22, PulseNet and epidemiology linked 29 listeriosis cases, including four deaths and three miscarriages/stillbirths, in 10 states since May, 2000 (NY, GA, CT, OH, MI, CA, PA, TN, UT, WI).
Deli turkey meat from Texas plant was implicated.
Implicated strain matched strain isolated from the same establishment in the 1980s.

12. The NFPA/FDA Retail Study
Purchased deli meat products at retail in two FoodNet catchment areas (Maryland and San Francisco Bay Area) used by The CDC for tracking foodborne illness trends.
Sampled intact products and meat/poultry sliced at deli counters.
31,705 samples collected and tested qualitatively over a month period (initiated in 2000). Lm-positive samples were reanalyzed for quantitative result.
Luncheon/deli meats, deli salads, bagged salads, smoked seafood, and three cheese categories.
Provided a wealth of data for updating the FDA/FSIS Risk Assessment.

13. NFPA/FDA Study Results
577 of 31,705 samples (1.8%) were Lm-positive (Gombas et al. 2003).
Seafood salads = highest prevalence (4.7%)
Luncheon/deli meats = 0.89% (82/9199)
Pre-packed product= 0.4% prevalence, deli-packed = 2.7%
42 at 10-1 CFU/g
20 at 100 CFU/g
10 at 101 CFU/g
2 at 102 CFU/g
7 at 103 CFU/g
1 at 104 CFU/g

14. Northeast U.S. Outbreak- 2002
MMWR, Oct 25, listeriosis cases, including seven deaths and three stillbirths/miscarriages, in eight states (PA, NY, NJ, DE, MD, CT, MA, and MI).
Turkey deli meat (deli sliced) was implicated.
Two establishments were possible sources:
One large PA plant had a variety of environmental L. monocytogenes strains from product contact and non-contact surfaces, including a pulsotype match to the outbreak strain associated with a drain, but not from product. However, this plant distributed nationally.
Product from second smaller PA establishment was Lm-positive and distributed only to the Northeast U.S. No environmental isolates were found from FSIS testing.

15. Listeria monocytogenes, 2002
MATCH
HUMAN
FSIS launches traceback on turkey deli meat
PRODUCT
ENVIRONMENT
PulseNet detects listeriosis cluster in Northeastern US
FSIS PFGE analysis of all samples, matches to outbreak strain are found
Epidemiological study associates illness with turkey deli meat
FSIS initiates recalls, 2 plants recall total >31 million pounds turkey deli meat
[Note: the PFGE images on this slide are merely examples to demonstrate the match – they are not the actual outbreak pattern.]
In the Fall of 2002, there was an outbreak of listeriosis in the northeastern United States. Our use of PFGE subtyping technology and our collaboration with PulseNet was critical to figuring out the source of this outbreak.
It all began when a cluster of illnesses was detected by public health authorities, and the L. monocytogenes isolated from every ill person shared the same uncommon PFGE pattern. An epidemiological investigation was launched (a case-control study), the results of which implicated consumption of turkey deli meat.
FSIS launched tracebacks of the most common brands of turkey deli meat from delis where the ill individuals had shopped, and found that there were four predominant processing plants supplying those delis. We immediately conducted intensified investigations, or food safety assessments, at those plants. Numerous samples from both products and the environments of these plants were taken for L. monocytogenes and PFGE analyses.
We submitted the PFGE patterns from positive samples, and discovered that one plant had the outbreak strain in their finished product, and we initiated a recall.
Additionally, we found a high number of environmental samples that were positive for L. monocytogenes in a second plant, one of which was the outbreak strain. Given the high level of contamination in this plant, the likelihood that its products had contained the outbreak strain was high. We initiated a recall in this plant as well. In total, over 31 million pounds of turkey deli meat was recalled.
Without the molecular subtyping network in place, this complicated investigation may not have been resolved.
[Full description of investigation below]
Late Aug 2002 PA health officials noticed rise in Lm cases
Early Sept PFGE identifies 7 = outbreak strain. Uncommon strain in PulseNet database.
No food isolate patterns in found to match.
FSIS verified that all PFGE were sent. Found FSIS lab had all in, and no delays in process.
CDC asked state health departments to immediately submit PFGE patterns to PulseNet
Case-control study launched
Asking people about foods they ate up to 4 weeks ago: recall bias
Turkey deli meat, crab, and shrimp of concern in earlier analyzes. FDA on point also.
Final analysis: Listeriosis statistically significant association with frequent consumption of turkey deli meat (turkey deli meat sliced at the deli counter).
What turkey deli meat? Case-control study didn’t answer that question.
Went to delis where ill individuals shopped
Inventory and product samples collected
Ready-to-eat USDA, FSIS-regulated products being sold
Products manufactured by 4 plants most common.
Products associated with illness? Tracebacks didn’t answer that question as the findings may have reflected market saturation? May delis had multiple brands of turkey deli meat for sale.
Focus on 4 plants, more commonly present in the delis where the ill individuals shopped
Food Safety Assessments: sent in a special, multi-disciplinary teams. 1 CDC accompanied us
regulatory compliance, observation of operations and sanitation practices, eval of micro data (plant and FSIS-generated), env swabbing, product testing
226 product samples analyzed
(collected at plants and at retail)
206 ennvironemntal swabs analyzed
PFGE analysis
Outbreak strain in product of one Plant A
Outbreak strain in environment of Plant B. This plant, however, had a high number of env positives and PFGE analysis revealed that env strains were found in their product. Suggesting the potential for other strains, such as the outbreak strain, to contaminate product
Both plants recalled product: number of pounds of product recalled we were not acting conservatively to maximize public health protection
Large recalls…Industry and Agency not acting conservatively…maximize protection of PH
FSIS Investigates 4 plants, products and environments sampled

16. The FDA/FSIS Risk Assessment 2003
23 food categories.
3 populations: perinatal, “elderly” (>60 years), intermediate (could not separate HIV, cancer and transplant patients).
Deli meats- highest risk per serving (77predicted cases/billion servings) and per annum (1599 predicted cases/year).
Non-reheated franks- “high” risk per serving (65 cases /billion) and per annum (30.5 cases/year).
Pates/meat spreads-”high” risk per serving (32 cases/ billion), “moderate” risk per annum (3.8 cases/year).
Reheated franks- “low” risk per serving (0.063 cases/billion) and per annum (0.4 cases/year).
Dry/semi-dry fermented sausages- “low” risk per serving (0.017 cases/billion) and per annum (<0.1 case/year).

17. The FSIS Listeria Risk Assessment 2003
Goal #1- Provide insight into relationship between Listeria spp. on food contact surfaces and L. monocytogenes on the product.
Conclusion- Food contact surfaces found positive for Listeria spp. greatly increased the likelihood of finding RTE product lots positive for L. monocytogenes.
Goal #2- Evaluate the effectiveness of environmental testing programs, pre- and post-packaging interventions, growth inhibitors, and combinations.
Increased frequency of food contact surface testing and sanitation leads to a proportionally lower risk of listeriosis.
Combinations of the above mitigation strategies are more effective than any single approach.
Frequency of contamination of food contact surfaces with Listeria spp. encompasses a broad timeframe, and, on average, contamination lasts about one week.

18. Prevalence of L. monocytogenes in RTE meat/poultry products for 1990 and 2000
Product type
Cooked/roast/corned beef 6.38% 2.24%
Sliced ham/lunch meats 7.69% 3.05%
Small cooked sausages 4.21% 1.26%
Large cooked sausages 5.32% 0.51%
Cooked poultry products 2.79% 1.24%
Salads/spreads/pates % 0.98%
Fermented sausages % 1.49%

19. FSIS RTE Testing
40 categories based on 10 physical descriptions (e.g., whole muscle, sliced, multicomponent, etc.) and 4 processing categories (e.g., fully cooked, heat treated/shelf stable, etc.):
2001 overall prevalence
1.32% (86 Lm+ from 6533 total samples)
2002 overall prevalence
1.03% (78 Lm+ from 7592 total samples)

20. The USDA ARS Frankfurter Study
Twelve frankfurter manufacturing plants in ten states prepared one custom lot of product each.
ARS tested first at five days after production, then periodically tested product stored at 4ºC and 10ºC.
Three plants had significant percentage of positive samples (i.e., 1.5%, 2.2% and 16%).
Recovery rates did not change over time.

21. Impact: Recalls Year Meat & Poultry Recalled (pounds) 2000 18 million
Referring to recalls related to Lm-contaminated RTE meat & poultry products in the US.

22. Impact: Recalls Year Listeria Recalls/All Recalls (%)
/126 (34%)
/96 (26%)
/75 (41%)
Referring to recalls related to Lm-contaminated RTE meat & poultry products in the US over ALL meat & poultry recalls

23. Thank you…
Any Questions?