| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Instituto de Investigación y Análisis Alimentarios, Universidad de Santiago de Compostela, Facultad de Veterinaria, Campus Universitario s/n. 27002, Lugo, Spain
1 Corresponding author: jlrotero{at}usc.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Key Words: Listeria spp. • prevalence • risk factor • dairy farm
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Listeria monocytogenes is included in the World Health Organization list of foodborne pathogens (WHO, 2002), and is the most important Listeria species in terms of public health risk and frequency of appearance in foodstuffs (Fenlon and Wilson, 1989). Listeria seeligeri and L. ivanovii may be pathogenic in humans (Wiedmann, 2003).
Listeria monocytogenes is ubiquitously present in diverse environments including water, grass, silage, decomposing organic matter, soil, and feces (Hassan et al., 2001). It has a high capacity for persistence under adverse conditions, surviving at refrigeration temperatures and over a wide range of pH (Low and Donachie, 1997). Both humans and diverse wild and domestic animals may be asymptomatic carriers of L. monocytogenes, which is widely present in the feces of cattle, pigs, chickens, turkeys, ducks, crustaceans, and flies. Similarly, cattle may shed these bacteria in milk over long periods without showing any symptoms of disease (Low and Donachie, 1997). Listeria monocytogenes can be pathogenic for animals, and the development of clinical listeriosis was associated with stress (Cooper and Walker, 1998). The principal source of infection for ruminants is spoiled silage (Low and Donachie, 1997; Vázquez-Boland et al., 2001; Wiedmann, 2003), although it may come from contaminated water or feed contaminated by avian or insect vectors (Cooper and Walker, 1998).
Listeria monocytogenes may reach bulk tanks as a result of exogenous contamination via the milking equipment, because of fecal contamination during milking, or, less frequently, by an intramammary route following generalized asymptomatic infection or mastitis (Cooper and Walker, 1998; Hassan et al., 2001; Winter et al., 2004).
The aims of the present study were to determine the prevalence of Listeria species in on-farm bulk tanks, to identify management practices associated with the presence of Listeria species in bulk tanks, and to assess the possible influence of Listeria species on milk hygienic quality.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The mean number of animals per herd was 61.3 (8 to 480, SD = 67.1), and the mean number of lactating cows was 35.0 (4 to 260, SD = 39.3). Of the 98 farms, 41.2% had free-stall housing with a milking parlor, and the remaining 58.8% had tie-stall housing with pipeline (43.3%) or bucket (15.5%) milking.
Data Collection: Questionnaire
Each farmer was interviewed on the farm at the time of sample collection and pH determination using a questionnaire about the following farm characteristics and aspects of management; however, some of these aspects were observed directly:
Somatic cell counts and bacterial counts (BC) in bulk-tank milk recorded for each farm as assessed monthly by the official dairy laboratory of the region.
Sample Collection and pH Determination
On each farm the following samples were collected in sterile containers following aseptic sampling procedures. 1) Bulk-tank milk was obtained at the upper opening of the tank after several minutes of stirring to ensure homogeneity; 2) pooled feces from 3 randomly selected, apparently healthy, lactating cows was taken directly from the rectum; 3) silage, both grass and corn, was sampled from the middle and lower parts of the silage storage container (bunker, trench, or bag).
All samples (98 milk samples, 97 fecal samples, and 83 silage samples) were transported to the laboratory under refrigeration, and all analyses were performed within 24 h of sampling.
In addition, silage pH was determined at 3 locations in the silage container (upper, middle, lower) using a portable pH meter with penetration electrode (Hanna Instruments, Woonsocket, RI) with an accuracy of ± 0.02 pH units.
Isolation and Identification of Listeria Species
Listeria spp. were determined as described by Menéndez et al. (1997). Twenty-five grams of sample was transferred to a flask containing 225 mL of FDA Listeria enrichment broth (Merck, Darmstadt, Germany) and incubated for 48 h at 30°C. After enrichment, samples were plated onto PALCAM (polymixin B sulfate, acriflavine HCl, lithium chloride, ceftazidime, aesculin medium) agar (Merck) and incubated for 72 h at 37°C. Morphologically typical black colonies (tellurite-positive) with a black halo (esculin hydrolysis) were confirmed as Listeria on the basis of the catalase reaction with 3% H2O2, motility at 25°C, and sheep blood hemolysis, using the overlay technique and the API Listeria kit (bioMérieux, Marcy l’Etoile, France).
Data Analysis
Data were analyzed using Excel (Microsoft Corp., Redmond, WA) and the statistical software package SPSS 12.0 (SPSS Inc., Chicago, IL) with the aim of detecting differences between farms with Listeria spp. (or L. monocytogenes) in milk, and farms without Listeria in milk. Categorical variables were compared by 
2 tests and quantitative variables by Student-Fischer t-tests (normally distributed data as indicated by Shapiro-Wilk test, P 
0.05) or Mann-Whitney U-tests (nonnormally distributed data). For identification of risk factors, odds ratios (OR) and corresponding confidence intervals (CI) were calculated. Statistical significance was indicated by P 
0.05 (Doménech, 2003).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
), a greater prevalence than the previously reported 4.1% in Finland (Husu, 1990), 8.3% in Ireland (Rea et al., 1992), 10.4% in the United States (Van Kessel et al., 2004), and 12.4% in Canada (Farber et al., 1988).
|
Analysis of bulk-tank milk is important in view of the ability of Listeria spp. to survive at refrigeration temperatures and because of the techniques used to manufacture certain products from raw milk. For example, L. monocytogenes was detected in 6.4% (Rudolf and Scherer, 2001) and 4.9% (Pak et al., 2002) of samples of cheese produced from raw milk; in the former study, L. innocua was detected in 10.6% of samples and L. seeligeri in 1.2% (Rudolf and Scherer, 2001). In addition, Listeria-contaminated milk constitutes a risk for industrial dairy plants, for which raw milk is one of the principal sources of contamination (Menéndez et al., 1997).
Listeria Prevalence in Silage and Feces
Listeria spp. were detected in 32.2% of the 180 samples of silage and feces. This percentage is lower than the 77.8% obtained by Sanaa et al. (1996), but similar to the 30.6% obtained by Husu (1990). These values confirm the importance of environmental sources of Listeria spp. in dairy farms. In 7.2% of silage samples and 3.1% of fecal samples, more than 1 species of Listeria was detected. Reported prevalences of Listeria spp. in feces of healthy animals vary widely, apparently because of differences in sampling procedures and analytical methods (Husu, 1990).
Table 1
shows the prevalence of different Listeria spp. in milk, feces, and silage. Listeria spp. were detected in 41.2% of the 97 fecal samples analyzed, greater than the reported 8% in Sweden (Unnerstad et al., 2000) and 9.6% in Finland (Husu, 1990), but lower than the 81% from France (Sanaa et al., 1996). The presence of Listeria spp. in the feces of clinically healthy animals suggests that feces from such animals may be a source of contamination of milk (Unnerstad et al., 2000). We detected Listeria innocua in 22.7% of the samples, a greater prevalence than the 2% obtained in Sweden (Unnerstad et al., 2000) and the 4.6% in Finland (Husu, 1990). Listeria monocytogenes was present in 9.3% of our samples, similar to the 6% obtained by Unnerstad et al. (2000) and 6.7% obtained by Husu (1990), but lower than the value of 39% obtained by Sanaa et al. (1996). Listeria welshimeri was present in 4.1% of the samples, greater than the 0.2% obtained by Husu (1990). In addition, we detected L. grayi in 4.1% of our samples and L. ivanovii in 1.0%. Neither of these species was cited in the previous studies. Listeria seeligeri was not detected in fecal samples, as in previous studies (Unnerstad et al., 2000); Husu (1990) did detect this species, but only in 0.1% of samples.
We detected Listeria spp. in 33.7% of silage samples, a value intermediate between 22.7% in Finland (Husu, 1990) and 62% in France (Sanaa et al., 1996). Listeria monocytogenes was present in 6.0% of our silage samples, greater than the 3% obtained by Rea et al. (1992) in Ireland, but lower than the values of 16% obtained in Finland (Husu, 1990) and 39% in France (Sanaa et al., 1996). Listeria innocua was detected in 19.3% of our silage samples, greater than the 6% obtained by Rea et al. (1992). Other Listeria species were detected in silage with lower prevalence: 4.8% for L. welshimeri, 2.4% for L. grayi, and 1.2% for L. seeligeri.
The pH of silage samples from which Listeria spp. were isolated ranged from 4.47 to 6.97, a different range from those observed in other studies: 3.8 to 5.2 in Rea et al. (1992) and 5.78 to 5.89 in Ryser et al. (1997). The range obtained in the present study was more consistent with the lower pH limit of 4.5 for multiplication of Listeria spp. (Sanaa et al., 1993). This pH cut-off was selected in view of previous reports indicating that insufficient acidification of silage may favor the growth of Listeria spp., notably if pH is >4.0 (Sanaa and Ménard, 1994; Ryser et al., 1997) or 5.5 (Low and Donachie, 1997). The presence of Listeria spp. in silage was associated (P = 0.05) with silage pH (Table 2): when the pH was 4.5, the risk of presence of Listeria was increased (OR = 6.28). Silage storage method (concrete bunker silo, unlined trench, bagging) did not affect the presence of Listeria spp. (Table 2).
|
shows associations between the presence of Listeria spp. in milk, silage, and feces. In previous studies the proportion of farms with Listeria spp. present in all 3 sample types was as high as 50% (Rea et al., 1992). According to Sanaa and Ménard (1994), the presence of L. monocytogenes in silage, feces, or both increases the risk of presence in milk by a factor of 20, a finding we could not support.
|
). Similarly to Husu (1990), we found no association (P = 0.92) between silage consumption and Listeria contamination of the feces of clinically healthy animals, although Ryser et al. (1997) found a similar association.
Risk Factors for the Presence of Listeria Species in Milk
Table 4 summarizes the results of associations between potential risk factors and the presence of Listeria spp. in milk. However, as pointed out by Hassan et al. (2001), interpretation of the effect of a given variable on the presence of Listeria should be based not solely on the presence or absence of a statistically significant association, but also on an understanding of the microorganism’s biology.
|
Protecting the milk from environmental contamination in stall milking systems is more difficult than in parlor milking systems, so that BC tends to be greater (Hassan et al., 2001), as was the case in the present study (pipeline average BC: 48 x 103 cfu/mL vs. parlor average: 32 x 103 cfu/mL). The risk of milk contamination by Listeria spp. was almost 3-fold greater on farms having a pipeline stall milking system (P = 0.01, OR = 2.87) than farms with parlor milking systems, as in previous studies (Rohrbach et al., 1992). Only 40.2% of farmers correctly identified cows with mastitis, and on these farms the probability of Listeria spp. in milk was lower (OR = 0.44; P = 0.17), but not different from farms with farmers who could not identify cows with mastitis. Although Listeria spp. are isolated with rather low prevalence from the mammary glands (i.e., they are infrequently associated with mastitis; Rea et al., 1992; Rohrbach et al., 1992; Sanaa et al., 1996), the risk of milk contamination from cows with mastitis due to Listeria exists. In addition, following an appropriate order of milking reduces the likelihood of bacterial transmission between animals and of contamination of the milking equipment, and thus, reduces the probability of appearance of Listeria spp. in the bulk tank. The risk of contamination of milk by Listeria was greater when a correct milking order was not followed (P = 0.01, OR = 0.26).
Association Among Listeria Contamination, SCC, and BC
Most of the farms had milk SCC <400 x 103 cells/mL (83.9%), with 34.5% of farms having <200 x 103 cell/mL. On 96.5% of farms, BC was <100 x 103 cfu/mL, and on 42.4% of farms <20 x 103 cfu/mL. Somatic cell counts and BC were not different (P > 0.64) in milk with Listeria than in milk without (SCC: 252 x 103 vs. 250 x 103 cells/mL; BC: 29 x 103 vs. 26 x 103 cfu/mL).
Considering SCC as a categorical variable (<200 x 103 cells/mL, 200 to 400 x 103 cells/mL, and >400 x 103 cells/mL), the prevalence of Listeria spp. in milk did not vary among the 3 categories (P = 0.32), as shown in other studies (Rohrbach et al., 1992; Hassan et al., 2000; Van Kessel et al., 2004). When BC was considered as a categorical variable (<20 x 103 cfu/mL, 20 to 100 x 103 cfu/mL, and >100 x 103 cfu/mL), the presence of Listeria spp. in milk again showed no significant variation among the 3 categories.
In conclusion, correct practices with respect to silage production and milking are essential for preventing introduction of Listeria into the herd, its spread within the herd, and its entry into milk. More studies are needed to clarify the significance of these risk factors, because only 1 sample was collected at each farm in the current study. The risk of contamination of milk by Listeria spp. increased when animals were fed low-quality silage, notably silage with pH 4.5. The prevalence of Listeria species and L. monocytogenes in bulk-tank milk was similar to that reported in previous studies.
Received for publication March 21, 2007. Accepted for publication July 31, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
2, and UD 11. Comparación de dos medias: Pruebas t. Signo, Barcelona, Spain.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
