Evaluation of the TEMPO Ò most probable number technique for the enumeration of Enterobacteriaceae in food and dairy products

Transcription

1 Journal of Applied Microbiology ISSN ORIGINAL ARTICLE Evaluation of the TEMPO Ò most probable number technique for the enumeration of Enterobacteriaceae in food and dairy products M. Owen, C. Willis and D. Lamph Health Protection Agency Food Water and Environmental Microbiology Network (Southampton Laboratory), Southampton General Hospital, Southampton, UK Keywords automated, dairy, Enterobacteria, enumeration, food, TEMPO. Correspondence Caroline Willis, Health Protection Agency Food Water and Environmental Microbiology Network (Southampton Laboratory), Level B South Block, Southampton General Hospital, Southampton SO16 6YD, UK. Caroline.willis@hpa.org.uk : received 16 March 2010, revised 19 May 2010 and accepted 23 June 2010 doi: /j x Abstract Aims: The aim of this study was to evaluate the TEMPO Ò EB automated most probable number (MPN) system for the enumeration of Enterobacteriaceae in comparison with three standard laboratory methods. Method and Results: Pure cultures were used to determine the sensitivity and specificity of the TEMPO Ò EB test. These studies indicated that Enterobacteriaceae were reliably detected with a detection limit below 10 CFU, whilst non- Enterobacteriaceae strains gave negative results at all inoculum levels tested. A total of 209 food, dairy and environmental samples were tested using the four methods, and results indicated that there was no statistically significant difference between the TEMPO Ò EB and either the MPN, pour plate or spread plate techniques. Conclusions: It is considered that, with a small number of exceptions, TEMPO Ò EB is a suitable method for the analysis of routine food, dairy and environmental samples as it has many advantages, and the results obtained are equivalent to those of standard methods. Significance and Impact of the Study: The TEMPO Ò system could provide significant benefits to Food Water and Environmental Microbiology laboratories, by responding to food and dairy hygiene problems rapidly (up to 4 days sooner than previously), reducing media and consumable costs and making savings in hands-on staff time. Introduction The Enterobacteriaceae are a large family of Gram-negative, rod shaped, facultatively anaerobic bacteria, which are often used in food microbiology as indicator organisms for assessing the overall hygiene of food products. Members of the Enterobacteriaceae family are useful indicator organisms as the group is well-defined taxonomically, and the traditional identification tests include lactose-fermenting organisms (i.e. coliform bacteria) and lactose nonfermenting organisms such as salmonellae (PHLS 2000). They are killed by the heat processes used in food production and can be readily removed from the production environment by appropriate cleaning procedures (HPA 2009). Therefore, the presence of Enterobacteriaceae in heat-treated food and dairy products generally indicates problems with process hygiene such as inadequate heat treatment or post-process contamination from raw ingredients or the environment. In swabs of food preparation surfaces, the presence of these organisms can indicate inadequate cleaning. Traditionally, these organisms have been enumerated by means of a pour plate or surface spread technique on selective agar. A confirmed result is usually obtained by these methods in 72 h (British Standards Institution, 2004b). In 2005, new EC legislation (European Commission, 2007) specified the use of a most probable number (MPN) method, BS ISO :2004 (British 1810 Journal of Applied Microbiology 109, ª 2010 The Society for Applied Microbiology

2 M. Owen et al. Evaluation of TEMPO â Enterobacteriaceae method Standards Institution. 2004a), for the enumeration of Enterobacteriaceae in certain dairy products including pasteurized milk. This international standard method is based on the inoculation of triplicate tubes of liquid medium with the food or dairy product at each of a series of consecutive tenfold dilutions. Following subculture and confirmatory biochemical tests, the Enterobacteriaceae count per gram or per millilitre is calculated using statistical tables. The method is particularly suitable when levels of Enterobacteriaceae are likely to be low, and also when the target organisms are likely to have been stressed by processes such as drying or freezing, because of the superior recovery of stressed micro-organisms in liquid growth medium (Health Protection Agency, 2005). However, it is a labour-intensive procedure and also relatively slow, with a confirmed result available after a minimum of four days. Moreover, in the case of dairy products, frequent overgrowth by organisms such as Pseudomonas species can delay the result further. An automated MPN method, TEMPO Ò, has been designed by BioMérieux [BioMérieux (UK) Ltd, Basingstoke, UK], which can be used for the enumeration of a variety of target organisms including Enterobacteriaceae, Escherichia coli, Total Viable Count and Coagulasepositive Staphylococci. The TEMPO Ò system is based on a 16 3 MPN technique in which a disposable MPN card is automatically filled with a pre-prepared dilution of the food sample, together with the TEMPO Ò reagent. The plastic card consists of three rows of 16 wells each, with a well volume of 225 ll in the first row, 22Æ5 ll in the second row and 2Æ25 ll in the third. This difference in volumes effectively achieves a tenfold dilution between one row and the next. The card is incubated and then placed in an automated reader, which detects fluorescence. Glucose fermentation by Enterobacteriaceae causes acidification of the reagent which results in the quenching of fluorescence in positive reaction tubes. A confirmed Enterobacteriaceae result can be obtained by the TEMPO Ò method in 24 h. BioMérieux has obtained AFNOR (Association Française de Normalisation) validation for the TEMPO Ò technique as an alternative analytical method to BS ISO :2004 for Enterobacteriaceae. Previous studies involving the TEMPO Ò system have found that it compares favourably to other methods of enumeration for a number of different organisms. In 2007, a study compared the TEMPO Ò system to the Petrifilm method for the detection of E. coli, Coliforms and Total Aerobic Plate Counts (TAPC) in samples of chicken (Cosby and Bailey 2007). The results were found to be comparable between TEMPO Ò and Petrifilm for TAPC. For E. coli and Coliforms, there was a correlation between TEMPO Ò and the cultural MPN technique but not between TEMPO Ò and Petrifilm. The authors attributed this to the inability of Petrifilm to distinguish low levels of background flora. Another study compared the TEMPO Ò E. coli test (TEMPO Ò EC) with the chromogenic agar, TBX, for the enumeration of E. coli in cheese (Torlak et al. 2008). Results produced by the two methods were found to be comparable. The TEMPO Ò system has been designed to be labour and time saving, and an evaluation of this was carried out as part of a study in TEMPO Ò was found to reduce the analysis time and required less laboratory personnel compared to the conventional laboratory methods used by the authors when analysing 24 food samples for Total Viable Count, Coliforms and E. coli. TEMPO Ò was noted to be particularly time saving in comparison with the time taken preparing media and inoculating plates with multiple sample dilutions (Kunicka 2007). The TEMPO Ò Enterobacteriaceae method (TEMPO Ò EB) was evaluated in comparison with the international standard method for detection and enumeration of Enterobacteriaceae using violet red bile glucose (VRBG) agar (BS ISO :2004) and also the Petrifilm Enterobacteriaceae method (Paulsen et al. 2008). The TEMPO Ò EB method was found to give equivalent results to the two alternative methods for confirmed Enterobacteriaceae. This evaluation also found the TEMPO Ò EB method to be less time consuming than the BS ISO method as the need for confirmation is eliminated. To date, there have been no reported evaluations of TEMPO Ò EB compared to the MPN method, ISO :2004. The aim of this study was therefore to evaluate the TEMPO Ò EB method for the enumeration of Enterobacteriaceae in food and dairy products in comparison with three standard laboratory methods, namely the MPN method, ISO :2004; the pour plate method, ISO :2004; and a surface spread technique based on ISO :2004. Materials and Methods Determination of sensitivity and specificity of TEMPO Ò EB: analysis of Enterobacteriaceae and Non-Enterobacteriaceae in pure cultures Cultures of eleven different Enterobacteriaceae strains [Escherichia coli NCTC 9001, Escherichia coli NCTC 13216, Proteus rettgeri NCTC 7475, Citrobacter youngae (wild strain), Klebsiella aerogenes NCTC 9528, Serratia marcescens NCTC 11935, Enterobacter aerogenes NCTC 10006, Edwarsdsiella tarda NCTC 11934, Salmonella Nottingham NCTC 7832, Yersinia enterocolitica NCTC and Escherichia coli O157 NCTC 12900] and four Journal of Applied Microbiology 109, ª 2010 The Society for Applied Microbiology 1811

3 Evaluation of TEMPO â Enterobacteriaceae method M. Owen et al. bacterial species that do not belong to the Enterobacteriaceae family [Pseudomonas aeruginosa NCTC10662, Enterococcus faecalis NCTC 775, Staphylococcus aureus NCTC 6571 and Aeromonas hydrophila (wild strain)] were prepared in nutrient broth (Oxoid Ltd, Basingstoke, UK) and incubated at 37 C for 18 h. Serial tenfold dilutions of each organism were prepared in Maximum Recovery Diluent (MRD) (Oxoid Ltd), and the counts were enumerated using the surface drop technique (ICMSF, 1988) on Plate Count Agar (Oxoid Ltd). The 10 )4 and 10 )6 dilutions of each of these cultures were also used to inoculate TEMPO Ò cards (i.e. 20 ll of each dilution was added to a TEMPO Ò EB reagent bottle containing 4 ml of sterile distilled water). Cards were filled and incubated, and results read according to the manufacturer s instructions (as described in section Enumeration of Enterobacteriaceae by the TEMPO Ò MPN method). Comparison of TEMPO Ò EB with international standard methods for the enumeration of Enterobacteriaceae: analysis of naturally contaminated and artificially spiked samples A total of 195 naturally contaminated food samples were tested. These consisted of dairy products (71 samples), dried foods (33), ready-to-eat meals (25), sandwiches (25), frozen foods (15), raw fruit and vegetables (11), raw meat (six) and environmental swabs and cloths (nine). In addition, a further 14 samples were artificially spiked with Escherichia coli NCTC 9001, by adding 1 ml of a 10 )5 dilution of overnight culture to 25 g of food. These were peppercorns (five samples), ice cream (three), orange juice (one), dried milk (one), thick cream (one), strawberry yoghurt (two) and strawberry milkshake (one). Sample homogenates were prepared as described below, and Enterobacteriaceae enumerated by the TEMPO Ò method and also by one or more standard culture methods (as described in sections Enumeration of Enterobacteriaceae using a MPN method including pre-enrichment, Enumeration of Enterobacteriaceae using a pour plate technique, Enumeration of Enterobacteriaceae using a spread plate technique). Preparation of food samples A 10 )1 dilution of each food sample was prepared by weighing out a 25 g aliquot of the food and diluting with 225 ml of MRD. The samples were then homogenized in a stomacher (AES Laboratoire, Combourg, France) for 60 s. The homogenate was prepared in a TEMPO Ò stomacher bag with mesh [BioMérieux (UK) Ltd], to reduce the chance of blockage of the filling tube to the TEMPO Ò card by particles of food. Preparation of environmental swab and cloth samples Each swab was homogenized with 10 ml of MRD, and the resulting suspension was then treated in the same way as a 10 )1 food homogenate. Cloths were rinsed by adding 150 ml of MRD to the entire cloth in a stomacher bag and placed in a stomacher for 60 s. The resulting liquid was used to inoculate growth media and TEMPO Ò cards as required. Enumeration of Enterobacteriaceae by the TEMPO Ò MPN method Three millilitres of sterile distilled water was added to a TEMPO Ò EB reagent bottle. One millilitre of sample homogenate (usually a 10 )1 dilution of food sample, or the suspension resulting from the rinsing of swabs and cloths) or undiluted sample, in the case of liquid dairy products, was then added and the bottle vortexed to thoroughly mix the resulting suspension. Cards and reagents were then loaded into the TEMPO Ò Preparation Station according to the manufacturer s instructions, to automatically fill the cards with the appropriate sample dilution. Cards were then incubated at 35 ± 1 C for h. Following incubation, cards were loaded into the TEMPO Ò Reading Station, and the results automatically read according to the manufacturer s instructions. Enumeration of Enterobacteriaceae using a MPN method including pre-enrichment Samples were tested according to Health Protection Agency Standard Method F18 (Health Protection Agency, 2005) which is based on BS ISO : 2004 (British Standards Institution. 2004a). Enumeration of Enterobacteriaceae using a pour plate technique Samples were analysed according to Health Protection Agency Standard Method F23 (Health Protection Agency, 2004), based on BS ISO : 2004 (British Standards Institution. 2004b). Enumeration of Enterobacteriaceae using a spread plate technique Samples were analysed using a method based on BS ISO : 2004, but with the difference that 0Æ5 -ml aliquots of sample homogenate were spread onto the surface of prepoured Violet Red Bile Glucose Agar (VRBGA; Oxoid Ltd.), rather than using pour plates. Statistical analysis Bacterial counts produced by the TEMPO Ò technique and the three traditional enumeration methods were 1812 Journal of Applied Microbiology 109, ª 2010 The Society for Applied Microbiology

4 M. Owen et al. Evaluation of TEMPO â Enterobacteriaceae method compared using a Student s t-test (at the 95% confidence level) and linear regression analysis. Results Determination of sensitivity and specificity: analysis of Enterobacteriaceae and Non-Enterobacteriaceae in pure cultures The eleven Enterobacteriaceae strains tested were detected using the TEMPO Ò technique at both low counts (between 3 and 16 CFU ml )1 ) and high counts (between 300 and 2600 CFU ml )1 ) (Table 1). There was no significant difference between the inoculum levels, as determined by the surface drop technique, and the TEMPO Ò Table 1 Detection and enumeration of pure cultures of Enterobacteriaceae and non-enterobacteriaceae strains by the TEMPO Ò EB method Bacterial strain Inoculum level (CFU ml )1 ) TEMPO Ò result (CFU ml )1 ) Enterobacteriaceae Escherichia coli NCTC* Æ Escherichia coli NCTC Æ Escherichia coli O157 NCTC Klebsiella aerogenes NCTC Æ Citrobacter youngae (wild strain) 9 8Æ Proteus rettgeri NCTC Æ Serratia marcescens NCTC Æ Enterobacter aerogenes NCTC Æ Edwardsiella tarda NCTC Æ Salmonella Nottingham NCTC Yersinia enterocolitica NCTC Æ Non-Enterobacteriaceae Pseudomonas aeruginosa NCTC <1 900 <1 Staphylococcus aureus NCTC <1 500 <1 Enterococcus faecalis NCTC <1 800 <1 Aeromonas hydrophila (wild strain) 3 <1 300 <1 *National collection of type cultures. results for these strains (Students t-test; P = 0Æ52). Organisms which are not members of the Enterobacteriaceae family were reported as not detected by the TEMPO Ò method for counts ranging from three to 900 CFU ml )1 (Table 1). Comparison of TEMPO Ò EB with international standard methods for the enumeration of Enterobacteriaceae: analysis of naturally contaminated and artificially spiked samples Comparison of TEMPO Ò with the MPN method. One hundred and eight samples were analysed by both TEMPO Ò and the MPN tube method (HPA method F18). These comprised 56 dairy samples (milk, ice cream, yoghurt and cream), 33 dried foods (seeds, fruit and biltong), 12 frozen foods (seafood, fruit and vegetables), three ready-to-eat meals, three mushroom samples and one sample of fruit juice. Of these, Enterobacteriaceae were not detected by either method in 33 samples. A further eight samples (seven peppercorns and one spiced biltong) were excluded from the analysis as the samples themselves appeared to cause complete quenching of fluorescence in the TEMPO Ò cards, resulting in false positive results. Analysis of the remaining 67 pairs of results indicated that there was no statistically significant difference between the two methods using the Students t-test (P =0Æ84). Linear regression analysis gave a correlation coefficient (R 2 )of0æ78 (Fig. 1). Comparison of TEMPO Ò with the pour plate method Seventy-one samples were analysed by both TEMPO Ò and the pour plate method (HPA method F23). These Log count per gram (MPN method) R 2 = Log count per gram (TEMPO method) Figure 1 Comparison of Enterobacteriaceae results (Log count per gram) obtained by the TEMPO Ò and most probable number methods. (u) Spiked samples and (r) naturally contaminated. Journal of Applied Microbiology 109, ª 2010 The Society for Applied Microbiology 1813

5 Evaluation of TEMPO â Enterobacteriaceae method M. Owen et al. comprised 26 dairy products (yoghurt, ice cream, cream and dried milk), 23 dried foods (seeds and biltong), five ready-to-eat meals, six raw meat samples, four cloths and swabs, three raw vegetables, three frozen foods (fruit and vegetables) and one sample of fruit juice. Of these, 13 samples gave Enterobacteriaceae results below the limit of detection by both methods. A further seven samples of peppercorns were excluded from the analysis because of the production of false positive results by this sample type. Analysis of the remaining 47 pairs of results indicated that there was no statistically significant difference between the two methods using the Students t-test (P =0Æ37). Linear regression analysis gave a correlation coefficient (R 2 )of0æ75 (Fig. 2). Comparison of TEMPO Ò with the spread plate method One hundred and twenty-eight samples were analysed by both TEMPO Ò and the spread plate method. These comprised 23 ready-to-eat meals, 25 sandwiches, 27 dairy samples (milk, ice cream and yoghurt), 23 dried foods (seeds and biltong), 11 raw vegetables, six raw meats, nine swabs and cloths, three frozen foods (fruit and vegetables) and one sample of fruit juice. Of these, Enterobacteriaceae were not detected by either method in 37 samples. A further seven samples of peppercorns were excluded from the analysis because of this sample type producing false positive results. Analysis of the remaining 84 pairs of results indicated that there was no statistically significant difference between the two methods using the Students t-test (P =0Æ09). Linear regression analysis gave a correlation coefficient (R 2 )of0æ78 (Fig. 3). Log CFU per gram (pour plate method) R 2 = Log count per gram (TEMPO method) Figure 2 Comparison of Enterobacteriaceae results (Log count per gram) obtained by the TEMPO Ò and pour plate methods. (u) Spiked samples and (r) naturally contaminated. Log CFU per gram (spread plate method) Log count per gram (TEMPO method) Discussion R 2 = Figure 3 Comparison of Enterobacteriaceae results (Log count per gram) obtained by the TEMPO Ò and surface spread plate methods. (u) Spiked samples and (r) naturally contaminated. The Enterobacteriaceae are important indicators of potential hygiene problems in food-processing environments. This group of bacteria includes species which can originate from the intestinal tract of humans and animals and so they are often used as indicators of poor food handler hygiene as well as cross-contamination from raw meat or food contact surfaces and as evidence of undercooking (Health Protection Agency 2009). They can also indicate poor time and temperature control in the food production process. To allow a prompt response to hygiene problems, any test designed to detect these organisms should ideally give a rapid and reliable result. The time taken to obtain results is of particular concern to manufacturers of products with a relatively short shelf-life, such as processing dairies, where there is often pressure to release products onto the market as soon as possible after production. The TEMPO Ò system is a 16 3 MPN technique miniaturized into a disposable test card, which is automatically filled by machine with a pre-prepared dilution of the food sample and the TEMPO Ò reagent. During incubation of the MPN card, the fermentation of glucose by Enterobacteriaceae and the resulting acidification of the growth medium leads to the quenching of the proprietary fluorescent ph indicator. Thus, after 24 h of incubation, the TEMPO Ò Reading Station automatically determines the Enterobacteriaceae count by detecting the number of nonfluorescing reaction tubes and comparing the result to statistical values in an MPN table. With over 4000 possible MPN combinations in the 16 3 table, this technique lends itself to an automatic reading platform, although examining the card manually under a UV light and comparing to the MPN table is theoretically possible Journal of Applied Microbiology 109, ª 2010 The Society for Applied Microbiology

6 M. Owen et al. Evaluation of TEMPO â Enterobacteriaceae method The evaluation described here was carried out primarily to compare the TEMPO Ò EB system to the MPN method for the enumeration of Enterobacteriaceae, BS ISO 21528:1:2004 (British Standards Institution. 2004a). This method is specified in European legislation (European Commission, 2007) for use in the analysis of certain dairy products including pasteurized milk. It is also favoured within the Health Protection Agency s Food Water and Environmental Microbiology Network for the analysis of frozen and dried food samples, because the liquid growth medium is known to improve the recovery of stressed organisms compared to solid media. However, it is an expensive and labour-intensive method and relatively slow to produce results. For dairy products, in particular, the overgrowth of Enterobacteriaceae by background flora such as Pseudomonas species can frequently present an additional problem. Subsequent efforts to obtain a pure culture for confirmatory tests can delay the reporting of results by a further 1 or 2 days. The TEMPO Ò system was viewed as a possible alternative to the British Standard as it retains the advantages of a MPN method (i.e. improved recovery of stressed organisms and low detection limit) whilst giving a more rapid result (24 h, compared to a minimum of four days in the case of the British Standard). This study also evaluated TEMPO Ò EB in comparison with traditional pour plate and surface spread techniques using selective agar (British Standards Institution, 2004b), because the ability to replace all traditional techniques for the enumeration of Enterobacteriaceae with the TEMPO Ò EB system would allow the laboratory to produce more rapid results whilst making savings in both labour and media costs. In the first stage of the study, pure cultures were used to determine the sensitivity and specificity of the TEMPO Ò test. These studies indicated that Enterobacteriaceae were reliably detected with a detection limit of below 10 CFU ml )1 whilst non-enterobacteriaceae strains gave negative results at all inoculum levels tested (Table 1). Comparison of TEMPO Ò with traditional methods for the enumeration of Enterobacteriaceae in artificially spiked and naturally contaminated samples indicated that there was no statistical difference between TEMPO Ò results and the three alternative methods (namely the MPN method, BS ISO 21528:1:2004; pour plate method, BS ISO 21528:2:2004; and the surface spread plate method based on BS ISO 21528:2:2004). These results are in agreement with the findings of Paulsen et al. (2008), who demonstrated that TEMPO Ò EB produced equivalent results to BS ISO :2004. In European legislation (European Commission, 2007), a lower action threshold of <1 per ml is specified for certain dairy products, above which the presence of Enterobacteriaceae is considered to warrant investigation. These product types include pasteurized liquid dairy products such as milk, cream, yoghurt and fromage frais. To achieve this low detection limit, the product must be tested neat or undiluted. Products such as cream and yoghurt are often very viscous and cannot be readily pipetted into the TEMPO Ò reagent bottle without dilution. Therefore, for these sample types, a 1 in 2 dilution was prepared and duplicate TEMPO Ò cards inoculated so that a total of 1 ml of product was examined. Although results produced by TEMPO Ò were comparable with traditional methods for the majority of sample types, occasional anomalous results were obtained for products such as peppercorns and ground pepper or products containing a large amount of spice or pepper (for example spiced dried meat). In these instances, false positive reactions in every tube within the MPN card resulted in the reporting of an Enterobacteriaceae count exceeding the upper detection limit of the method (usually greater than CFU g )1 ). The reason for this was unclear, although the possibility of blockage of the filling tube (leading to a lack of fluorescence, and therefore an apparent positive result, in every well) was excluded by means of a visual inspection of each card after filling. Another possible reason for results appearing to exceed the detection limit for products such as pepper or spice could be a substance inherent to the product which is interfering with the reaction and quenching the fluorescence. One way to overcome this effect is to dilute the product further before testing. Alternatively, the TEMPO card may be examined visually under a UV light before incubation. Immediately before incubation, all tubes in the card should be fluorescent so if any tubes are negative for fluorescence at this stage, there may be another substance present which has the effect of quenching the fluorescence. Some highly coloured products also led to false positive results in our study as the colour of the product interfered with the reaction. This was mainly observed in products that were tested in their undiluted state to achieve a low detection limit (e.g. chocolate milk and chocolate fromage frais). It is therefore recommended that traditional test methods be used in place of TEMPO Ò when examining these products for the presence absence of Enterobacteriaceae. Although the manufacturers of TEMPO Ò EB cite acidic products such as fruit juice and intensely coloured mushrooms as further examples of unsuitable food types for this technique, samples of these product types tested in our study showed comparable results with traditional techniques. This is likely to be because of the fact that the preparation of a 1 in 10 dilution of the food samples before inoculation of the TEMPO Ò card resulted in a Journal of Applied Microbiology 109, ª 2010 The Society for Applied Microbiology 1815

7 Evaluation of TEMPO â Enterobacteriaceae method M. Owen et al. diluting effect on the colour acidity that might otherwise interfere with the fluorescence reaction. The most significant advantage of the TEMPO Ò method was the reduced time taken to obtain a result when compared to traditional methods. Moreover, the staff time involved in processing samples was greatly reduced compared to the British Standard MPN method (and similar to that required for plating methods). The lack of confirmation steps required by TEMPO Ò leads to further savings on staff time and media costs. Media preparation time is also reduced, as is the cost of transporting large quantities of liquid media in glass tubes from the media production site to the laboratory. Given its many advantages, and the equivalence of results compared to standard methods, it is considered that TEMPO Ò EB is a suitable method for the analysis of routine food, dairy and environmental samples (with a small number of exceptions, such as chocolate-flavoured dairy products and products containing a high proportion of pepper). The manufacturer, BioMérieux, produces TEMPO Ò kits for the detection of a variety of organisms in addition to Enterobacteriaceae (including Total Viable Count, Total Coliforms, E. coli, Coagulase-positive Staphylococci, Lactic Acid Bacteria, Yeasts and Moulds). Once the TEMPO Ò hardware has been installed in the routine laboratory, it may prove beneficial to evaluate this methodology for additional microbiological test parameters in the future. Acknowledgements The authors would like to thank BioMérieux UK for the provision of equipment and advice during the course of the study. References British Standards Institution (2004a) BS ISO :2004. Microbiology of Food and Animal Feedingstuffs Horizontal Methods for the Detection and Enumeration of Enterobacteriaceae Part 1: Detection and Enumeration by MPN Technique with Pre enrichment. London: BSI. British Standards Institution (2004b) BS ISO :2004. Microbiology of Food and Animal Feedingstuffs Horizontal Methods for the Detection and Enumeration of Enterobacteriaceae Part 2: Colony Count Method. London: BSI. Cosby, D.E. and Bailey, J.S. (2007) Comparison of the TEMPO Ò system, petrifilm Ò and cultural MPN procedure for enumeration of E. coli, coliforms and total aerobic plate counts. Int Assoc Food Prot 161, 2 58 [Abstract]. European Commission (2007) Regulation (EC) No of 5 December 2007 amending Regulation (EC) No on microbiological criteria for foodstuffs. Official J Eur Union L 322, Health Protection Agency (2004) Enumeration of Enterobacteriaceae by the colony count technique. National Standard Method F23; org.uk/pdf_sops.asp. Health Protection Agency (2005) Detection and enumeration of Enterobacteriaceae. National Standard Method F18; Health Protection Agency (2009) Guidelines for Assessing the Microbiological Safety of Ready-to-Eat Foods Placed on the Market. HPAwebStandard/HPAweb_C/ ? p= ICMSF (1988) Micro-organisms in Foods 1: Their Significance and Methods of Enumeration, 2nd edn. Toronto: University of Toronto Press. Kunicka, A. (2007) Evaluation of the TEMPO Ò system: an automated method for food microbiological quality control. J Biotechnol 131, S69 S72. Paulsen, P., Borgetti, C., Schopf, E. and Smulders, F.J.M. (2008) Enumeration of Enterobacteriaceae in various foods with a new automated most-probable-number method compared with petrifilm and international organization for standardization procedures. J Food Prot 71(2), PHLS (2000) Guidelines for the microbiological quality of some ready-to-eat foods sampled at the point of sale. Commun Dis Public Health 3, Torlak, E., Akan, I.M. and Gökmen, M. (2008) Comparison of TEMPO Ò EC and TBX medium for the enumeration of Escherichia coli in cheese. Lett Appl Microbiol 47, Journal of Applied Microbiology 109, ª 2010 The Society for Applied Microbiology