Food Survey Reports 1999-2000

Microbiological Quality of Meats other than Chicken

NOVEMBER 1999 APRIL 2000

Report prepared by Geoff Millard and Simon Rockliff

OBJECTIVE

  • Determine the microbiological status of raw retail meats other than raw retail chicken i.e. raw meat sold in the ACT;
  • Compare these results to those available from Australian and overseas sources and to a previous raw retail chicken (raw chicken) survey conducted in the ACT to illustrate the unacceptable microbiological quality of raw chicken.

BACKGROUND

Literature sources indicate that raw meat has lower rates of contamination with the organisms Escherichia coli, Thermophilic Campylobacter sp (Campylobacter), Salmonella sp and Listeria monocytogenes than raw chicken(3). This short survey was undertaken to confirm that the ACT was consistent with the literature.

  • In 1996 Meat and Livestock Australia (4) (MLA) implemented the Microbial Food Safety Key program, in line with the Meat Industry Strategic Plan (MISP) prepared by the Meat Industry Council for the period 1996 2000. One aim of the MISP was to introduce a comprehensive quality assurance system based on the Hazard Analysis of Critical Control Points (HACCP) to all levels of the meat industry. To measure the effectiveness of the MISP the industry undertook baseline studies on the microbiological status of carcass and carton meats in both 1993-4 and 1998. In February 2000 MLA (4) produced a report comparing these baseline studies. The report "was based on samples excised from carcass surfaces which had been chilled for at least 12 hours and from pieces of meat drilled from cartons of frozen manufacturing meat."

As recognition of this quality assurance system approach, the USA in 1999 accepted the Australian Quarantine Inspection Service (AQIS) Australian Meat Safety Enhancement Program as equivalent to the U.S. inspection system. (5)

STANDARDS

There are no Australia New Zealand Food Authority Food Standard Code Standards for this product.

SURVEY

During the period 08/11/1999 to 11/04/2001, 53 raw meat samples consisting of 18 Beef, 11 Pork, 9 Lamb/mutton, 9 Fish, 3 Veal, 2 Ox and 1 Prawn were collected from 13 ACT retail establishments. These samples were tested to determine the presence of Escherichia coli (E. coli), Listeria monocytogenes (L. monocytogenes), Salmonella sp. and Thermophilic Campylobacter sp.

RESULTS

The isolation rates for the different types of ACT meat are given in Table 1.

Table 1

 

Meat [n]

E. coli (%)

L. monocytogenes (%)

Salmonella (%)

Campylobacter (%)

Veal [3]

0 (0.0)

0 (0.0)

0 (0.0)

0 (0.0)

Prawns [1]

0 (0.0)

0 (0.0)

0 (0.0)

0 (0.0)

Fish [9]

0 (0.0)

1 (11.1)

0 (0.0)

0 (0.0)

Beef [18]

3 (16.6)

3 (16.6)

0 (0.0)

0 (0.0)

Pork [11]

1 (9.1)

3 (27.3)

0 (0.0)

0 (0.0)

Lamb/ mutton [9]

2 (22.2)

0 (0.0)

0 (0.0)

0 (0.0)

Ox [2]

2 (100)

0 (0.0)

0 (0.0)

0 (0.0)

Samples Positive

8 (15.1)

7 (13.2)

0 (0.0)

0 (0.0)

 

[n] = number of samples tested.

(%) = percentage of samples positive.

Table 2 compares the results of the ACT raw meat survey with those of the ACT raw chicken survey (3) and meat results from the literature. (1)(2)(4)

Table 2

 

Organism

Meat Survey % Pos

Raw chicken Survey
% Pos

Literature 
%Pos

E. coli

15.1

70.8

2.3 – 31.0

Listeria monocytogenes

13.2

36.0

2.0 – 47.9

Salmonella sp.

0.0

39.9

0.4 – 45.6

Campylobacter

0.0

20.6

0.0 - 5.1

 

Pos = positive result

DISCUSSION

While the survey sample numbers for some of the cuts and types of meat are statistically small we believe that the overall trends are valid but the actual percentage isolation or non-isolation rate may not be representative.

Escherichia coli

The presence of E. coli on the sample is an indicator of faecal contamination and its isolation rate gives some indication of the degree of contamination for the individual types of meat. E. coli was isolated from beef, lamb/mutton, pork and ox samples. The ACT study showed that 16.6% beef, 22.2% lamb/mutton. 9.1% pork and 100% ox samples carried E. coli. Only 2 ox samples were tested in the survey and the high isolation rate could be due to low number of samples. The overall isolation rate was 15.1% which is approximately one fifth the isolation rate for raw chicken and in the mid range reported in the literature (Table 2).

The overall E. coli isolation rate in 1998 (4) for beef carcasses and boneless meat was 10.3% and 5.3 % respectively and 29.1% and 24.5% respectively for sheep carcasses and boneless meat. The ACT isolation rate from retail beef samples is higher than beef carcasses while sheep carcasses and retail sheep samples tend to have similar isolation rates. It is difficult to determine from the survey the reason for the differing beef and sheep meat results.

Listeria Monocytogenes

Listeria Monocytogenes was isolated from fish, beef and pork samples at individual rates of 11.1%, 16.6 % and 27.3% respectively. The overall isolation rate was 13.2% which is less than half the isolation rate for raw chicken and in the lower third of the range reported in the literature (Table 2).

Listeria Monocytogenes analysis was not performed by the Meat and Livestock Australia Surveys.

Salmonella

Salmonella was not found in any of the survey samples, which is below the range reported in the literature (Table 2). The overall isolation rate for raw chicken was reported as 39.9%. The Microbiology of Australian Meat 1998 results indicate that, with an overall isolation rate for beef carcasses of 0.2% and 0.1% for sheep carcasses (4) this organism is not prevalent at slaughter.

Campylobacter

Thermophilic Campylobacter was also not found in any of the samples and is at the lowest end of the range reported in the literature (Table 2). Meanwhile the overall isolation rate for raw chicken was reported as 20.6%. In 1993-4 (4) the overall isolation rates for Thermophilic Campylobacter for beef and sheep carcasses were 0.26% and 1.3% respectively. Thermophilic Campylobacter tests were not undertaken in 1998.

CONCLUSION

This was only a small survey with low sample numbers in each of the food types but the results indicate that the original premise as stated in the BACKGROUND was correct. Raw meat does have a much lower contamination rate of E. coli, L. monocytogenes, Salmonella and Thermophilic Campylobacter than raw chicken meat. This lower contamination rate (in some cases zero) indicates a lower risk to the public from these undercooked, mishandled and cross-contaminated foods. The HACCP (4) approach introduced into the meat industry appears to be achieving a small but significant improvement in some microbiological quality criteria of beef and sheep carcasses. A similar approach in the poultry industry could result in a reduction of overall levels of contamination for raw chicken carcasses.

RECOMMENDATION

That a follow up survey be conducted in the future, to ascertain if the present situation has continued.

BIBLIOGRAPHY

Nationwide Microbiological Baseline Data Collection Program

Flemming Bagar Ed. 2000, DANMAP 99. Consumption of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, food and humans in Denmark. Danish Veterinary Laboratory.

Microbiological Quality of Raw Chickens, 1999-2000. ACT Health Protection Service.

Meat and Livestock Australia, The Microbiology of Australian Meat 1998, Feb. 2000

Australian Meat Safety Enhancement Program (MSEP) Approved

Microbiological Quality of Meats other than Chicken

Microbiological Status of Raw Chilled Chicken

JULY 1999 - AUGUST 2000

Report prepared by Geoff Millard and Simon Rockliff

OBJECTIVE

  • Determine the microbiological status of raw chilled retail chicken (raw chicken) in the ACT and compare the results with a previous ACT survey conducted in 1995-1996;
  • Determine the correlation between human Thermophilic Campylobacter strains and those isolated from raw chicken;
  • Establish the serotypes of Salmonella sp isolated from the raw chicken samples and compare to the serotypes isolated from the ACT population during the same time period;
  • Investigate any health implications of the findings;
  • Establish the organism prevalence in cuts of chicken and apply a statistical test to see if the organism prevalence differs by cut.

BACKGROUND

A Health Protection Service (HPS)(17) survey conducted in 1995-1996 into raw chicken determined that it is a food that consistently harbours three pathogens causing the majority of foodborne diseases notified in the ACT ie Thermophilic Campylobacter sp, Salmonella sp and Listeria monocytogenes. The incidence of Listeria monocytogenes in the ACT is very low. Raw chicken that is cooked is an increasingly consumed product throughout Australia. (23) Per capita, Australians eat approximately 17.3 kilos of chicken muscle meat each year (20), while Americans eat approximately 23 kilos a year. (1) Chickens have been implicated on epidemiological grounds in a number of human Thermophilic Campylobacter and Salmonella infections. (10)(11)(14)(21) This survey was set up to further investigate the possible links between raw chicken and human disease in the ACT.

The ACT isolation rates for all three pathogens are based on the reported number of foodborne disease notifications. In the ACT a foodborne disease notification occurs when a proven causative agent is identified from faeces or a normally sterile site by a pathology laboratory and is reported to the health department. Notification processes are passive reporting schemes, which produce significant under reporting ie underascertainment of the true level of infection. (2) Where appropriate the authors have attempted to determine the true incidence of the disease in the ACT and estimate its relative cost to the community. Diagram 1 illustrates the degree of under reporting by notification schemes.

Diagram 1

 

Burden of Illness Pyramid US FoodNet cases 1997 graph

Burden of Illness Pyramid US FoodNet cases 1997 (3)

There are many species of Campylobacter but only those capable of growth at 42C ie Thermophilic sp. are able to cause disease in humans. Almost all of the isolates in this survey belonged to the Thermophilic species Campylobacter jejuni. Professor Coloe and Dr Ben Fry of the Royal Melbourne Institute of Technology (RMIT), hypothesise that the Cla1 gene is highly conserved in chicken Campylobacter strains and can be used as a stable genetic marker for epidemiological purposes. The presence of this gene can then be used to differentiate Campylobacter strains of chicken origin from human strains. (15) Therefore, the incidence of human Campylobacteriosis being caused by chicken strains can be determined utilising the presence or absence of the Cla1 gene in chicken and human Thermophilic Campylobacter isolates. This aspect of the survey involved collaboration with RMIT with Thermophilic Campylobacter sp isolated from both raw chicken samples and human specimens collected from pathology laboratories being transported to RMIT for further analysis.

STANDARDS

There are no Australia New Zealand Food Authority Food Standards Code Standards for this product.

SURVEY

During the period 19/07/1999 to 28/08/2000 266 raw chicken samples were collected from 62 ACT retail establishments, most establishments were only visited once. These samples were tested to determine the presence of E. coli, Listeria monocytogenes (L. monocytogenes), Salmonella sp. and Thermophilic Campylobacter sp.

RESULTS

Overall Isolation rates

The overall isolation rates of the organisms for the two ACT surveys are given in Table 1. A complete tabulation of organism distribution by individual cuts for the 1999 - 2000 survey are given in Appendix 1.

Table 1

 

Organism

Number of samples positive (%) in 1999 – 2000 survey, n=266

Number of samples positive (%) in 1995-1996 survey, n=112

E. coli

188 (70.7)

110 (90.2)

Listeria monocytogenes

96 (36.0)

34 (27.9)

Salmonella sp.

109 (41.0)

49 (40.2)

Thermophilic Campylobacter

55 (20.6)

15 (12.3)

 

n= Number of samples

Salmonella Serotyping

Table 2 compares the notified Salmonella serotype results from clinical isolates for the 1999-2000 period and Salmonella serotypes isolated from raw chicken by the two ACT surveys. For all Salmonella isolation data see table at Appendix 2. The occurrence of common serotypes could indicate a possible link between Salmonella contamination of raw chicken and cases of human Salmonellosis.

Table 2

 

Serotype

Clinical 1999-2000

n=120

Chicken 1999-2000

n=100

Chicken 1995 – 6

n=51

No of Isolates

No of Isolates

No of Isolates

S typhimurium 135

8

6

3

S typhimurium 9

32*

2

6

S. typhimurium 64

1

5

1

S. typhimurium untypable

2

2

0

S typhimurium RDNC

2

1

0

S. Singapore

2

0

3

S. agona

1

0

3

 

  • n= Number of samples

*higher than normal due to an outbreak

Thermophilic Campylobacter and the Cla1 gene

Table 3 presents the number of clinical and raw chicken Thermophilic Campylobacter isolates that possess the Cla1 gene as a percent.

Table 3

 

Source of Isolate (number of results)

Percentage possessing Cla1 gene

Clinical (n=81)

62

Chicken (n=20)

90

 

The numbers of Thermophilic Campylobacter isolates sent off for Cla1 testing and number of results received back are given in Table 4.

Table 4

 

Source of isolate

Number

Results received back

Clinical isolates sent to RMIT

296

81

Chicken isolates sent to RMIT

55

20

Total

351

101

 

A number of the clinical investigations performed on ACT residents were undertaken outside of the ACT and were not included in this study. A number of isolates sent to RMIT failed to grow on subculture. This is a common feature of Thermophilic Campylobacter and the current failure rate is 26.2%. The authors are still waiting on approximately 150 results from RMIT. While less than one third of the Thermophilic Campylobacter isolates have been tested for the presence of the Cla1 gene, the ratio between the human and raw chicken isolates has not changed significantly with the addition of each new result and we expect this trend to continue.

DISCUSSION

General

During the period of the survey there were 120 Salmonella sp, 333 Thermophilic Campylobacter sp and one Listeria monocytogenes clinical notifications in the ACT. Listeria monocytogenes causes listeriosis, which is an uncommon but severe disease. The subsequent investigation into the single case of listeriosis determined that the disease was acquired outside of ACT but diagnosed and reported within the ACT.

Prevalence of pathogens in ACT raw chicken and comparison to literature sources.

The prevalence figures given in Table 1 would indicate that the pathogen levels in raw chicken in the ACT have varied over the two surveys. The prevalence of E. coli appears to have declined by some 20% since the 1995-1996 survey while both Listeria monocytogenes and Thermophilic Campylobacter have increased by some 8%. Some of the Thermophilic Campylobacter increase may have been due to increased sensitivity from changes in methodology but there is no way of quantifying the effects of the change in methodology. The prevalence of Salmonella in raw chicken was similar for both surveys. The results obtained for ACT raw chilled chicken can also be compared to the prevalence levels of pathogens in chicken found in the literature. See Table 5.

Table 5

 

Organism

Source 1

Source 2

ACT 1999 – 2000 survey

E. coli

54% Denmark (7)

99.6% USA(12)

70.7%

L. monocytogenes

60% UK

15% USA(12)

36%

Salmonella

15% Denmark (7)

20% USA(12)

41%

Campylobacter

34% Denmark(7)

40–88%USA(12)

20.6%

 

The ACT raw chicken has twice the prevalence level for Salmonella, when compared to the literature (Table 6). ACT raw chicken is however, in the mid-range for Listeria monocytogenes and E. coli and in the lower range for Thermophilic Campylobacter when compared to the literature sources.

Australia and the ACT have an overall significantly higher notification rate for Salmonella and Thermophilic Campylobacter than the USA as illustrated by the data presented in Table 6. The authors do not believe that this is due to better reporting in Australia and the ACT.

Table 6

 

Organism

Notification rate per 100,000 for year 2000

ACT

Australia

USA

Thermophilic Campylobacter

108.4

113.8*

20.1

Salmonella

32.3

33.1

12.0

 

*excluding NSW

For the years 1994 to 2000, Salmonella and Thermophilic Campylobacter have been responsible for numerous acute infections in the ACT (See Graph 1). Graph 1 also indicates that the incidence of notified cases of Salmonella and Thermophilic Campylobacter are on the increase in the ACT.

Graph 1

 

ACT Salmonella and Campylobacter rates graph

E. coli

The generic E. coli test is used as an indicator of process control and faecal contamination of the carcass during and after slaughter. The prevalence rates (Tables 1and 6) indicate that the present processing and handling of raw chicken allows for the cross contamination and proliferation of this indicator organism to occur easily. No investigation was undertaken into the prevalence rates for the pathogen Escherichia coli O157:H7.

Listeria Monocytogenes.

While Listeria Monocytogenes is present on raw chicken the epidemiological data from the literature indicates that cooked chicken represents a minor risk. (18) The authors do not know the reason why Listeria Monocytogenes prevalence on raw chicken does not translate into a larger number of incidents of foodborne disease, Listeriosis. It could be that Listeriosis is under reported, as it tends to present itself as a mild disease in healthy populations and only has its full effect in the very young, immunocompromised, elderly and the unborn child. It could also be due serotypical behaviour of the organism, as only certain serotypes are particularly pathogenic and may not reproduce to high numbers against a competitive background flora of bacteria present on raw chicken. Whatever the reason, the cooked food source is currently regarded as low risk.

Salmonella

Our study found 11 different Salmonella serotypes (Table 3) from the 109 positive raw chicken samples sent to the Microbiological Diagnostic Unit (MDU) in Melbourne for serotyping. As up to three isolates per sample were sent to MDU a number of samples were found to have multiple serotypes (Table 4) and some isolates could not be serotyped for various reasons. The 120 notified clinical specimens yielded 39 different serotypes.

Comparison of the Salmonella serotypes (Table 2) obtained from both clinical specimens and raw chicken samples indicate that seven Salmonella serotypes were common to both raw chicken and humans. Four of the serotypes occurred more than once, in both the raw chicken samples and clinical specimens which tends to indicate that they are persisting in both groups. The detection of the same Salmonella serotypes in both raw chickens and humans suggests that a number of ACT Salmonella infections could be acquired from chicken.

Further support for a direct connection between chicken and human Salmonellosis comes from a number of epidemiological studies from around the world which have identified the consumption of undercooked chicken as a significant risk factor. (14)(20)(21)

During the study period there were 120 notified cases of Salmonella which, as explained in the BACKGROUND, is probably an underestimation of the true rate of Salmonellosis in the ACT. A UK study found 3.2 times as many community cases of Salmonella as that notified. (2) Extrapolating this UK study to the ACT on a per capita basis would result in a figure of 384 cases for the ACT. Graph 2 illustrates the difference between the ACT reported notified cases and the estimated actual community cases.

Graph 2

 

Notification Rates ACT graph

To illustrate the possible savings for the ACT of reducing the prevalence of Salmonella on raw chicken, one of the larger American studies (14) calculated that a significant risk was associated with eating chicken prepared outside the home. This study, while it found that chicken cooked in the home was not a significant risk factor, determined that the Population Attributable Risk (PAR) for Salmonella enteritidis (SE) infection, due to the consumption of chicken outside of the home, was 25%. This indicates that if no chicken was consumed outside of the home, 25% of the SE infections in the USA could be avoided. While SE is not a significant cause of clinical cases in the ACT we believe this study can be extrapolated to the ACT as the Joint FAO/WHO Risk Assessment for Salmonella sp determined that (SE) is not significantly different from other non-typhoidal strains.(20) If this study is extrapolated to Salmonella infections in the ACT, 40 notified cases and 130 community cases could be prevented.(2)

Campylobacter

Table 5 indicates that 90% of the Thermophilic Campylobacter from chicken samples and 62% from clinical specimens possess the Cla1 gene. This indicates that at least 54.9% of human ACT Campylobacter infections could be derived from chickens. This is consistent with the recently proposed FDA figure of between 47 - 70% for the USA (1).

There were 333 notified cases of Thermophilic Campylobacter during the study period but the actual number of ACT infections is most probably many times this figure. A UK investigation found 7.6 times as many community cases compared to those that were notified (2). Once again the extrapolation of the UK data to the ACT would put the actual figure at 2,531 cases per year for the ACT. Graph 3 illustrates the difference between the reported ACT notified cases and the under reported actual community cases.

Graph 3

 

ACT Community andNotification Rates for Campylobacteriosis graph

Case Control Studies of local Campylobacteriosis have found that eating of poultry cooked outside of the house is a significant risk factor (9)(11). Based on the ACT prevalence rate for Thermophilic Campylobacter in chicken, if the chickens could be sold Thermophilic Campylobacter free this could save 181 notified or 1377 community cases.

Cost to the community

In 1999 the Australia New Zealand Food Authority (ANZFA) estimated that Australia as a whole experiences approximately 4.2 million cases of food poisoning a year, at a cost to the Australian community in excess of 2.6 billion dollars (13). When these figures are extrapolated to the ACT on a per capita basis the cost to the ACT community is approximately $42,500,000 a year.

The cost of foodborne disease cannot only be measured in money but also in physical and emotional costs as well. The anguish felt by other family members when seeing one of there loved ones, taken to hospital is hard to quantify in purely financial terms. In America 10% of Thermophilic Campylobacter, 15% of Salmonella and 89% of Listeria monocytogenes cases are hospitalised with 15% of Listeriosis patients dying.(3) One Australian study of Thermophilic Campylobacter culture proven cases found that 43% of the cases attended casualty and 31% were admitted to hospital. (11) In the ACT for the year 1999 - 2000 there were 15 (5.8% of notifications) admissions associated with Thermophilic Campylobacter and 8 (7.8% of notifications) for Salmonella.

What is probably less appreciated, is the chronic sequelae caused by these infections, which may occur in 2 3% of cases. The sequelae are often more detrimental to health than the acute disease (5). Probably the best known is Guillain Barre Syndrome but the list also includes cases of Rheumatoid, Reactive Arthritis, Reiters Syndrome, renal disease (5) and possibly Irritable Bowl Syndrome (6). A study in the USA has indicated that approximately one case of Guillain Barre Syndrome (GBS) occurs for every 1000 cases of Campylobacteriosis with up to 40% of patients with GBS having evidence of recent Thermophilic Campylobacter infection. GBS is a subacute, acquired, inflammatory demyelinating polyradiculoneuropathy, with approximately 20% of cases being left with some form of disability and 5% dying. (8) A number of serotypes of Thermophilic Campylobacter appear to be associated with GBS and its variants around the world. The biological mechanism appears to involve molecular mimicry where human peripheral nerves share epitopes with surface antigens of certain Thermophilic Campylobacter serotypes. An immune reaction triggered by the Thermophilic Campylobacter in the intestine initiates an autoimmune reaction against the peripheral nerves. (5)

Additionally GBS cases can spend considerable time in hospital. Between 1991 and 1998 The Canberra Hospital (TCH) had an average of 13.9 admissions per year for this disease with an average length of stay of 30.5 days. From an economic perspective GBS is a costly disease. In America the portion of GBS cases attributable to Campylobacter is estimated to cost the country between 0.2 and 1.8 billion US$ a year. (23)

The good news is that so far the GBS implicated serotypes have not been detected from the 20 clinically derived Thermophilic Campylobacter which have been serotyped.

Pathogen Reduction Strategies

In 1996 the U.S. Department of Agriculture introduced a Pathogen Reduction and Hazard Analysis of Critical Control Points (HACCP) rule on the chicken processing industry for the reduction of pathogens on chicken carcasses. (3) Preliminary, year 2000 FoodNet data from the USA indicates that the notification rate of Campylobacter infections (per 100,000) in humans has fallen from 23.5 in 1996 to 20.1 in 2000 and for Salmonella from 14.5 to 12.0. (4) This demonstrates that by reducing the prevalence of significantly contaminated carcasses leaving the processing houses you can reduce the incidence of human infections.

Incidence of pathogens on cuts of chicken.

The surface of meats themselves do not normally, inherently contain pathogenic organisms but can acquire the organisms from their own faecal matter or from cross contamination at slaughter. The organisms tend to remain on the surface or just under it. The final numbers of bacteria present becomes a factor of the original mix of bacteria present, type of carcass processing, storage and transport conditions. A persons likelihood of acquiring an infection from the meat will depend on a number of factors, for example, how many of the pathogenic organisms are consumed and how well their body deals with the organisms. The best possible situation is for meat to be:

  • free of contamination or alternatively only lightly contaminated during slaughter;
  • transported and stored under controlled chilled conditions as this ensures minimal increase in bacterial numbers;
  • thoroughly cooked at the consumption stage i.e. achieving a temperature of at least 71 C in the centre of the meat and preferably holding that temperature for at least 15 seconds. To ensure the killing of Listeria monocytogenes this cooking time should be extended to two minutes. It should be remembered that visual checks are not always reliable as meat may appear cooked below the safe cooking temperature ie ground beef will start to turn brown as low as 57 C.

A visual inspection of the raw isolation rates of organisms from different cuts revealed that their numbers were not evenly distributed. As a result of this observation a simple statistical analysis was performed to see if the observed differences were statistically significant. Utilising Chi-square c 2 (applying Yates correction) at P= <0.05 the statistically significant observations are shown in Table7.

Table 7

 

Organism

L. monocytogenes

E. coli

Salmonella

Observation

Lower on breasts but higher on drum sticks and thigh fillets

Lower on drumsticks

Higher on wings

 

This indicates that certain pathogens are more prevalent on particular cuts of chicken and need to be cooked well.

CONCLUSION

To compare the pathogen levels on chicken to other raw meats, the Health Protection Service undertook a small survey in 1999-2000 into meats other than chicken. (19) Comparison of these surveys indicates that raw chicken has a much higher prevalence of pathogens than other types of raw meat.

Worldwide the majority of Salmonellosis, Campylobacteriosis and Listeriosis are regarded as being foodborne in origin. USA data attributes 80% of Campylobacter, 95% of Salmonella and 80% of Listeria (17) cases to a food source. Chicken is thought to represent 47 - 70% of Campylobacter (1) and approximately 25% of Salmonella (14) cases, with cooked chicken regarded as a minor risk for Listeria monocytogenes. (18)

This ACT survey has determined:

  • that a strong correlation exists between raw chicken and human cases of Salmonellosis and Campylobacteriosis,
  • foodborne disease is of significant health concern in the ACT as it afflicts approx one in four people every year at a cost of over $143 per person per year,
  • pathogens associated with chickens can be effectively avoided by an appropriate pathogen reduction program as instituted in the USA and a publicity campaign to establish a thorough cooking regime could considerably reduce the cost of foodborne illness to the ACT,
  • if no one in the ACT were to eat undercooked chicken or cross contaminate other foods while handling raw chicken, there would be 221 less notified foodborne disease cases which could equate to 1507 less community cases.

RECOMMENDATION

That the HPS actively lobby relevant government agencies:

  • to implement a pathogen reduction scheme in chicken slaughterhouses; and
  • to institute adequate quality control procedures for the storage and handling of the product from the farm yard to the plate. This scheme should be aimed at reducing both the prevalence and concentration of pathogens on chicken carcasses.

That consumers be made aware of the differing risks posed from different cuts of chicken eg wings have a higher incidence of the main pathogens than breasts.

BIBLIOGRAPHY

Risk assessment of fluoroquinolone use in poultry

Stuart Handysides. 1999 Underascertainment of infectious intestinal disease, Communicable Disease and Public Health. Vol 2 (2) pp 78-80.

FoodNet Surveillance Report for 1999 (Final Report)

Preliminary FoodNet data on the Incidence of Foodborne IllnessSelected Sites, United States, 2000

Chronic Sequelae of Foodborne Disease

Neal RK, Hebden J, Spiller R, 1997.Prevalence of gastrointestinal symptoms six months after bacterial gastroenteritis and risk factors for development of irritable bowl syndrome: postal survey of patients, British Medical Journal, (314), pp 779-82

Flemming Bagar Ed. 2000, DANMAP 99. Consumption of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, food and humans in Denmark. Danish Veterinary Laboratory.

Campylobacter jejuni Strains from patients with Guillain-Barre Syndrome

Alkekruse S.F. et al Campylobacter jejuni- An Emerging Foodborne Pathogen

Risk Factors for Sporadic Campylobacter Infections in the United States: A Case control Study on FoodNet Sites.

Leanne Unicomb, Andrew Voetsch, Craig Dalton. 2001.Risk Factors for Foodborne Campylobacteriosis in the Hunter Valley, NSW, Australia. Hunter Public Health Unit. NSW OzFoodNet Site.

Nationwide Baseline Data Collection Program

ANSFA, 1999 Food Safety Standards Costs and Benefits. an analysis of regulatory impact of the proposed national food safety reforms

Kimura A et al 1998 Chicken, a newly identified risk factor for sporadic Salmonella serotype Enteritidis infections in the United States: A case-control study in FoodNet sites.

Korolik Victoria and Peter Cloe 1997 Tummy ache bugs. Microbiology Australia, July pp 16-17

Mead et al, 1999. Food-Related Illness and Death in the United States. Emerging infectious Diseases. Sep-Oct 5(5):pp 607-25

Microbiology Quality of Raw Chicken. 1995-6. ACT Health Protection Service.

Bad Bug Book. Listeria monocytogenes USFDA,CFSAN

Microbial Quality of Meats other than Chicken 1999 2000. ACT Health Protection Service.

FaZil A et al. Risk assessment: Salmonella spp. In broilers and eggs (Preliminary report) Joint FAO/WHO Expert Consultation on Risk Assessment of Microbiological Hazards in Food, July 2000

Bryan F L., Doyle MP. Health Risks and consequences of Salmonella and Campylobacter jejuni in raw poultry. Journal of Food Protection. 1995. 58:326-

Buzby C.J. and Roberts T, Estimated Annual Costs of Campylobacter-Associated Guillain-Barre Syndrome Agricultural Economics Report No. 756 40 pp. July 1997

World Poultry Consumption

APPENDIX

Appendix 1

Organism distribution by individual cuts for the 1999 - 2000 survey

 

CUT

No. of Samples

No. of samples Positive for:

E. coli

L. monocytogenes

Salmonella

Campylobacter sp

Bones

4

2

0

2

1

Breast (skinless)

33

21

1

13

4

Breast chunks

9

9

1

6

2

Breast fillet

21

15

5

7

6

Breast mince

1

1

1

0

0

Casserole cuts

1

1

1

0

1

Cordon bleu

1

0

0

0

0

Diced

2

0

0

0

0

Drummetts

4

2

4

1

1

Drumsticks

54

32

31

18

9

Fillet

5

4

0

3

1

Giblets

1

1

1

0

1

Kiev

1

0

0

0

0

Legs

7

5

2

1

2

Liver

1

1

0

1

1

Maryland

2

2

2

1

1

Medallion

1

0

0

0

0

Mince

2

2

1

1

0

Neck

1

1

1

1

0

Neck (no skin)

1

1

0

1

1

Pieces

3

2

0

2

1

Sausage

1

1

1

0

0

Schnitzel

2

1

0

1

0

Stir fry

4

1

0

1

0

Stroganoff

1

1

1

0

0

Tenderloin

15

14

3

6

0

Thigh cutlets

5

5

2

1

1

Thigh fillets

19

17

12

6

3

Thighs

24

20

9

8

7

Whole

1

1

0

0

0

Wings

39

26

17

24

12

TOTAL(%)

266

188(70.7)

96(36.0)

106(39.9)

55(20.6)

 

Appendix 2

Full table of isolates from both chicken surveys and clinical isolates 1999-2000

 

Serotype

Chicken 1999-2000

n=100

Chicken 1995 – 6

n=51

Clinical 1999-2000

n=120

No of Isolates

No of Isolates

No of Isolates

S. Agona

0

3

1

S. Albany

0

0

1

S. Anatum

0

1

0

S. Arizonae

0

0

1

S. Binza

0

0

1

S. Birkenhead

0

0

1

S. Blockley

0

0

1

S. Bovismorbificans

0

0

1

S. Brandanburg

0

0

1

S. Bredeney

0

0

1

S. Enteritidis

0

0

5

S. Hadar

0

0

1

S. Havana

0

0

1

S. Heidelburg

0

0

1

S. Hvittingfoss

0

0

1

S. Infantis

0

0

1

S. Javiana

0

0

1

S. Kiambu

18

0

0

S. Mgulani

0

0

3

S. Muenchen

0

0

1

S. Ohio

0

2

0

S. Oslo

0

0

2

S. Paratyphi B bv Java

0

0

5

S. Rislen

0

0

1

S. Saint Paul

0

0

5

S. Schwarzongrand

0

0

1

S. Singapore

0

3

2

S. Sofia subsp II

54

32

0

S. Subsp II rough

2

0

0

S. Subsp 6.7 fg

0

0

1

S. Typhimurium untypable

2

0

2

S. Typhimurium RDNC

1

0

2

S. Typhimurium 6

0

0

1

S. Typhimurium 8

0

0

1

S. Typhimurium 9

2

6

32

S. Typhimurium 22

0

0

1

S. Typhimurium 64

5

1

1

S. Typhimurium 101

0

0

2

S. Typhimurium 120

0

0

1

S. Typhimurium 135

6

3

8

S. Typhimurium 135a

1

0

0

S. Typhimurium 170

0

0

1

S. Typhimurium 179

0

2

0

S. Typhimurium 193

1

0

0

S. Typhimurium U290

0

0

1

S. Virchow

0

0

8

S. Welteveden

0

0

1

S. Zanzibar

1

0

0

Download

Microbiological Status of Raw Chilled Chicken

Microbiological Quality of Seed Sprouts

APRIL - JUNE 2001

Report prepared by Geoff Millard and Simon Rockliff

OBJECTIVE

  • To determine the microbiological status of fresh seed sprouts sold in the ACT.

BACKGROUND

Literature sources indicate that seed sprouts have been implicated in a number of foodborne illness outbreaks around the world. Between 1973 and 1998 there have been a 15 worldwide outbreaks involving seed sprouts. The organisms involved were Salmonella sp.10 outbreaks, Escherichia coli (E. coli) O157: H7 3 outbreaks, E. coli O157: NM and Bacillus cereus (B. cereus) one each. There has not been a B. cereus outbreak since the original one in 1973. (1) Recorded outbreaks so far have effected around 8266 people with one death. This death occurred in one of the 153 hospitalised cases from the approximately 6000 community cases caused by E. coli O157: H7 in Japan in 1996. The carrier in this case was white radish sprouts. (2) Australian alfalfa seeds were implicated in two of the Scandinavian Salmonella Bovismorbificans outbreaks in 1994 (1). The US Food and Drug Administrations concern over these cases prompted them to issue a guidance document in 1999 requiring the microbiological testing of spent seed irrigation water. This testing should be performed approximately 48 hours after sprouting and no later than 48 hours prior to harvesting. (3)

While there is no epidemiological evidence for foodborne illness outbreaks from sprouts in Australia it was considered prudent to examine this product as it is sold in stores thoughout the ACT. This short survey was undertaken to determine the microbiological status of certain microbiological indicators and pathogens in seed sprouts.

STANDARDS

The Draft Australia New Zealand Food Authority Guidelines for Ready-to-eat-food is applicable to this product. See table 1.

Table 1

Microbiological Quality (colony forming units per gram) Satisfactory Marginal Unsatisfactory Potentially Hazardous  
Indicator          
Escherichia coli <3 3-100 ≥ 100 *  
Pathogens          
Coagulase positive Staphylococci < 102 102-103 103-104 ≥ 104 SET+VE  
Bacillus cereus and other pathogenic Bacillus sp < 102 102-103 103-104 ≥ 104  
Salmonella sp Not detected in 25g     detected  
Listeria monocytogenes Not detected in 25g     ≥ 102**  

*Pathogenic strain of E. coli, such as E. coli O157, should be absent

**The detection of Listeria monocytogenes in ready-to-eat foods prepared specifically for "at risk" population groups eg the elderly, immunocompromised and infants should also be considered as potentially hazardous

SET+VE = Staphylococcus enterotoxin positive

SURVEY

During the period 02/04/2001 to 26/06/2001, 62 samples of seed sprouts representing 15 different types of seeds from 13 ACT retail establishments were collected. These samples were tested to enumerate the presence of E. coli, coagulase positive Staphylococci, Bacillus cereus and the presence or absence of Listeria monocytogenes (L. monocytogenes) and Salmonella sp. In response to a positive samples six extra samples were collected and tested for E. coli and one additional sample was collected and tested for coagulase positive Staphylococci only. The seven additional samples are not included in the results.

Testing for pathogenic strains specific serotypes of E. coli such as E. coli O157 and Staphylococcus enterotoxin were not undertaken due to the cost involved in testing a low number of positive samples.

RESULTS

Overall isolation rates for the 62 samples is given in Table 2.

Table 2

Organism (n = 62) No. of samples positive (%) Range of positive Results*
E. coli 11(17.7) 3->1100
Coag Positive Staphylococci 1(1.6) 21000
B. cereus 2 (3.2) 50-150
Salmonella sp. 0(0.0) NA
Listeria monocytogenes 0(0.0) NA
TOTAL 14(22.6) NA

One of the six extra samples contained E. coli

NA= Not applicable.

*expressed as colony forming units per gram

E. coli

E. coli was isolated from 11 of the samples with a median and mode of four organisms per gram.

See Table 3.

Table 3

Microbiological Quality (%) Satisfactory Marginal Unsatisfactory Potentially Hazardous
Escherichia coli 51(82.3) 10(16.1) 1(1.6)*  

*Further testing from the same batch of Mung Beans resulted in E. coli levels in the potentially hazardous category. A further five samples of Mung Beans from the same manufacturer were obtained from the next lot and these were all found to be satisfactory for E. coli.

Coag Positive Staphylococci

Coag Positive Staphylococci was only isolated from one sample of Snow Pea shoots. See Table 4.

Table 4

*A further tested sample of Snow Pea shoots from the same manufacturer and outlet but different lot was found to be satisfactory for Coag Positive Staphylococci.

Microbiological Quality (%) Satisfactory Marginal Unsatisfactory Potentially Hazardous
Coag positive Staphylococci 61(98.4) 0(0.0) 0(0.0) 1(1.6)*

B. cereus

B. cereus was detected in two of the samples. See Table 5.

Table 5

Microbiological Quality (%) Satisfactory Marginal Unsatisfactory Potentially Hazardous
B. cereus 60(96.8) 1(1.6) 1(1.6)* 0(0.0)

*Only one sample (Crunchy combo) was found to have B. cereus in the unsatisfactory category. The level of 150 cfu/g would not be considered hazardous. No other sample of crunchy combo was found to contain B. cereus.

Salmonella sp and Listeria monocytogenes were not isolated from any of the samples in this study.

DISCUSSION

Sprouts are generally eaten fresh and pathogens can exceed 107 per gram without adversely affecting the appearance of the product. (1) During the germination stage sprouts are considered to be potential potent bacterial amplifiers. (1)

While the survey detected one sample of mung beans with E. coli and one sample of snow peas with Coag Positive Staphylococci in the potentially hazardous category further investigations and testing indicated that these high counts appeared to be restricted to particular batches.

Pathogen levels (in this case E. coli and B. cereus) falling into the marginal category are likely to be restricted in reaching the unsatisfactory or potentially hazardous categories, if the germinated sprouts are stored under refrigeration conditions.

CONCLUSION

The quality of seed sprouts sold in the ACT is generally good, but occasional batches may contain potential pathogens. Germinated sprouts should be stored under refrigeration conditions.

RECOMMENDATION

  • That a future survey be conducted to confirm the continued quality of the product.
  • That people are informed to wash sprouts thoroughly before eating.

BIBLIOGRAPHY

Taormina et al, 1999. Infections Associated with Eating Seed sprouts: An International Concern.

Thompson S, Powell D. 2000. Risks Associated With the Consumption of Fresh Sprouts.

FDA ,.1999. Sampling And Microbiological Testing Of spent Irrigation Water During Sprout Production.

Download

Microbiological Quality of Seed Sprouts