Sampling and testing

For many years, microbiology in the brewery has suffered a bad press. Of its many perceived failings, Quain (1999) focused on its slowness and the equivocal nature of microbiological results. The former consideration is the most damning. Methods that require between three and seven days to deliver a result can offer little to monitoring and process control. Such a lack of real time delivery has undermined the importance of microbiology and reduced it to a 'comfort' analysis that is tracked via charts of compressed and averaged data. The second point ('What do the results mean?') has triggered countless furrowed brows and sleepless nights. Microbiologists - who are frequently seen as bringers of bad tidings - are left to explain to harassed team leaders or departmental managers that they are unable to predict the significance of the (bad!)

results they bring. This is invariably exacerbated by the 'historical' nature of the results ('The beer has gone to trade.') and the uncertainty around the details of the result ('Is it a beer spoiler? 'Is there enough to spoil?'). If this is not enough, lurking beneath such a debate are concerns about the general robustness of brewing microbiology. In particular, the 'quality' of the sample and, particularly, just how representative it is of the vessel it was taken from. For those with the 'knowledge' (often ex-microbiologists), further uncertainty can be introduced by questioning the ability of the media to support the exclusive growth of 'beer spoilers'. Failing these arguments, there is always the indefensible review of the occasions when routine testing failed to flag a spoilage problem and, conversely, the time when the results anticipated a problem which failed to materialise!

Ironically, despite the above pessimism, brewing microbiology has been re-energised through the enhanced profile and importance of 'hygiene' (see Section 8.2.1) together with spin-off developments from the wider food industry. This later point reflects the high-profile food scares that have triggered public concern about micro-organisms in general and pathogens in particular! Consequently, there has been an explosion of interest and funding of initiatives to validate cleaning in real time, and to improve sampling, testing and identification. These have all to a lesser or greater extent had a positive impact on brewing microbiology.

8.3.1 Philosophy

The approach to routine microbiological sampling and testing has fundamentally changed in recent years. Although, unlike analytical testing, still very much the responsibility of 'specialists', the focus of microbiology has changed fundamentally from 'blanket' coverage of the process to more specialised, targeted testing. This recognises the industry-wide reduction in resources and the need to focus testing in the most appropriate areas.

8.3.1.1 Inspection (QC) versus prevention (QA). The most fundamental change in philosophy has been the move from 'control' to 'assurance'. Accordingly, many breweries have evolved their quality control departments (QC) into quality assurance (QA) departments. This is much more than a fashionable, cosmetic change but a change in emphasis from 'inspection' to 'prevention'. Driven by the implementation of 'systems' (HACCP, ISO 9000), microbiological sampling and testing is now focused on the process rather than the product. The basic premise is that microbiological effort should be directed at the 'critical control points' of the process which together determine product quality.

8.3.1.2 Sampling plans and specifications. Sampling plans and the associated specifications make brewing microbiology happen! Typically, a sampling plan will be highly prescriptive and detail the inevitable (sample, frequency) as well the method and sample volume. The size of the sample is important and, crucially is often too small to realistically achieve occasional counts of micro-organisms. It is all too easy to be deluded into believing that a CCP is in control because of routine zero counts. Accordingly, the sample volume should be stretched so that microbial counts are occasionally observed. This confirms that the sample 'window' is appropriate and gives the assurance that changes in performance can be tracked. Further, it facilitates useful comparison, or benchmarking, of performance against other breweries. Sampling plans and specifications require periodic, at least annual, review to ensure they remain appropriate, focused and responsive to improvement initiatives.

Given the inevitable historical nature of the results, the 'specifications' are invariably a 'target' and, unlike measurements of colour, ABV or C02 etc., have little authority or impact on the process. Results are tracked to varying degrees of statistical sophistication and regularly reviewed, thereby enabling changes in performance to be readily identified and responded to.

Despite the importance of microbiological sampling plans and specifications, surprisingly little has been published. In a relatively short report for such a big subject, Avis (1990) described the application of HACCP to microbiological process control. Usefully he described CCPs and associated principal sample points (PSPs) which are used to monitor the process. This paper concluded with an example of the then sampling plan for a keg line (see Table 8.15). More recently, Alan Kennedy (Anonymous, 1999) described the minimum sampling plan used by the Scottish Courage Group. (In passing, sampling plans are invariably described as being the 'minimum', a rider that is often forgotten in the translation into practical reality!) Returning to 'specifications', Kennedy usefully described the use of red/amber/green descriptors to 'weight' microbiological results according to micro-organism and process location. In another example, Ichikawa and Takinami (1992) described their sampling plan to meet the more exacting demands of sterile filtration and aseptic packaging.

8.3.2 Methods - 'traditional'

Despite the very real developments in 'rapid' methods (see Section 8.3.3), traditional methods continue to be used for the vast bulk of microbiological activities in breweries large and small. These methods have remained relatively unchanged since the days of Louis Pasteur (Ogden, 1993). Typically, these methods revolve around the 'plating out' of a sample, either directly (or after centrifugation) as 'spread plates' (<0.2 ml), 'pour' plates (5 ml) or after membrane filtration (<500 ml). These approaches differ fundamentally. With the spread plate and membrane filtration approaches, the micro-organisms grow on the surface of the media. With the pour plate technique the sample is added directly to the hot molten media, this then cools and solidifies. Accordingly, the micro-organisms are distributed throughout the media which, together with concerns about the effects of heat shock, has led many microbiologists to stop using pour plates. Irrespective of the approach, after three to seven days' incubation at typically 27°C any microbial colonies that are present are counted and, where deemed necessary, identified. Identification is often 'intuitive' from the selective nature of the media or limited to microscopic examination as 'bacterial rods or cocci' or 'yeast'. More detailed traditional analysis includes the Gram test and catalase test (Section 8.1.2) for bacteria and, as a last resort, diagnostic strip tests (e.g. API, Biolog etc.) for yeast (Section 4.1.1.1) and bacteria (Section 8.1.2).

Table 8.15 Minimum sampling plan for

a keg line (after Avis, 1990).

Brew Your Own Beer

Brew Your Own Beer

Discover How To Become Your Own Brew Master, With Brew Your Own Beer. It takes more than a recipe to make a great beer. Just using the right ingredients doesn't mean your beer will taste like it was meant to. Most of the time it’s the way a beer is made and served that makes it either an exceptional beer or one that gets dumped into the nearest flower pot.

Get My Free Ebook


Post a comment