Enumeration of Escherichia coli in swab samples from pre- and post-chilled pork and lamb carcasses using 3M™ Petrifilm™ Select E. coli and Simplate® Coliforms/E. coli.

The aim of the study was to compare two analytical methods; 3M Petrifilm™ Select E. coli and SimPlate® Coliforms &E. coli, for detection and enumeration of E. coli using swab samples from naturally contaminated pork and lamb carcasses that were collected before and after chilling. Blast chilling was used for pork carcasses. Swab samples (n=180) were collected from 60 warm and 60 chilled pork carcasses, and 30 warm and 30 chilled lamb carcasses, and analysed in parallel. The concordance correlation coefficient between Petrifilm and SimPlate was 0.89 for pork and 0.81 for lamb carcasses. However, the correlation was higher for warm carcasses (0.90) than chilled carcasses (0.72). For chilled lamb carcasses, the correlation was only 0.50, and SimPlate gave slightly higher results than Petrifilm (P=0.09). Slower chilling gave slightly lesser agreement between methods than for blast chilling, however, both Petrifilm and SimPlate methodologies are suitable and recommended for use in small laboratories in abattoirs.

The significance of clean and dirty animals for bacterial dynamics along the beef chain.

This study investigated the bacterial dynamics along the beef chain for clean and dirty cattle in the slaughter and processing lines, using classic quantitative methods and molecular analyses. In addition, the Norwegian national guidelines for Good Hygiene Practices in Norway were evaluated. In these guidelines, cattle presented for slaughter are categorised according to hide cleanliness, resulting in separate processing lines for meat from very dirty animals and reduced prices to farmers. The study was conducted in two commercial abattoirs in Norway. Two groups were compared; 40 visually clean cattle and 40 visually dirty cattle presented for slaughter, with 20 from each group at each abattoir. The same animals were sampled at five sampling sites: hides, carcass surfaces after dehiding, just before chilling, after chilling, and meat trimmings. Meat trimmings were sampled in only one abattoir. Three hundred and sixty samples were collected by swabbing 100 cm(2) of the brisket area at the first four sampling sites, and sampling 200 g of meat trimmings at the fifth site. The results showed that the hides of dirty cattle had more Enterobacteriaceae and higher Aerobic Plate Counts (APC) than visually clean cattle (P<0.05), however there was no significant difference for Escherichia coli. For the other sampling sites, there were no differences between the dirty and the clean group. An effect of chilling/drying of the carcass surfaces was demonstrated by the significant reduction in the number of carcasses on which E. coli and Enterobacteriaceae were detected; from 11% and 39% before chilling to 1% and 16% after chilling, respectively. Enterobacteriaceae and E. coli were detected in only three and one of the meat trimming samples, respectively. Amplification and sequencing of the 16S rRNA gene from 643 Enterobacteriaceae colonies derived from 107 samples demonstrated that Escherichia/Shigella were dominant within this family on the hides. However, after dehiding, after grading, and after chilling, the genera Citrobacter and Enterobacter dominated. The meat trimmings were dominated by the genera Kluyvera, Hafnia, and unclassified Enterobacteriaceae. The relative proportions of Escherichia/Shigella were higher for dirty animals than for clean animals, and were higher on hides than from sampling sites further down the chain (P<0.05). The minor differences in contamination on carcass surfaces and meat trimmings between clean and dirty cattle indicate that separate processing lines in Norwegian abattoirs seem to be unnecessary.

Effects of hygienic treatments during slaughtering on microbial dynamics and contamination of sheep meat.

The aims of this study were to investigate bacterial dynamics in the sheep meat chain, from fleece to meat trimmings, using both quantitative and qualitative analyses, and to study the effects on microbial load associated with the hygienic interventions of: i) shearing sheep immediately before slaughter, ii) manual steam vacuum pasteurisation, iii) hot water pasteurisation of carcasses, followed by iv) chilling. A further aim was to provide evidence to determine whether or not unshorn sheep should be handled in a processing line separate from that of shorn sheep in Norwegian abattoirs. A total of 176 surface swab samples were collected from three sites along the value chain: i) on fleeces, ii) on carcasses at the end of the slaughter line, and iii) on carcasses after chilling for 24h, and 32 samples were collected from meat trimmings. The results showed that Aerobic Plate Counts (APC) were lower for the shorn group compared to the unshorn group, both on carcasses before chilling and after chilling (difference of 0.8 and 0.9logCFU/1000cm(2) (p≤0.05), respectively) and in meat trimmings (difference of 0.5logCFU/g (p≤0.05)). Hygienic treatments were used on carcasses derived from unshorn sheep, and steam vacuum treatment reduced Escherichia coli, Enterobacteriaceae, and APC before chilling by 1.2, 1.0, and 0.6logCFU/1000cm(2) (p≤0.05), respectively, and hot water pasteurisation, in addition to chilling, reduced E. coli, Enterobacteriaceae, and APC by 0.7, 1.0, and 0.9logCFU/1000cm(2) (p≤0.05), respectively, compared with untreated carcasses. The effect of chilling was shown by the significant reduction of number of carcasses where E. coli were detected; from 65% (13/20) of the shorn group before chilling to 35% (7/20) after chilling, and from 90% (36/40) to 45% (9/20) of the unshorn group. Sequencing of the 16S rRNA gene derived from 316 colonies of Enterobacteriaceae showed a tendency for the relative proportion of the genus Escherichia/Shigella, compared with other genera within Enterobacteriaceae, to be greater for unshorn, untreated sheep than from the other groups at the sampling locations along the meat chain. The study showed that steam vacuum and hot water pasteurisation reduced the contamination of carcasses derived from unshorn sheep, down to the level of the shorn group, and thus can replace the separate processing line for unshorn sheep. Indeed, the low microbial contamination in meat trimmings for all groups indicates that the separate processing line is unnecessary.

Impact of rainfall on the hygienic quality of blue mussels and water in urban areas in the Inner Oslofjord, Norway.

The effects of precipitation on the hygienic quality of water and blue mussels collected from five different localities in the urban areas in the Inner Oslofjord were investigated, with samples analysed for Escherichia coli, Salmonella spp., pathogenic Vibrio spp., Norovirus, Sapovirus, Cryptosporidium spp. and Giardia duodenalis. The sampling sites were located at varying distances from the outlet of combined sewer overflows (CSO)-impacted rivers/streams. In general, 1-3 log10 increases in fecal indicator bacteria and human pathogens were observed after heavy rainfalls. Blue mussels appeared to be a useful indicator of the impact of sewage at these sites, and generally a good correlation was identified between concentrations of E. coli and other human pathogens in the mussels. Provision of general advice to the public of avoiding areas near the outlets of CSO-impacted rivers after heavy rainfall may reduce the risk of gastroenteritis by bathers and others that may swallow water during recreational activities.

Effect of water treatment on the growth potential of Vibrio cholerae and Vibrio parahaemolyticus in seawater.

In laboratory experiments we added Vibrio cholerae and Vibrio parahaemolyticus to bottles with seawater previously treated by filtration, UV, chlorine or ozone. The purpose was to investigate the influence of different treatment techniques on the growth potential of these bacteria in simulated ballast water tanks. Residual oxidants were removed before inoculation, and the bottles were incubated at 21 ± 1 °C. The growth potential of the vibrios was investigated in two different experimental setups, i.e. in presence and absence of added natural microorganisms. In general, V. cholerae and V. parahaemolyticus rapidly lost their culturability after inoculation and storage in untreated seawater, but showed increased survival or growth in the treated water. Highest growth was observed in water previously exposed to high concentrations of ozone. Addition of natural microorganisms reduced the growth of V. cholerae and V. parahaemolyticus.

Evaluation of the SimPlate method for enumeration of Escherichia coli in swab samples from beef and lamb carcasses.

Abattoirs have to enumerate Escherichia coli on carcass surfaces as part of compulsory HACCP monitoring and they therefore need rapid and reliable methods. The objective of this study was to compare a conventional plating method with a faster, simpler method for detection and enumeration of E. coli in samples from naturally contaminated carcasses. The two methods were the conventional pour plate method of the Nordic Committee on Food Analysis; NMKL Method No. 125, and the enzymatic method of SimPlate Coliforms &E. coli. Materials were 588 cotton-cloth samples used for swabbing 100 cm(2) areas on four sites on cattle and lamb carcasses in three commercial abattoirs in Norway. E. coli was detected by at least one of the methods in 270 (46%) of the samples. Forty-five samples (8%) were positive only by SimPlate while 28 samples (5%) were positive only by NMKL125. Cohen's kappa value was 0.74 for detection/non-detection results, which showed a high level of agreement between the two methods. E. coli counts determined by the conventional NMKL125 method showed a high concordance correlation (ccc 0.80, slope 0.99) with most probable number (MPN) values obtained with SimPlate. SimPlate is a rapid and reliable method for detection and enumeration of E. coli and a suitable alternative method for use with swab samples from cattle and lamb carcasses.

Microcystin poisoning in roe deer (Capreolus capreolus).

Acute cyanobacterial hepatotoxicosis in a wild roe deer (Capreolus capreolus) from Norway is reported. The diagnosis was based upon the demonstration of typical liver lesions and high liver concentrations of microcystins. The liver was markedly enlarged and histopathological examination revealed diffuse hepatocellular dissociation, degeneration and necrosis and perisinusoidal haemorrhage. Liquid chromatography-mass spectrometry demonstrated the presence of 1361 ng microcystin-YR, -LR and -RR per gram liver (wet weight). This is believed to be the first report of cyanobacterial intoxication in wild mammalian species as confirmed by demonstration of high toxin levels in the animal's tissues.

Occurrence of Vibrio parahaemolyticus, V. cholerae, and V. vulnificus in Norwegian Blue Mussels (Mytilus edulis).

Vibrio parahaemolyticus, V. cholerae, and V. vulnificus were isolated from 10.3%, 1.0%, and 0.1% of 885 blue mussel samples, respectively. Four of the samples contained trh(+) V. parahaemolyticus, while no tdh-positive isolates were detected. The V. cholerae isolates were non-O:1/non-O:139 serotypes and were ctxA negative.

An evaluation of sampling- and culturing methods in the Norwegian action plan against Campylobacter in broilers.

The Norwegian Action Plan against Campylobacter in broilers was implemented in May 2001 with the objective of reducing human exposure to Campylobacter through Norwegian broilers. From each flock, samples collected at the farm about one week prior to slaughter, and then again at the slaughter plant, are examined for the presence of Campylobacter. All farmers with positive flocks are followed up with bio-security advice. Sampling of broiler products at retail level is also included in the Action Plan. The aim of this study was to evaluate the existing sampling and culturing methods of the Norwegian Action Plan against Campylobacter in broilers. The material collected was pooled faecal samples, pooled cloacae samples and caecae samples from individuals. The highest number of positives, from culturing of the pooled faecal samples, the pooled cloacae swabs and the caecae swabs from individuals, were obtained at incubation temperature 41.5 degrees C. When comparing the results at incubation temperature 37 and 41.5 degrees C, the faecal samples from the farms demonstrated a high concordance, with a kappa value of 0.88. The results from culturing cloacae swabs and caecae samples from slaughter plant level at two temperatures did not agree very well with a kappa value of 0.21 and moderate value of 0.57, respectively, but were both disconcordant at a level of 0.05. Modelling farm level data indicated that if increasing the number of pooled samples per flock from two (in existing regime) to three, the flock sensitivity increases from 89% to 95%. Modelling of slaughter plant data indicated that three pooled cloacae swabs are needed to identify 90% of the positive flocks. The results from the modelling of caecae data indicated that samples from seven individuals are sufficient to identify 90% of the positive flocks and caecae samples could thus be an alternative to cloacae sampling at slaughter plant level.

Survival of Campylobacter on frozen broiler carcasses as a function of time.

In the Norwegian Action Plan against Campylobacter in broilers, carcasses from flocks identified as positive before slaughter are either heat treated or frozen for 5 weeks to reduce the number of Campylobacter. The objective of this study was to estimate the effect of freezing time and predict the number of Campylobacter on naturally infected or contaminated broiler carcasses following freezing for 2, 4, 6, 8, 10, 13, 21, 35, and 120 days by nonparametric and parametric linear statistical models. From each of the five flocks, 27 carcasses were sampled. Each carcass was cut in two pieces along the chest bone. Half was put into the freezer (-20 degrees C), whereas the other was deskinned and quantitative culturing was conducted from a 10-g sample of the skin. Fifteen frozen halves were selected at random at each time point following freezing from 2 to 120 days, and skin samples from these were cultured quantitatively and qualitatively. In regard to the log reduction of Campylobacter, almost similar results were obtained using three statistical methods; median regression on the change in Campylobacter counts, zero-inflated negative binomial regression, and a Bayesian Markov chain Monte Carlo (decay) model on original counts. Overall, a 2-log reduction of Campylobacter was obtained after 3 weeks of freezing. Only a marginal extra effect was observed when extending the freezing to 5 weeks. Although freezing appears to be an efficient way to reduce the level of Campylobacter on broiler carcasses, in 80% of the carcasses Campylobacter could still be detected using quantitative culturing following 120 days of freezing. Based on the high number of zeros, these data should be modeled by a zero-inflated model. The best statistical fit in regard to goodness-of-fit measures was the zero-inflated negative binomial log link model, closely followed by the Poisson model. Thus, in our continued search for a better way to describe the data, we used the Poisson distribution in the mixed Bayesian decay models.