View Report Details

Review of raw milk analyses methods and assessment of effectiveness as pathogen makers and indicators of farm hygiene

Project Code: B12002

31/08/2004

Direct Laboratories Ltd
Hutchison, M

Background and Objectives

Although milk is heat treated by pasteurisation to remove pathogenic microorganisms before sale, it is still important that it is produced, collected and stored in a hygienic manner. Prevention of milk contamination is safest for consumers because pathogenic microorganisms such as E. coli O157 can produce toxins that accumulate in milk and are subsequently unaffected by pasteurisation. Furthermore, some classes of bacteria can survive pasteurisation by forming heat-resistant spores (e.g.bacillus) or forming aggregations which shield cells in the centre of the clump from lethal temperatures (e.g. Mycobacterium avium subspecies paratuberculosis). From a commercial and logistical viewpoint, milk which is not produced hygienically or which is not stored appropriately has a reduced shelf life. For these reasons, milk produced on UK farms is subject to microbiological examination by dairy companies as an indicator of its quality; and hence its commercial value. Most commonly, a count of total bacterial numbers (TVC) in milk is undertaken to assess its quality. In order to safeguard consumers, the Dairy Hygiene Inspectorate carries out periodic hygiene inspections of dairy farms on behalf of the Food Standards Agency. However when microbiological results are compared with hygiene audit scores there can be, in some cases, unclear correlations between total bacterial counts in milk and dairy hygiene inspectorate farm audit scores. This study was undertaken to assess if there are other bacterial groups which are better suited for indicating the levels of on-farm hygiene

Approaches

Task 1: Extensive review of pertinent scientific literature

An extensive 60-page review was prepared by collating and summarising all of the most important scientific and technical publications covering milk hygiene over the last 50 years. The report listed the bacteria commonly found in raw milk and their likely sources of contamination. Traditional and newly-reported microbiological methods in various stages of development were theoretically-assessed as potential indicators of on-farm hygiene practices. In addition, the suitability of the analytical methods for adoption by high throughput testing laboratories was determined. A conclusion of the report was that pathogenic microorganisms were not appropriate for use as hygiene indicators since most are not commonly detected in raw milk (a conclusion later supported by Task 2 results). However, for completeness, methods that can be used to detect the most commonly-encountered pathogens in milk were also reviewed.
The review recommended that a number of bacterial types be practically evaluated further as indicators of on-farm hygiene during the experimental phases of the project. TVC determined by a commercially available machine called a BactoScan was included to allow comparisons between this standard commercial test, and the proposed specific bacterial indicators.

Task 2: Measurement of the uncertainty of bacterial indicator levels and estimation of practical aspects of each methodology.

Traditional microbiological analyses protocols can be used to estimate bacterial levels in milk. This section of the study was concerned with determining how accurate each method was for estimating bacterial levels. The output gave information on the reliability of the results provided by each analyses method. The most accurate methods were those used to measure groups of bacteria that favour low temperatures and which are called psychrotrophs, TVC and a class of bacteria which is commonly found in fresh faeces called Bifidobacterium. The laboratory costs for each analysis were calculated to be around £3.00 per sample except for TVC by BactoScan which cost £1.20.

Task 3: Practical appraisal of bacterial indicators of on-farm hygiene

A comprehensive 60-question checklist covering cow cleanliness on entry to dairy parlours, general hygiene awareness and the practices of farm staff during milking, cleaning of equipment, parlour clean-down after milking, and the maintenance of livestock housing conditions was drawn up. A number of commercial dairy farms (24) were recruited to the study which had a history of consistent TVC results and which covered the full range from poor to excellent hygiene. Hygiene audits were carried out on these farms during summer 2003 and winter 2003/04. During the audits numerical scales were used to assess each of the five above areas. At the same time as the farms were audited, samples of raw milk were collected from farm bulk tank stores and analysed microbiologically for bacterial indicators. Raw milk from 100 randomly-selected farms was also analysed to determine prevalence of selected bacterial pathogens. For the summer audits, hygiene audit scores and bacterial indicator levels were analysed for statistically significant correlations. No statistically significant correlations were found between cow cleanliness on entry to the parlour or housing conditions and levels of any bacterial indicators. Total bacterial counts showed significant correlations with hygiene audit scores for milking practice and parlour cleaning. Psychotropic pseudomonads were the only bacterial group which were significantly correlated with hygiene audit scores. High hygiene standards during the milking routine made it more likely that high standards would be found for the plant cleanliness and parlour washing sections of the audit. Although some statisticallysignificant relationships were identified for the summer audits, none of these correlations were strong enough to allow firm conclusions to be drawn between the hygiene standards on individual farms and bacterial indicator levels in milk. The highest correlation observed for the winter audits was between the cleanliness of equipment in the parlour during milking and thermoduric bacteria which favour high temperatures. There was also a relationship between total bacterial counts and the effectiveness of parlour cleaning after milking had finished. As with the summer audits, none of the winter bacterial counts showed a strong enough correlation with the corresponding hygiene audits to allow valid assumptions to be made regarding hygiene standards at the single farm level.
When both the summer and winter audit datasets were merged together and analysed, relationships between hygiene audit scores and bacterial counts tended to decrease. This finding is important because it is strongly indicative that types and levels of bacteria in milk and their relationship with on farm hygiene practices may be different in summer and winter. The prevalence of Listeria monocytogenes in raw milk was 5% (5/100) and Mycobacterium avium subspecies paratuberculosis was 4% (4/100). Salmonella and E. coli O157 were not detected in any of the milk samples tested. Since pathogens are rarely isolated from raw milk, they and not suitable for use as hygiene indicators.

What it means and why it’s important

The bacterial indicators that were identified as potentially useful by this study were total bacterial counts, psychrotrophs, coliforms and pseudomonas. Only coliforms, which were related to milking plant cleanliness, had significant relationships which were preserved across both the summer and winter seasons. One of the difficulties encountered during this study was the high degree of variation in bacterial levels in milk which meant the relationships identified are not suitable for “snapshot” determinations of farm standards based on a single analysis. Total bacterial counts are currently used by dairy companies as farm hygiene and milk quality indicators. This study was unable to identify any alternative bacterial groups which provided a better measure of on-farm hygiene practices and the likelihood of pathogens being present in raw milk.

Main File

Some of the files on this site may be in a format that your computer can't read. However, you can download Readers and Viewers for the following document types below: