Dairy Science & Technology · Dairy Microbiology · also Microbiological Quality Control, Dairy Laboratory Practice
How dairy laboratories sample, count, detect and certify: criteria, indicator organisms, rapid methods and lab quality systems.
In the exams: ICAR AICE subject 36 Unit 6 and subject 35b Units 1 and 3; ICAR AIEEA-PG group 16 (enumeration, isolation and identification of pathogens)
Microbiological quality assurance is the part of dairy quality management that decides, from samples and tests, whether milk, products, processes and plant environments meet microbiological requirements. It rests on a chain of judgements: which microorganisms to look for and why, how many samples to take and how to interpret the results, which methods to use and how fast they must be, where to sample in the factory, how to keep laboratory staff and the plant safe from the organisms being tested, and how to prove that the laboratory's results can be trusted. This subject covers that chain: microbiological criteria and sampling plans, indicator organisms, enumeration methods, the detection of pathogens, rapid methods such as ATP bioluminescence, PCR and immunoassays, biosensors, environmental monitoring programmes, biosafety, and laboratory accreditation.
The subject matters because testing is expensive, slow and statistically limited, and because decisions based on it have consequences: a lot released on a false negative can cause illness, and a lot held on a false positive costs money. A microbiologist who understands what a sampling plan can and cannot detect, what an indicator count means, how a rapid method was validated and how to keep a laboratory under quality control can design testing that genuinely supports food safety rather than giving false reassurance.
The organisms are described in Microbiology of Raw and Market Milk, Microbiology of Dairy Products and Dairy Pathogens, Toxins and Microbial Risk; general methods, indicators and sampling statistics are introduced in Food Technology's Food Microbiology; method validation, reference materials, z-scores and ISO/IEC 17025 for chemical analysis appear in Dairy Analysis and Instrumental Methods; and cleaning and hygienic design in the subject on cleaning-in-place and hygienic design. This manual refers to them and concentrates on microbiological practice. It is examined in Unit 6 of ICAR AICE-JRF/SRF subject 36 (Dairy Microbiology: quality assurance and rapid methods) and Units 1 and 3 of subject 35b (Food Safety and Quality), and in group 16 of ICAR AIEEA-PG (enumeration, isolation and identification of pathogens). Expect questions on criteria, sampling plans and their probabilities, indicators, counting methods and calculations, reference methods for pathogens, the principles of rapid methods and biosensors, environmental zoning, biosafety levels and accreditation.
A microbiological criterion defines the acceptability of a product, a lot of food or a process, based on the absence, presence or number of microorganisms, or the quantity of their toxins or metabolites, per unit of mass, volume, area or lot. The Codex Principles and Guidelines for the Establishment and Application of Microbiological Criteria (CXG 21-1997, revised in 2013) state that a criterion consists of its purpose; the food, process or food safety control system to which it applies; the specific point in the food chain where it applies; the microorganism or toxin and the reason for choosing it; the microbiological limits; the sampling plan, with the number and size of sample units and the acceptance number; the analytical method; and the action to be taken when the criterion is not met. Criteria are used in three broad ways. Microbiological standards are mandatory criteria in laws or regulations. Guidelines are advisory criteria that indicate whether hygiene practices are working. Specifications are criteria agreed between buyers and suppliers in purchase contracts. The European Union regulation on microbiological criteria distinguishes food safety criteria, which define the acceptability of a product or batch placed on the market and concern pathogens such as Salmonella, Listeria monocytogenes and Cronobacter, from process hygiene criteria, which indicate the acceptable functioning of the production process and concern indicator organisms such as Enterobacteriaceae, Escherichia coli and coagulase-positive staphylococci. Examples for dairy products are given in Microbiology of Dairy Products.
Criteria should be based on scientific evidence and be achievable, meaningful and verifiable. Food safety criteria are ideally derived from risk assessment and from public health goals expressed as a food safety objective or performance objective, as described in Dairy Pathogens, Toxins and Microbial Risk: the limit, the sampling plan and the method are chosen so that lots exceeding the acceptable level of hazard are likely to be rejected. Process hygiene criteria are usually set from data on what well-run processes achieve, for example a limit that good plants meet most of the time, so that a result above it signals a process problem needing investigation rather than an unsafe product. Criteria should be applied only where they add value: testing a product for a hazard that the process reliably controls, or for which the sampling plan has little chance of detecting a problem, wastes resources. The action when a criterion is not met must be stated in advance: for a food safety criterion, rejection, reprocessing or recall of the lot; for a process hygiene criterion, corrective action on the process, such as improved hygiene or review of raw materials. Microbiological criteria are only one element of food safety management. They verify that good hygiene practice and hazard analysis and critical control point (HACCP) systems are working, but they cannot replace them, because no feasible amount of end-product testing can guarantee the absence of a pathogen present at low levels in a lot.
Sampling plans for microbiological criteria, developed by the International Commission on Microbiological Specifications for Foods (ICMSF), are attributes plans: each sample unit is classified by its result, and the lot is accepted or rejected according to the number of units in each class. In a two-class plan, a limit m separates acceptable from unacceptable units; n sample units are tested, and the lot is accepted if no more than c units exceed m. For pathogens, m is usually absence in a stated quantity, such as 25 grams, and c is 0. In a three-class plan, used for counts, there are two limits: m separates good from marginally acceptable quality, and M separates marginal from unacceptable. The lot is accepted if no unit exceeds M and no more than c units fall between m and M. The plan therefore tolerates some marginal results, reflecting normal variation in good processes, while rejecting any clearly unacceptable unit. The stringency of a plan increases with larger n and smaller c. The ICMSF classified foods into fifteen cases according to the severity of the hazard and whether conditions of handling and use would reduce, not change or increase the hazard, with more stringent plans for more severe hazards and for foods in which the hazard could increase; for example, the most stringent case, for severe hazards in foods for vulnerable populations under conditions that may increase risk, uses n = 60 and c = 0. Sample units should be taken at random from the lot to be representative, and composite samples may be used for presence or absence tests when the method has been validated for them.
A sampling plan is described by its operating characteristic curve, which shows the probability of accepting a lot against the true proportion of defective units in it. For a two-class plan with c = 0, the probability of acceptance is (1 minus p) raised to the power n, where p is the proportion of defective units. With 5 per cent of units defective, a plan with n = 30 accepts the lot about 21 per cent of the time. When c is greater than 0, the binomial distribution is used: with n = 10, c = 1 and 10 per cent defective units, the probability of acceptance is 0.9 to the power 10 plus 10 times 0.1 times 0.9 to the power 9, about 0.74. The number of samples needed to detect a given level of contamination with a stated confidence follows from the same formula: to have a 95 per cent probability of finding at least one positive when 5 per cent of units are contaminated requires n of at least the logarithm of 0.05 divided by the logarithm of 0.95, which is 59 samples; for 1 per cent contamination, about 299 samples are needed. These numbers show why testing cannot assure safety when contamination is rare and uneven: realistic sampling plans detect gross contamination but often miss low-level contamination. Concentrations in foods are usually log-normally distributed between units, and ICMSF's later approaches use this to relate plans to the mean concentration and its standard deviation. Plans are chosen by balancing the consumer's risk of accepting a bad lot against the producer's risk of rejecting a good one, together with the cost of testing.
Testing for every pathogen in every sample is impossible, so laboratories use indicator organisms, whose presence or numbers signal a condition, such as inadequate processing, contamination after processing or poor hygiene, that could also allow pathogens to be present. A distinction is sometimes made between indicators of process quality and index organisms, whose presence suggests the possible presence of a related pathogen. A good indicator is easily and quickly detected, present when the condition it indicates exists, absent or low when it does not, and behaves like the pathogens of concern during processing and storage. Coliforms are Gram-negative, non-spore-forming, facultatively anaerobic rods that ferment lactose with production of acid and gas within 48 hours at 30 to 37 degrees Celsius; the group includes Escherichia, Citrobacter, Enterobacter, Klebsiella and other genera, many of which live in the environment as well as the intestine. In milk and dairy products, which are pasteurised, coliforms are killed by the heat treatment, so their presence in a pasteurised product indicates post-pasteurisation contamination or, less often, inadequate pasteurisation, as described in Microbiology of Raw and Market Milk. They are counted on violet red bile lactose agar, following international standards such as ISO 4832 (incubation at 30 or 37 degrees Celsius depending on the standard applied) or by most probable number methods. Many dairy standards, including the US Pasteurized Milk Ordinance limit of 10 coliforms per millilitre in pasteurised milk, use coliforms.
Enterobacteriaceae are a broader group of Gram-negative, oxidase-negative, facultatively anaerobic rods that ferment glucose; they include the coliforms and also lactose-negative organisms such as Salmonella, Shigella and many strains of Cronobacter and Yersinia. Because they include pathogens that coliform tests would miss, the European Union uses Enterobacteriaceae rather than coliforms for many process hygiene criteria, for example in pasteurised milk, milk powders and infant formula, where their presence in dry processing areas and products indicates conditions under which Salmonella and Cronobacter could occur. They are counted on violet red bile glucose agar following ISO 21528. Escherichia coli is the indicator most specific for faecal contamination, because it lives mainly in the intestines of warm-blooded animals; thermotolerant (faecal) coliform tests at 44 to 44.5 degrees Celsius were used as a proxy, but E. coli is now usually counted directly using its beta-glucuronidase activity on chromogenic media such as tryptone bile glucuronide agar at 44 degrees Celsius (ISO 16649). E. coli is used as a process hygiene criterion for raw milk cheeses and butter made from raw milk. Interpretation requires context: coliforms and Enterobacteriaceae in raw milk are expected and indicate milking hygiene; in pasteurised products they indicate recontamination; in fermented products they decline over time, so a negative result may mask earlier contamination; and E. coli suggests faecal contamination and a raised risk of enteric pathogens. Total viable counts, yeasts and moulds, and aerobic spore counts are other commonly used quality indicators.
The standard plate count, also called the aerobic plate count or total viable count, estimates the number of viable aerobic mesophilic microorganisms. Decimal dilutions of the sample are prepared in a diluent such as peptone saline; measured volumes are mixed with molten plate count agar in Petri dishes (pour plate) or spread on the surface of set agar (spread plate); and plates are incubated, under the international method ISO 4833-1 at 30 degrees Celsius for 72 hours, or under the US standard methods for dairy products at 32 degrees Celsius for 48 hours. Colonies are counted on plates within the countable range, commonly up to about 300 colonies per plate. Under ISO 7218, when two successive dilutions are counted, the count per millilitre or gram is calculated as a weighted mean: N equals the sum of colonies on all counted plates divided by the product of the volume plated, (n1 plus 0.1 times n2) and d, where n1 and n2 are the numbers of plates at the first and second dilutions and d is the first dilution counted. For duplicate plates of 1 millilitre with 232 and 244 colonies at a 1 in 100 dilution and 33 and 28 colonies at 1 in 1,000, N equals 537 divided by (1 times 2.2 times 0.01), about 24,000, reported as 2.4 times 10 to the power 4 per millilitre. Counts are reported to two significant figures, and microbiologists often work with log10 values, here 4.38. Plate counts are affected by the medium, temperature, time and the physiological state of cells, so the method must be stated with the result.
Several other methods are used in dairy laboratories. The most probable number (MPN) method inoculates replicate tubes of liquid medium at three or more successive dilutions and estimates the concentration statistically from the pattern of positive tubes using tables; it is useful for low numbers and for organisms such as coliforms in liquids, though less precise than plate counts. Membrane filtration concentrates organisms from large volumes of liquid, such as water or clear rinse samples, on a filter that is then placed on a medium. Dry rehydratable film methods contain medium and indicators in a film that is inoculated with 1 millilitre of sample; they are convenient and widely used for coliforms, E. coli and total counts after validation against reference methods. Spiral plating deposits a sample in a decreasing amount along a spiral, allowing a wide range of counts on one plate. Flow cytometry, which stains and counts individual cells passing a laser, is used for rapid total bacterial counts in raw milk payment testing, with results converted to colony-forming-unit equivalents, as described in Microbiology of Raw and Market Milk. Impedance and conductance methods measure changes in the electrical properties of a medium as bacteria grow; the detection time is inversely related to the initial count. The quality of all enumeration depends on correctly prepared and tested culture media, as specified in ISO 11133, and on understanding counting uncertainty: colony counts follow approximately a Poisson distribution, so a plate with 100 colonies has a standard deviation of about 10, a relative uncertainty of 10 per cent, before other sources of error are added.
Pathogens are usually present in foods at low numbers, often injured by processing, among a large background of other organisms, so their detection relies on enrichment followed by selective isolation and confirmation. A typical reference method has four stages. Pre-enrichment in a non-selective or mildly selective broth allows injured cells to recover and multiply. Selective enrichment in one or two broths containing agents that inhibit competitors increases the proportion of the target. Selective and differential plating on agars, usually two different ones, allows colonies of the target with characteristic appearances to be recognised. Confirmation of suspect colonies uses biochemical tests, increasingly miniaturised test strips or automated systems, serological tests, and, more and more, identification by matrix-assisted laser desorption ionisation time-of-flight mass spectrometry (MALDI-TOF MS), which identifies bacteria in minutes from their protein profiles. Reference methods take several days: about 3 to 5 days for a negative result and 5 to 7 days for a confirmed positive. Results are usually reported as detected or not detected in a stated quantity, commonly 25 grams, or, for some organisms such as Listeria monocytogenes, coagulase-positive staphylococci and Bacillus cereus, as counts per gram by direct plating. Laboratories must include positive and negative controls and use reference strains, which, together with enrichment cultures, must be handled under the biosafety conditions described in chapter 9.
International standards specify reference methods for the main dairy pathogens. ISO 6579-1 for Salmonella uses pre-enrichment in buffered peptone water at 37 degrees Celsius, selective enrichment in Rappaport-Vassiliadis soya broth at 41.5 degrees Celsius and in Muller-Kauffmann tetrathionate-novobiocin broth at 37 degrees Celsius, and plating on xylose lysine deoxycholate agar and a second selective agar, followed by biochemical and serological confirmation. ISO 11290-1 for Listeria monocytogenes uses primary enrichment in half-strength Fraser broth and secondary enrichment in Fraser broth, with plating on a chromogenic agar on which L. monocytogenes forms blue-green colonies with an opaque halo, and a second agar; ISO 11290-2 covers enumeration. ISO 6888 covers coagulase-positive staphylococci on Baird-Parker agar or a medium with rabbit plasma and fibrinogen, with confirmation by the coagulase test. ISO 7932 covers presumptive Bacillus cereus on mannitol egg yolk polymyxin agar. ISO 22964 covers Cronobacter, ISO 16654 Escherichia coli O157 using immunomagnetic separation after enrichment, and ISO 10272 Campylobacter using microaerobic conditions. Methods for staphylococcal enterotoxins in milk products use immunoassays after extraction and concentration. Laboratories may use alternative methods only if they have been validated against the reference method, under ISO 16140-2, by independent schemes such as AOAC International, NordVal or MicroVal certification, and verified in the laboratory under ISO 16140-3. Whole-genome sequencing is now used to characterise isolates for typing, outbreak investigation and tracing of persistent strains in plants.
Rapid methods give results faster than conventional culture, from minutes to a day or two, and are used for hygiene monitoring, screening of raw materials and products, and positive release. ATP bioluminescence measures adenosine triphosphate, present in all living cells and in food residues, using the firefly enzyme luciferase, which in the presence of ATP, luciferin, oxygen and magnesium produces light; the light is measured in a luminometer and expressed in relative light units (RLU). It is used to verify the cleanliness of surfaces after cleaning: a swab is taken, read within minutes, and compared with pass, caution and fail thresholds set for each site from experience. Because it detects ATP from food residues as well as microorganisms, and cannot identify organisms, it is a measure of overall cleanliness, not a microbial count, and it complements rather than replaces microbiological swabs. Immunoassays detect pathogens or toxins using antibodies. Enzyme-linked immunosorbent assays (ELISA) in sandwich format detect bacterial antigens after enrichment, because they need of the order of a hundred thousand to a million cells per millilitre; lateral flow devices give results within minutes from enriched cultures; and immunomagnetic separation uses antibody-coated magnetic beads to capture target cells from enrichment broths, concentrating them before plating or other detection. Automated immunoassay systems are widely used for Salmonella and Listeria screening. Principles of immunoassay formats are described in Dairy Analysis and Instrumental Methods.
The polymerase chain reaction (PCR) amplifies a specific DNA sequence by repeated cycles of denaturation of double-stranded DNA at about 94 to 95 degrees Celsius, annealing of two primers to the target at a lower temperature, often 50 to 65 degrees Celsius, and extension by a heat-stable DNA polymerase at about 72 degrees Celsius. With perfect efficiency, the amount of target doubles each cycle, so 10 copies become about 10 to the power 10 after 30 cycles. In real-time (quantitative) PCR, fluorescence is measured during each cycle, using dyes that bind double-stranded DNA or sequence-specific probes such as hydrolysis probes, which release a fluorescent signal when cleaved during extension. The cycle threshold (Ct), the cycle at which fluorescence crosses a threshold, is inversely related to the logarithm of the starting quantity: a tenfold difference in target corresponds to about 3.3 cycles at 100 per cent efficiency. Efficiency is calculated from the slope of a standard curve of Ct against log concentration as 10 raised to the power of minus one divided by the slope, minus one; a slope of minus 3.32 corresponds to 100 per cent and a slope of minus 3.6 to about 90 per cent. Real-time PCR after a short enrichment detects Salmonella, Listeria, E. coli O157 and other pathogens within one to two days, and multiplex assays detect several targets at once. A key limitation is that PCR detects DNA from dead as well as live cells; enrichment, which favours live cells, and viability treatments with dyes such as propidium monoazide that block DNA from dead cells reduce false positives. Isothermal methods such as loop-mediated isothermal amplification (LAMP) amplify DNA at a single temperature and suit simpler instruments.
A biosensor is an analytical device that combines a biological recognition element with a physical or chemical transducer that converts the recognition event into a measurable signal. The recognition element, or bioreceptor, may be an antibody, an enzyme, a nucleic acid probe, an aptamer (a short single-stranded nucleic acid selected in the laboratory to bind a target, by a process called SELEX), a bacteriophage or its receptor-binding protein, a receptor protein, or whole cells. Transducers fall into three main groups. Electrochemical transducers measure current (amperometric), potential (potentiometric) or impedance (impedimetric) changes when the target binds or an enzyme reaction occurs; they are cheap and easily miniaturised. Optical transducers measure changes in light, including surface plasmon resonance, which detects the change in refractive index at a metal surface when molecules bind, fluorescence, chemiluminescence and colour changes, for example of gold nanoparticles that change colour when they aggregate. Piezoelectric transducers, such as the quartz crystal microbalance, measure the change in resonant frequency of a crystal when mass binds to its surface. Nanomaterials such as gold nanoparticles, carbon nanotubes and graphene are widely used to increase sensitivity. The aims of biosensor development are speed, sensitivity, specificity, portability, low cost and, ideally, on-line or at-line use in processing plants and at milk collection centres.
Biosensors have been developed for most dairy safety targets. Antibiotic residues in milk are detected by receptor-based and immunological sensors, including surface plasmon resonance systems and strip readers used at milk reception, as described in Dairy Analysis and Instrumental Methods. Aflatoxin M1 sensors based on antibodies or aptamers aim at rapid screening at collection centres. Sensors for pathogens such as Listeria monocytogenes, Salmonella and Escherichia coli O157 use antibodies, aptamers, phages or nucleic acid probes. Somatic cell counting and mastitis detection in milking systems use electrical conductivity and other sensors. Adulterants such as melamine, urea and detergents have been targeted with enzyme, nanoparticle and electrochemical sensors. In practice, relatively few biosensors have moved from research into routine dairy use, and those that have are mainly for residues and somatic cells. The main limitations are the complexity of the milk matrix, in which fat, proteins and calcium interfere with recognition and signal; insufficient sensitivity for pathogens, which are present at very low numbers and so still require enrichment; limited stability and shelf life of biological components; difficulty in distinguishing live from dead cells; and the need for validation against reference methods under standards such as ISO 16140 before results can be used for decisions. A biosensor result used for regulatory or release decisions must be supported by validation data for the specific matrix, by quality control, and by confirmation of positives with reference methods, just as for other rapid methods.
An environmental monitoring programme (EMP) samples the processing environment to verify that hygiene controls are working and to find and eliminate harbourage sites of pathogens before they contaminate products. It is particularly important for organisms that persist in plants, such as Listeria monocytogenes in wet ready-to-eat areas, where products such as soft cheeses, ice cream and butter can be recontaminated after heat treatment, and Salmonella and Cronobacter in dry areas producing milk powder and infant formula. Sampling sites are organised in zones by their proximity to exposed product. Zone 1 comprises food-contact surfaces, such as fillers, slicers, conveyor belts, cheese moulds and utensils. Zone 2 comprises non-food-contact surfaces close to zone 1, such as equipment frames, control panels and the exteriors of machines. Zone 3 comprises more remote surfaces within the processing room, such as floors, walls, drains and overhead structures. Zone 4 comprises areas outside the processing room, such as corridors, locker rooms, warehouses and maintenance areas. Most samples are usually taken from zones 2 and 3, because finding the organism there gives warning before it reaches food-contact surfaces, while zone 1 positives have direct implications for product. Hygienic zoning of the plant itself, separating raw and finished product areas and wet and dry areas with controls on movement of people, materials and air, is described in the subject on cleaning-in-place and hygienic design.
Samples are taken with sterile sponges or swabs moistened in a neutralising broth, which inactivates residual sanitisers so that they do not kill the organisms collected; large areas are sampled with sponges for presence or absence of Listeria or Salmonella, while smaller defined areas, such as 100 square centimetres, are swabbed for counts. Samples are taken during production, often several hours after start-up, when organisms dislodged from harbourage sites are most likely to be found, as well as after cleaning to verify sanitation. Targets are often indicator groups, such as Listeria species rather than only L. monocytogenes, or Enterobacteriaceae in dry areas, because they are easier to detect and signal conditions favourable to the pathogen. Frequencies are set by zone and risk, with more frequent sampling in high-risk areas, and sites are rotated so that the whole environment is covered over time. A positive result triggers a planned response: cleaning and re-sampling of the site and surrounding areas, investigation of the cause, increased sampling (sometimes called a seek-and-destroy approach) to find the harbourage, and, for zone 1 positives, evaluation of affected product, which may need to be held and tested. Results are trended over time to identify recurring sites and seasonal patterns, and isolates are characterised, increasingly by whole-genome sequencing, to tell persistent resident strains from transient introductions. Because an EMP aims to find organisms, a programme that never finds anything may be sampling the wrong places or at the wrong times.
Microbiology laboratories handle organisms that can infect staff and, if they escape, contaminate products and plants, so biosafety is essential. Microorganisms are classified into risk groups according to their ability to cause disease, its severity, its transmissibility and the availability of prevention and treatment. Risk group 1 organisms are unlikely to cause disease, such as starter cultures; risk group 2 organisms can cause disease but are unlikely to spread widely, and effective treatment is available, such as Salmonella, Listeria monocytogenes, Staphylococcus aureus and Bacillus cereus; risk group 3 organisms cause serious disease and may spread, such as Mycobacterium bovis, Brucella and Coxiella burnetii; risk group 4 organisms cause very serious disease with no treatment. Some agents are classified differently in different countries; Shiga toxin-producing Escherichia coli O157, for example, is placed in a higher hazard group in some national schemes because of its low infective dose and severe complications. Laboratories are built and operated at matching biosafety levels: BSL-1 for basic teaching, BSL-2 for most food microbiology including pathogen testing, with restricted access, biological safety cabinets for procedures that generate aerosols and access to an autoclave; BSL-3 for risk group 3 agents, with negative-pressure containment; and BSL-4 for the most dangerous agents. The World Health Organization's Laboratory Biosafety Manual, fourth edition (2020), emphasises a risk-based approach in which controls are chosen from a local risk assessment of the procedures and organisms involved. Brucella is notable for having caused many laboratory-acquired infections.
Good microbiological practice and procedure is the foundation of laboratory safety. It includes no mouth pipetting; no eating, drinking or applying cosmetics in the laboratory; wearing laboratory coats, gloves where needed and eye protection; washing hands after work and before leaving; decontaminating work surfaces with appropriate disinfectants, such as 70 per cent ethanol or hypochlorite; minimising aerosols by careful technique and by using sealed centrifuge buckets and biological safety cabinets; safe handling and disposal of sharps; and decontaminating all cultures and contaminated materials before disposal, usually by autoclaving at 121 degrees Celsius for at least 15 to 20 minutes. Class II biological safety cabinets, the type most used in food microbiology, protect the worker, the sample and the environment by inward airflow and high-efficiency particulate air filtration; Class I cabinets protect the worker and environment but not the sample; Class III cabinets are sealed glove boxes for the most dangerous agents. Laboratories need written procedures for spills and accidents, training and records, and medical surveillance where appropriate. A particular issue for dairies is the location of laboratories inside processing plants. Enrichment cultures and positive control strains of pathogens contain enormous numbers of the very organisms the plant is trying to exclude, so many companies send pathogen testing to external laboratories or confine it to a separated, controlled laboratory with its own entrance, air handling and waste control, away from production. Plant laboratories that test only indicators and use validated rapid methods without pathogen controls reduce this risk.
ISO/IEC 17025, General requirements for the competence of testing and calibration laboratories, most recently revised in 2017, is the international standard against which laboratories are accredited. Its general requirements cover impartiality and confidentiality. Structural requirements define the laboratory's legal identity, management and responsibilities. Resource requirements cover personnel, who must be competent, trained and authorised for the tasks they perform; facilities and environmental conditions; equipment, which must be calibrated, maintained and checked; metrological traceability of measurements to national or international standards; and externally provided products and services, such as media and reference materials. Process requirements cover the review of requests and contracts; the selection, verification and validation of methods; sampling; handling of test items; technical records; evaluation of measurement uncertainty; ensuring the validity of results through internal quality control and participation in proficiency testing or interlaboratory comparisons; reporting of results; complaints; and control of nonconforming work. Management system requirements cover documents, records, risks and opportunities, improvement, corrective action, internal audits and management reviews. Accreditation is granted by national accreditation bodies, such as India's National Accreditation Board for Testing and Calibration Laboratories, which are signatories to the International Laboratory Accreditation Cooperation mutual recognition arrangement, so that accredited results are accepted across borders. Accreditation applies to a defined scope of tests, not to the laboratory as a whole, and many regulators require official control laboratories to be accredited.
Several standards support the application of ISO/IEC 17025 to microbiology. ISO 7218 gives general requirements and guidance for microbiological examinations, including laboratory design, equipment, media preparation, dilutions, colony counting and calculation of results. ISO 11133 specifies the preparation, production, storage and performance testing of culture media, including the use of test strains to confirm productivity and selectivity of each batch. ISO 16140-3 describes how a laboratory verifies that it can perform a validated reference or alternative method correctly in its own hands, for example by testing artificially contaminated samples at low levels. ISO 19036 describes the estimation of measurement uncertainty for quantitative determinations, recognising that microbiological counts have large uncertainties, often expressed as a standard deviation of the log10 count. Internal quality control includes duplicate counts, control charts of results for reference materials, checks of incubator and water bath temperatures, sterility checks of media and diluents, and positive and negative controls for detection methods. Proficiency testing schemes, operated by providers accredited to ISO/IEC 17043, send samples with known organisms and counts; for quantitative tests, performance is usually assessed by z-scores calculated on log10 counts, so a laboratory reporting log 4.30 against an assigned value of log 4.00 with a standard deviation of 0.25 obtains z = 1.2, a satisfactory result, while absolute values of 2 to 3 are questionable and 3 or more unsatisfactory, as explained in Dairy Analysis and Instrumental Methods. For detection tests, performance is judged by correct detection and absence of false results.
A student uses this subject in every dairy microbiology practical: preparing and checking media, making dilutions and plate counts and calculating results by the weighted mean, counting coliforms, Enterobacteriaceae and E. coli, running enrichment-based detection under supervision, using ATP swabs, running a PCR, and working safely at the bench. Learn the components of criteria, how sampling plans work and their probabilities, the reference methods and their principles, and the logic of environmental monitoring, because examiners ask them and every laboratory job depends on them.
Five years into work as a dairy microbiologist or laboratory manager, you set testing plans for raw materials, products and environment; interpret indicator results and decide on corrective action; choose and verify rapid methods; run the environmental monitoring programme and respond to positives; keep media, equipment and methods under quality control; participate in proficiency testing; prepare for accreditation audits; and explain to production managers what test results do and do not show. You also review supplier certificates, investigate customer complaints and train plant staff in aseptic sampling.
A senior quality director with twenty-five years of experience designs the microbiological verification system for a group of plants: criteria and specifications agreed with customers and regulators, sampling plans matched to risk, use of external and internal laboratories, positive-release policies, investment in rapid methods and genomic typing, environmental monitoring standards and laboratory accreditation. Such a person decides how much testing is enough and defends those decisions to auditors and authorities, and manages the response when a pathogen is found in product or environment, including holds, recalls and root-cause investigation.
A researcher with long experience develops and validates new methods, from biosensors and isothermal amplification to culture-independent and genomic approaches; studies the statistics of sampling and measurement uncertainty; designs environmental monitoring strategies using genomic data; and contributes to international standards for methods and criteria. Such a person ensures that new methods are judged by their performance in real dairy matrices, not only in the laboratory. Such researchers also train analysts, review method validation studies for journals and standards bodies, and advise regulators on whether a proposed criterion can be verified with the methods available.
This manual explains principles and gives the main standard methods by number; it does not reproduce full procedures, media formulations or incubation details, which must be taken from the current standards, as they are revised. National criteria and limits are given only as examples. Commercial test kits, instruments and their performance are not described by name. Culture-independent sequencing of microbial communities and predictive microbiology are mentioned only briefly. Biosafety classifications differ between countries; the national list should be consulted. Chemical analysis and its validation are covered in Dairy Analysis and Instrumental Methods, and the organisms themselves in the dairy microbiology subjects.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.