SEARCH

SEARCH BY CITATION

Keywords:

  • food safety;
  • food-borne pathogens;
  • ham;
  • Listeria monocytogenes;
  • microbiological survey;
  • pâté

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Aims:  To gauge the effectiveness of pâté and ham manufacturers’ management of the microbial safety and quality of their products.

Methods and Results:  A survey of 60 batches of prepackaged pâté showed that 41·7% of the batches had aerobic plate counts (APC) exceeding 105 CFU g−1, one of pâté sample contained a Bacillus cereus count of >5000 CFU g−1 and another contained 1700 CFU g−1 of Listeria monocytogenes. No other pathogens were isolated from any of the samples. The survey of prepackaged ham showed that only 1% (1/104) of the ham samples were positive for L. monocytogenes (50 CFU g−1).

Conclusions:  The presence of microbial hazards in these foods has generally declined since the early 1990s in New Zealand. Noncomplying APC levels may be due to an over-estimation of product shelf life or poor food handling practices during manufacture.

Significance and Impact of the Study:  Few of the samples tested contained pathogens at significant levels. The prevalences of L. monocytogenes in pâté and ham were low. The presence of 1700 CFU g−1 of L. monocytogenes in a pâté sample indicates that occasionally, the population can be exposed to levels of L. monocytogenes above the zero tolerance level set in New Zealand.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Processed ready-to-eat foods have been implicated in the transmission of foodborne disease in New Zealand (Brett et al. 1998; Sim et al. 2002). Packaged foods with a long refrigerated shelf life need to be free of cold tolerant pathogens such as Listeria monocytogenes as growth can occur over time so increasing the level of exposure of the population to this foodborne pathogen (Tompkin 2002). The presence of L. monocytogenes in pâté has recently been surveyed in Spain (Dominguez et al. 2001) and Europe (Nichols et al. 1998). A study in the UK included microbiological quality of sliced meat as well as that of pâté (Elson et al. 2004), as did a Belgian survey (Uyttendaele et al. 1999). The findings of these studies showed a general improvement in the microbiological quality of ready-to-eat pâté and ham at retail level.

In New Zealand, concerns about the microbial quality of pâté resulted in the formation of the New Zealand Pâté Manufacturers Association and the introduction of a Code of Practice in 1990. Since 1996, HACCP based Food Safety Programmes have been promoted following the introduction of legislation. Survey results from a previous study (Gower 1996) showed some improvement following implementation of these control measures but highlighted the need to encourage manufacturers to adhere to the code of practice and to widen the implementation of the programmes within the industry.

Similarly, surveys of ready-to-eat meats carried out previously have shown prevalences of Listeria of the order of 33–50% although most of the isolates were identified as Listeria innocua (Hudson et al. 1991, 1992). For pork products, a prevalence of 2·9% was noted for L. monocytogenes.

To update the microbial information of these two products, two separate surveys were carried out between 2002 and 2004. The pâté survey, carried out in 2002 was conducted for two reasons; first to gauge the effectiveness of current pâté manufacturers’ management of the safety and microbial quality of their product and secondly, to provide quantitative data for future risk assessment purposes. The results, although collected differently to the previous studies, provide an overview of the direction of food safety in the manufacture of pâté in New Zealand.

The ham survey was conducted over one full year, between April 2003 and March 2004, for three reasons; first to obtain prevalence data of L. monocytogenes to allow comparison with those produced a decade earlier, secondly to provide quantitative data for risk assessment purposes, so filling a data gap identified in a risk profile examining L. monocytogenes in ready-to-eat meats (Lake et al. 2002) and thirdly to examine ready-to-eat meats in light of an outbreak of noninvasive febrile listeriosis caused by L. monocytogenes in cooked meat (Sim et al. 2002).

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Sample collection

Preliminary excursions to supermarket outlets and from searches of advertising on the internet revealed that there were nine brands (A–I) of prepackaged ready-to-eat pâté available on the retail market in New Zealand. The range of pâté varieties produced by each company differs with brand C producing seven different varieties compared with brands F and G, which produce only one type (Table 1). Between August and October 2002, five retail packs of each batch (bearing the same ‘best before’ date) of different varieties for each brand, were purchased from supermarket outlets in Auckland and Christchurch. Purchasing of samples was staggered over a 2-month period to capture the varieties that were available in the retail market. For each variety of each brand, different lot numbers were purchased from retail outlets in both cities. Where the distribution of some varieties of one brand was limited to one city, only one batch was purchased in that city. Seven of nine brands were distributed in both cities.

Table 1.  Results of pâté samples that have exceeded the New Zealand Microbiological Criteria for Foods (meat paste or spread – including pâté)
BrandNo. varieties availableNo. lots purchased (no. samples)No. noncomplying lotsNo. (%, 95% CI) samples exceeding the criteriaNo. samples showing presence of mould
APC > M (105 CFU g−1)B. cereusM (5 × 103 CFU g−1)Bacillus spp. > M (103 CFU g−1)L. monocytogenes (25 g−1)
  1. APC, aerobic plate counts.

  2. *Listeria monocytogenes of 1·7 × 103 CFU g−1 were enumerated from sample.

A48 (40)1  1 (2·5, 0·1–13·2)   1
B612 (60)9 33 (55, 41·6–67·9)1 (1·7, 0–8·9)  3
C78 (40)0     
D510 (50)2   7 (14, 5·8–26·7)  
E68 (40)5 13 (32·5, 18·6–49·1)  4 (10, 2·8–23·7) 5
F11 (5)1  1 (20, 0·5–71·6)    
G11 (5)0     
H48 (40)8 40 (100, –)  1 (2·5, 0·1–13·2)*1
I34 (20)2  6 (30, 11·9–54·3)    
Total 60 (300)28105 (35, 29·6–40·7)1 (0·3, 0–1·8)11 (3·7, 1·8–6·5)1 (0·3, 0–1·8)10

Packaged ham samples were purchased from retail outlets in Auckland, Wellington and Christchurch between April 2003 and March 2004. Sample collection was spread throughout the year on an ad hoc basis to achieve a target of at least 100 samples and covering as many styles and varieties of ham and brands that were available at retail. On each sampling occasion, only one sample per variety (100–250 g pack) was purchased.

All samples were collected and held in an insulated container containing frozen cooling packs to ensure samples are kept chilled at <8°C during transportation back to the laboratory. On receipt of samples at the laboratory, details were recorded of the premises from which the samples were purchased, batch number, ‘best before’ date, brand, and variety. All samples were held at 4°C and tested at the end of shelf life.

Microbiological analysis

Pâté samples were analysed for microbial quality according to the New Zealand Microbiological Reference criteria for meat paste or spread – including pâté (Anon. 1995). Five pâté samples of each lot were pooled for Salmonella analysis utilizing the TECRA Salmonella visual immunoassay (TECRA International Pty Ltd, Frenchs Forest, NSW, Australia) according to the manufacturer's instructions. Samples that were screened positive by TECRA were further confirmed according to the method of Andrews et al. (2001). Analyses for Bacillus cereus, Clostridium perfringens and aerobic plate count (APC) followed by American Public Health Association methods (Bennett and Belay 2001; Labbe 2001; Morton 2001). All Bacillus-like colonies were confirmed for sporangium morphology by Gram stain. The procedures for Campylobacter were adapted from Wong et al. (2004) where 10 g samples were used for isolation and identification procedures were according to Stern et al. (1998). Listeria spp. were tested by the TECRA Listeria visual immunoassay method and enumerated as previously detailed (Sim et al. 2002). A positive Listeria screen by TECRA was followed by culture isolation and confirmation to species level according to Hitchins (1998).

Ham samples were screened for the presence of Listeria species only, enumerated and confirmed to species level for positive TECRA samples following the methods described above.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Pâté

The microbiological results obtained from pâté testing were assessed for compliance with the New Zealand Microbiological Reference Criteria for Foods (Anon. 1995), which gives the following criteria for meat paste or spread – including pâté:

  1. For Bacillus spp., the B. cereus criterion was followed.

  2. *n represents the number of samples from each batch/lot required for examination.

  3. c represents the maximum allowable number of defective samples after which whole batch/lot is rejected.

  4. m represents the acceptable level above which samples are marginally acceptable.

  5. M represents the values above this level are unacceptable. One or more sample to exceed this level causes rejection of whole batch/lot.

Microbiological analytesn*cmM
Aerobic plate count (CFU g−1)52104105
Clostridium perfringens count (CFU g−1)52102103
Bacillus cereus count (CFU g−1)52102103
Campylobacter (10 g−1)50 0 
Salmonella (25 g−1)50 0 
Listeria monocytogenes (25 g−1)50 0 

Pâté samples of nine major brands (A–I), made up of 60 batches, each comprising of five units to give a total of 300 samples, were analysed. The range of pâté varieties produced by each manufacturer varies and samples were purchased according to their availability.

The criteria apply to pâté at any point in the shelf life of the product. Of 60 batches of pâté surveyed, only one individual sample within a batch of brand H basil pesto pâté did not comply with the L. monocytogenes criteria (batch prevalence of 1·7%) as it had a count of 1700 CFU g−1 (Table 1). In all of the other samples L. monocytogenes was not detected. The APC of 25 batches exceeded M (105 CFU g−1) (Fig. 1). Eight of those batches were from brand H where the APC in every sample (n = 40) exceeded 107 CFU g−1. A sample from brand B contained a count of >5000 CFU g−1 of B. cereus. This sample, a cracked pepper pâté, also had an APC of 106 CFU g−1. Three batches from brands D and E, exceeded M levels for other Bacillus spp. Two of the three batches of pâté with Bacillus spp. at levels exceeding 103 CFU g−1 were also of the cracked pepper variety.

image

Figure 1. Distribution of aerobic plate count data from pâté. (bsl00036), counts of <102 CFU g−1; (□), counts of 1·1 × 102–1·0 × 103 CFU g−1; (bsl00014), counts of 1·1 × 103–104 CFU g−1; (bsl00015), counts of 1·1 × 104–105 CFU g−1; (bsl00001), counts of >105 CFU g−1

Download figure to PowerPoint

Salmonella and Campylobacter were not isolated from any of the samples and C. perfringens counts were all <10 CFU g−1, which was the limit of detection of the method. Although moulds were not specifically cultured from these samples, mould colonies were observed on the APC plates of ten samples. One of these samples showed visible mould growth in an area where the packaging seal was faulty (Table 1).

Packaged ham

One company dominates the New Zealand market and sampling was designed to reflect this in that 36% of the samples came from this manufacturer (brands 8 and 10). A total of 104 ham samples were purchased and analysed.

The criterion used for L. monocytogenes in packaged ham is based on zero tolerance (c = 0) for n = 5 (Anon. 1995). Of 104 samples of packaged ham, only seven (6·7%), contained Listeria and only one sample from brand 1 (1·0%) contained L. monocytogenes (Table 2). Of the seven positive samples, five belonged to brand 5. Two of these samples were contaminated with Listeria welshimeri (<50 and 450 CFU g−1) and three with L. innocua (<50, 200 and 400 CFU g−1). The only other sample belonging to brand 9 contained <50 CFU g−1 of L. innocua.

Table 2. Listeria isolation and counts from the 16 brands of retail prepacked ham
Brand*No. samples showing presence/absence of Listeria 25 g−1Total
AbsentL. welshimeriL. monocytogenesL. innocua
  1. *Brand 8 was also processed and packed by brand 10.

  2. †Figures in parenthesis represent counts of Listeria in CFU g−1 in each positive sample.

 12 1 (50)† 3
 23   3
 38   8
 41   1
 502 (<50, 450) 3 (<50, 200, 400)5
 62   2
 75   5
 812   12
 91  1 (<50)2
1025   25
111   1
1211   11
131   1
147   7
155   5
1613   13
Total97214104

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This survey of retail prepackaged pâté showed that, generally, the products did not contain pathogens with the exception of one potentially hazardous sample in a batch returning a count of 1700 CFU g−1 of L. monocytogenes at the end of the stated shelf life. The growth of L. monocytogenes in this sample may have been inhibited by the background APC of 108 CFU g−1 also recorded at the end of the shelf life, as inhibition of growth of L. monocytogenes by competing bacteria has previously been shown in food (Bredholt et al. 1999). No other pathogens were detected at significant levels in these samples.

The prevalence of L. monocytogenes in pâté has declined since 1990 and 1996 when 21·9 and 19·2%, respectively, of batches did not meet the zero tolerance criterion (Gower 1996) compared with 1·7% shown in this survey. However, these comparisons, must be interpreted with caution as the samples used for the earlier surveys were sampled immediately after production from the manufacturers’ premises while the samples for the 2002 survey were retail packs and analysed at the end of shelf life. Compared with the lower noncomplying APC results noted in the 1990/1996 surveys, it could be concluded that higher numbers of samples exceeding the acceptable value of M in 2002 shows that the APC of pâté increases over time during storage, as would be expected. It could also be argued that the 2002 samples are just of poorer microbial quality, although this seems unlikely given the unequivocal improvement in Listeria status. It is possible that the emphasis placed on environmental Listeria control by manufacturers as part of their Food Safety Programmes has played a part in this improvement.

The risk posed by the 12 samples of pâté that contained 103 CFU g−1 of Bacillus spp. (one of which was B. cereus) was minimal. Nevertheless, a food poisoning problem could develop if the Bacillus population increased through temperature abuse. Eight of these 12 samples contained cracked pepper and this spice is known to harbour Bacillus (Seenappa and Kempton 1981; Wolfgang and Berger 1994; Hudson and Hasell 1998).

High APC results do not necessarily constitute a risk to public health (Gillespie et al. 2000), but they can indicate a general lack of hygiene. As about a third of the samples in the current survey exhibited unsatisfactory APC levels, it may be concluded that either shelf life has been over-estimated or poor food handling practices have occurred during manufacture. The fact that 10 samples of pâté contained visible mould colonies tends to support both suppositions. Also, the pâté survey found all 40 samples from one manufacturer and 33 of 60 from another exceeded M for the APC. This may indicate that some pâté manufacturers need to reconsider their processes and shelf-life criteria to produce quality foods. As samples from some brands showed a high proportion complying with the microbiological criteria, this survey showed that it is within the capability of the pâté manufacturing industry to produce food that is in compliance with the criteria.

Data for L. monocytogenes in ham samples also suggest an improvement over the last decade. Less than 7% of samples contained Listeria and only 1%L. monocytogenes. The data for Listeria suggest a significant reduction and for L. monocytogenes a modest improvement from 2·9% for ready-to-eat pork products (Hudson et al. 1992) to 1%. Given that the samples were tested at the end of their stated shelf life and that L. monocytogenes can grow on ham during refrigerated storage (Glass and Doyle 1989) the number of the pathogen detected in the one positive sample could be considered as low and as posing a low risk to the population. Recent analysis has estimated that only 0·207% (median value) of the incidence of listeriosis in the USA can be attributed to foods containing <102 CFU g−1 of L. monocytogenes (Chen et al. 2003) and that most of the risk resides in foods contaminated with higher numbers of this pathogen.

The finding that all five samples from brand 5 were positive for Listeria highlights the existence of problems in the processing and/or packaging areas of this manufacturer. The presence of any Listeria species in a cooked product like ham indicates failure of critical control points in processing. It is possible that the two species of Listeria found in these samples have colonized the plant causing subsequent recontamination of end products. Although L. monocytogenes was not isolated in the samples from this manufacturer, this pathogen also resides in similar environments to the other two species and could potentially be introduced by incoming materials. Strains of L. monocytogenes have been found to establish and persist in the food-processing environment and subsequently contaminate foods implicated in causing illness (Tompkin 2002).

In an outbreak of febrile noninvasive listeriosis involving ham, corned beef and luncheon meat contaminated with L. monocytogenes in New Zealand (Sim et al. 2002), products were traced back to production over a 3-month period. The indistinguishable serotypes isolated indicate that the pathogen must have colonized the plant leading to recontamination of products over the period of investigation. During the course of this survey one manufacturer recalled ham because of contamination by L. monocytogenes.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank the New Zealand Ministry of Health and Foundation for Research Science and Technology for financial support and ESR Public Health Laboratories, Jenny Bennett and Maurice Wilson for collection and analysis of samples.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Andrews, W.H., Flowers, R.S., Silliker, J. and Bailey, J.S. (2001) Salmonella. In Salmonella. Compendium of Methods for the Microbiological Examination of Foods, 4th edn, Ch. 37 ed. Downes, F.P. and Ito, K. pp. 357380. Washington, DC: American Public Health Association, ISBN 0-87553-175-x.
  • Anon. (1995) Microbiological Reference Criteria for Food, Version 2.0, October 1995. Wellington, New Zealand: Food Administration Section, Ministry of Health.
  • Bennett, R. W. and Belay, N. (2001) Bacillus cereus. In Compendium of Methods for the Microbiological Examination of Foods, 4th edn, Ch. 32 ed. Downes, F.P. and Ito, K. pp. 311324. Washington, DC: American Public Health Association, ISBN 0-87553-175-x.
  • Bredholt, S., Nesbakken, T. and Holck, A. (1999) Protective cultures inhibit growth of Listeria monocytogenes and Escherichia coli O157:H7 in cooked, sliced, vacuum- and gas- packaged meat. Int J Food Microbiol 53, 4352.
  • Brett, M.S.Y., Short, P. and McLauchlin, J. (1998) A small outbreak of listeriosis associated with smoked mussels. Int J Food Microbiol 43, 223229.
  • Chen, Y., Ross, W.H., Scott, V.N. and Gombas, D.E. (2003) Listeria monocytogenes: low levels equal low risk. J Food Prot 66, 570577.
  • Dominguez, C., Gomez, I. and Zumalacarregui, J. (2001) Prevalence and contamination levels of Listeria monocytogenes in smoked fish and pate sold in Spain. J Food Prot 64, 20752077.
  • Elson, E., Burgess, F., Little, C.L. and Mitchell, R.T. (2004) Microbiological examination of ready-to-eat cold sliced meats and pâté from catering and retail premises in the UK. J Appl Microbiol 96, 499509.
  • Gillespie, I., Little, C. and Mitchell, R. (2000) Microbiological examination of cold ready-to-eat sliced meats from catering establishments in the United Kingdom. J Appl Microbiol 88, 467474.
  • Glass, K.A. and Doyle, M.P. (1989) Fate of Listeria monocytogenes in processed meat products during refrigerated storage. J Food Prot 55, 15651569.
  • Gower, J. (1996) National Pate Project Report. Assessment of the Effectiveness of the New Zealand Pate Manufacturers Code of Practice. Auckland: Auckland Healthcare.
  • Hitchins, A.D. (1998) Listeria monocytogenes. In FDA Bacteriological Analytical Manual (Revision A), 8th edn, Ch. 10. pp. 10.110.13. Gaithersburg, MD: AOAC International.
  • Hudson, J.A. and Hasell, S.K. (1998) The bacteriological quality of herbs and spices imported into New Zealand. New Zealand J Environ Health 21, 2526.
  • Hudson, J.A., Lowry, B.D., Boerema, J.A. and DeLacy, K.M. (1991) Prevalence of Listeria spp. in retail meat products. Comm Dis New Zealand 92, 5657.
  • Hudson, J.A., Mott, S.J., Delacy, K.M. and Edridge, A.L. (1992) Incidence and coincidence of Listeria spp., motile aeromonads and Y. enterocolitica on ready-to-eat fleshfoods. Int J Food Microbiol 16, 99108.
  • Labbe, R.G. (2001) Clostridium perfringens. In Compendium of Methods for the Microbiological Examination of Foods, 4th edn, Ch. 34 ed. Downes, F.P. and Ito, K. pp. 325330. Washington, DC: American Public Health Association, ISBN 0-87553-175-x.
  • Lake, R., Hudson, J. and Cressey, P. (2002) Listeria monocytogenes in Processed Ready-to-Eat Meats. http://www.nzfsa.govt.nz/science-technology/risk-profiles/listeria-in-rte-meat.pdf .
  • Morton, R.D. (2001) Aerobic plate count. In Compendium of Methods for the Microbiological Examination of Foods, 4th edn, Ch. 7 ed. Downes, F.P. and Ito, K. pp. 6367. Washington, DC: American Public Health Association, ISBN 0-87553-175-x.
  • Nichols, G., McLauchlin, J. and de Louvois, J. (1998) The contamination of pate with Listeria monocytogenes– results from the 1994 European Community – Coordinated Food Control Program for England and Wales. J Food Prot 61, 12991304.
  • Seenappa, M. and Kempton, A.G. (1981) A note on the occurrence of Bacillus cereus and other species of Bacillus in Indian spices of export quality. J Appl Bacteriol 50, 225228.
  • Sim, J., Hood, D., Finnie, L., Wilson, M., Graham, C. and Hudson, J.A. (2002) A series of incidents of Listeria monocytogenes non-invasive febrile gastroenteritis involving ready-to-eat meats. Lett Appl Microbiol 35, 409413.
  • Stern, N.J., Line, J.E. and Chen, H.C. (1998) Campylobacter. In Compendium of Methods for the Microbiological Examination of Foods, 4th edn, Ch. 31 ed. Downes, F.P. and Ito, K. pp. 301310. Washington, DC: American Public Health Association, ISBN 0-87553-175-x.
  • Tompkin, R.B. (2002) Control of Listeria monocytogenes in the food-processing environment. J Food Prot 65, 709725.
  • Uyttendaele, M., Troy, P.D. and Debevere, J. (1999) Incidence of Listeria monocytogenes in different types of meat products on the Belgian retail market. Int J Food Microbiol 53, 7580.
  • Wolfgang, K. and Berger, E. (1994) Microbiological criteria of random samples of spices and herbs retailed on the austrian market. J Food Prot 57, 893901.
  • Wong, T.L., Whyte, R.J, Cornelius, A.J. and Hudson, J.A. (2004) Enumeration of Campylobacter and Salmonella on chicken packs. Br Food J 106, 651662.