Microbiological testing

 

Medicinal products may only be produced and circulated to be dispensed to the consumer within the scope of the law, if the substances contained within them and their dosage forms comply with the recognised pharmaceutical rules. In all European countries, there are similar regulations. These quality requirements also include requirements for a certain microbiological quality, which differs depending on the place and type of application, but which must be monitored. This means that corresponding testing must be carried out for the microbial count and for germ species defined in more detail in the pharmacopoeia.

The methods required to monitor the specified microbiological quality are also given in the pharmacopoeia: 2.6.12 Microbiological Examination Of Nonsterile Products: Microbial Enumeration Tests and 2.6.13 Microbiological Examination of Nonsterile Products: Tests for Specified Microorganisms. The pharmacopoeia defines that, for starting materials, for which the microbiological requirements are listed in a pharmacopoeia monograph, the total microbial count is to be determined using the pharmacopoeia methods (2.6.12). But the methods from 2.6.12 and 2.6.13 should also be used to monitor the microbiological quality of preparations and starting materials for which no corresponding requirements are given in a pharmacopoeia monograph.

This chapter deals with the practical execution of these investigations (chapter 11 Total microbial count and chapter 12 Specified microorganisms). Above all, the work required in addition to the actual performance of the test should be taken into consideration (see chapter 14 Miscellaneous tests).

The methods for determining the total microbial count and for detecting specified microorganisms are both subject to international harmonisation. The aim of this harmonisation is to use the same methods in the USP, in the Japanese pharmacopoeia and in the Ph. Eur. This will make it much easier to export medicinal products, avoiding multiple tests which might be required for export into the scope of a different pharmacopoeia. The harmonised text was published in Europe in Supplement 5.6 and in the USA in USP Suppl. 2. These supplements entered into force on 1st January 2007 (Europe) or will enter into force on 1st August 2007 (USA). The Japanese pharmacopoeia will publish the harmonised texts in September 2007 in JP 15 Suppl. 2. The texts will enter into force in Japan on 1st October 2007.

The European pharmacopoeia contains two sections each in the corresponding chapters:

A. Methods of the European pharmacopoeia (previous text)
B. Harmonised methods

Both methods should coexist for a few years, in order to facilitate the transition (revision of the pharmacopoeia monographs with microbiological requirements and any method revalidations that become necessary). From around 2010, only the harmonised text will remain valid.

However, the investigator is not left free to choose a method. In all three chapters, the pharmacopoeia uses the same wording to define that the first (previous) method must always be used if the conformance with a pharmacopoeia monograph is to be reviewed, unless the consent of the responsible authorities is obtained to use the new method. The second (harmonised) method should be used as soon as possible in all cases in which there is no reference to monographs, that is, for new substances. In the USA is planned also launch step by step. The planned final implementation of the harmonised chapters is postponed from 1st August 2007 until 1st May, 2009. The current chapter will remain effective until 1st May, 2009 (both texts are included in USP 30). Implementation of the postponed harmonised general chapters prior to the May 2009 date is at the discretion of the user and may be subject to regulatory consideration.

There is no information about the practical steps for implementation in Japan.

As both methods will coexist for a few years, both methods are illustrated below.

11 Total microbial count

Chapter 5 of the European pharmacopoeia (Ph. Eur.) contains information on the microbiological quality of various pharmaceutical preparations (without liability).

There are requirements regarding the microbial count (figure 14.K-1) and regarding the absence of certain microorganisms (figure 14.K-12).

This chapter has been revised in the context of the harmonisation and will now also contain the "old" and "new" requirements until around 2010.

To understand the harmonised requirements, please note one important innovation. While the number of bacteria was previously defined as an average count of the colony forming units on the "bacteria medium", the number of fungi as an average count of the colony forming units on the fungal medium, such as Sabouraud Glucose medium, and the total microbial count as a total of the number of bacteria and fungi, it is now the case that the total aerobic microorganism count (TAMC) is the number of colony forming units on Casein soya bean digest agar including fungi and the total yeasts and moulds count (TYMC) is the number of colony forming units on Sabouraud Dextrose medium incl. bacteria (requirements regarding the microbial count are given in figure 14.K-2 and requirements regarding the absence of certain microorganisms in figure 14.K-13).

Even if this section is only a recommendation (the following text is for information only), the health authorities expect information on the microbiological status of all medicinal products to be available. The same applies in the USA, where the chapter USP <1111> now has the title Microbiological Examination of Non-sterile Products: Acceptance Criteria for Pharmaceutical Preparations and Substances for Pharmaceutical Use and until 2009 the "old" chapter <1111>: Microbiological Attributes of Pharmacopeial Articles with the harmonised requirements.

But it is not only the final products that have to be inspected in terms of their microbiological quality. The raw materials used also have to be dealt with. Indeed the microbiological quality of raw materials that are used in the manufacturing of pharmaceutical preparations have an importance which is not to be underestimated. During manufacturing, it is not always possible to carry out organism-reducing steps to ensure a final product with an acceptable quality. It is self-evident that faultless microbiological quality of the raw material, that is, the lowest possible microbial count and absence of certain microorganisms, is an important prerequisite for a faultless final product.

Although the microbiological quality of raw materials used in the manufacturing of pharmaceutical preparations is very important, we are only given little data on the required quality in the official papers. Respectively the European pharmacopoeia, for example, only contains requirements for about 80 raw materials in the substance monograph (figure 14.K-3). In addition to the microbial count limit, the methods to be used are also specified.

In the context of marketing authorisation, the health authorities expect at least random microbiological testing of the medicinal products, in order to obtain information on their quality. If the requirements for raw materials were determined, the harmonised proposal is very helpful:

• TAMC - 10³ CFU/g or ml
• TYMC - 102 CFU/g or ml (figure 14.K-15)

Raw materials for which there are no requirements in the pharmacopoeia must also be monitored at least through random sampling. In practice, every fifth batch, but at least one per quarter is often used as the rule of thumb (see chapter 13 Testing frequencies).

11.1 Determination of the total microbial count

The following section covers the most important issues of the test. As already mentioned, the pharmacopoeia contains both the previous method (A. Method of the European pharmacopoeia) and the harmonised method (B. Harmonised method). As both methods are to coexist until around 2010, both are also dealt with below.

In both texts, the following methods are specified for counting the total multiplying organisms:

• Membrane filtration
• Plate count methods
• Pour plate method
• Surface-spread method
• Most probable number method (MPN method)

The latter can only be used with restriction due to its low reliability. The precision and accuracy of the MPN method are lower than those of the membrane filtration method or the plate-count method. Particularly unreliable results are obtained by counting moulds. For this reason, the MPN method must only be used to determine the bacterial count if no other method is available. The use of the MPN method must be justified.

Environmental conditions

The microbial count must be tested under conditions that avoid accidental contamination of the test product during the test. Aseptic conditions are recommended.

As a general rule, the tests will be carried out under laminar flow. If you use a laminar flow bench, there are often difficulties with the height required to fill the sample and spray volume into the filter stand with the membrane filter method. You must note the necessary distance between the table and LF filter when purchasing a laminar flow unit. If it is not possible to carry out the work under a laminar flow and the test is carried out on the bench, reinforced monitoring must be carried out (see chapter 14.1 Monitoring of the hygiene status) to demonstrate the low-microbial count conditions.

Sampling plan/sample number

The previous text of the pharmacopoeia describes in detail that product sampling must be carried out in accordance with an established plan, with this sampling plan being dependent on factors such as batch size, the risk of infection associated with an impermissible level of contamination in the product, the product properties and the expected contamination level.

The pharmacopoeia names, for example, the three-class sampling plan (see figure 14.K-4), which is particularly suitable for products with problematic distribution of the microorganisms. This three-class sampling plan has only established itself slightly in the routine, but it is used to test raw materials during repeat testing.

Testing of a combined sample from several containers, that is, around ten tubes or blister strips, is much more common, in line with the usual sampling keys for raw materials and bulk products.

In the harmonised text, this behaviour of the investigator is now accounted for, as it is demanded that a random sample be drawn from the bulk material or from the drug product containers and a combined sample produced in order to obtain the required sample quantity.

The previous text does not give a closer indication of the number of samples. It only speaks of a sufficient number. But the harmonised text is different: Here, ten containers are suggested for liquids or solids in aerosol form. This number can be reduced for small batches for clinical trials. This results in the sample quantities specified in figure 14.K-5.

Amount used for the test

The guide value is specified as 10 g or 10 ml. This quantity is also usually easily available. However, there is always the question of how much has to be tested if the available amount is very low or if the preparation only contains a very small quantity of active substance. There is now information on this matter in the harmonised text (see figure 14.K-6)

.

It is recommended that these specifications be used already in all cases, with the prerequisite that this is possible in terms of the validation.

Sample preparation

Before you can carry out the test, the sample must first be prepared. This sampling is now opened in the harmonised text if the described procedure does not lead to a satisfactory result. Please note that an alternative method should be developed if one of the methods described in the pharmacopoeia cannot be used successfully. This sample preparation differs for each product.

• Water soluble products: The test quantity (figure 14.K-6) must first be dissolved. The European pharmacopoeia specifies sodium chloride peptone buffer pH 7.0 or another suitable liquid. USP prescribes a phosphate buffer pH 7.2. Use of the USP buffer is recommended; this is also easily possible in line with the Ph. Eur. (or another suitable liquid). This makes it easy to obtain a test that complies with both the Ph. Eur. and the USP. Then, a dilution of 1:10 or a dilution corresponding to the validation is created. If the medicinal product has an inherent anti-microbial effect, this anti-microbial effect can be eliminated by the addition of an inactivation agent (see section "Activation/neutralisation of anti-microbial properties of drug products").
• Non-fatty products insoluble in water: According to the previous valid text of the European pharmacopoeia, the corresponding test quantity (see figure 14.K-6) must first be suspended in a sodium chloride peptone buffer pH 7.0 or a different suitable liquid. This is often facilitated by the addition of suitable surface active substances, such as polysorbate 80 (1 g per litre). When selecting surface active substances, it must be ensured that they do not have any inherent anti-microbial effect. For example, it is known that heat sterilised isopropylmyristate contains anti-microbial substances. Isopropylmyristate must therefore only be used if sterile-filtered. If the preparation contains anti-microbial substances, you must also add an inactivation agent (see section Neutralisation/removal of anti-microbial activity on page 14). If necessary, you must correct the pH value of the solution to around pH 7.0-7.2. Further dilutions may also be produced with the same diluent.
• Fatty products: The required test quantity (figure 14.K-6) is first homogenised in a quantity of sterile polysorbate 80 corresponding to at most half the mass. If you wish to use a different, suitable, surface active substance, it must be ensured that it does not have an inherent anti-microbial effect. If necessary, you can heat the same briefly to 40 (45) °C during homogenisation. This temperature must not be exceeded under any circumstances, as there is otherwise a risk that any organisms within the product will be damaged and will be more difficult to detect. The emulsion is carefully blended, maintaining the temperature in a water bath or oven, if necessary. To achieve a 1:10 dilution of the starting product, a sufficient quantity of preheated sodium chloride peptone buffer solution pH 7.0 is added and carefully blended. The higher temperature should be maintained for as short a time as possible (not longer than 30 minutes at the most). If the preparation contains anti-microbial substances, an inactivation agent must be added (see section "Activation/neutralisation of anti-microbial properties of drug products"). If necessary, you must correct the pH value of the solution to around pH 7.0-7.2. If necessary, further dilutions are produced with the same diluent (addition 0.1 % polysorbate 80 or another suitable surface active substance).
• Fluids or solids in aerosol form: The harmonised text also contains information on sample preparation of liquid and solid aerosols. The product is to be sprayed into a membrane filter unit or into a suitable sterile container under aseptic conditions. The total content of the container or a certain number of doses for each tested container should be tested.
• Transdermal patches: The protective film is initially removed from ten transdermal patches using sterile tweezers and the patches placed in sterile glass or plastic trays with the adhesive layer facing upwards. The patches are covered with a suitable sterile porous material (gauze or filter-type grating of woven monofilament polymer) to prevent them from sticking together. The patches are shaken with a suitable inactivator for at least 30 minutes in at least 500 ml sodium chloride peptone buffer pH 7.0.
  The harmonised text's description of sample preparation is very similar. There are only certain changes for the solution or dilution liquids:
• Water soluble products: Sodium chloride peptone buffer pH 7.0, phosphate buffer pH 7.2 (new) or Casein soya bean digest broth (new).
• Non-fatty products insoluble in water: Sodium chloride peptone buffer pH 7.0, Phosphate buffer pH 7.2 (new) or Casein soya bean digest broth (new) - surface active substances, such as polysorbate 80 (1 g/l), can be added to make it easier to suspend substances that are difficult to wet.
• Fatty products: In Isopropylmyristate, dissolve sterilised (new) through filtration or mix with the minimum required quantity of sterile polysorbate or another suitable, sterile, surface active substance. Other dilutions with selected solvents - previously sodium chloride peptone buffer pH 7.0.
• Liquids or solid substances in aerosol form (new): Directly without further dilution in membrane filter unit or a sterile container.
• Transdermal patches: Selected liquid, which contains inactivators such as polysorbate 80 and/or Lecithin - previously sodium chloride peptone buffer solution pH 7.0, which contains a suitable inactivator such as polysorbate 80 and/or Lecithin.

Neutralisation/removal of anti-microbial activity

If the medicinal product has an inherent anti-microbial effect, this can be eliminated by the addition of an inactivation agent. The previous text of the pharmacopoeia specifies the following as typical, liquid neutralising agents:

If this neutralising fluid is not sufficient, you can also try another agent (figure 14.K-7). If necessary, you must correct the pH value of the solution to around pH 7.0-7.2. Further dilutions with the same diluent may also be produced. Although the above-mentioned standard agent is no longer mentioned in the harmonised text and only special substances are now specified (figure 14.K-7), the previous substances can still be used in future after corresponding validation (chapter 11.6 Validation of the method).

11.2 Product testing

The pharmacopoeia contains very detailed method descriptions for

• Membrane filtration
• Plate count methods
• Pour-plate method
• Surface-spread method
• Most-probable-number method

Membrane filtration

In our experience, the membrane filter method should always be used if permitted by the drug product, due to the possibility of also testing larger sample volumes and removing the anti-microbial components of the drug product relative easily. The method description is very similar in the previous text and in the harmonised text.

Membrane filters with a nominal pore width of at most 0.45 mm should be used, whose salt rejection for the bacteria to be detected has been proven. Repeatedly, there are discussions on whether or not the use of membrane filters with a maximum pore width if 0.22 mm is necessary for aqueous preparations. In comparative investigations, we were able to ascertain that 0.45 mm is still sufficient here. When selecting the filter material, it must be ensured that the salt rejection for bacteria is not impaired by the components of the product to be tested. For aqueous and oily solutions and for solutions with a low ethanol content, filters of cellulose nitrate should be used - on the other hand, for solutions with a high ethanol content, filters of cellulose acetate should be used. Generally, white filters are used; an imprinted grating makes it easier to count colonies that have formed, especially with a high bacterial microbial content. If, according to experience, you get very light or clear colonies, such as in water investigations with R2A agar, the use of coloured filters (e.g. black or green) is very helpful. The filtration device must be designed in such a way that the filter can be transferred to the culture medium. Generally, the filtration device can be opened under aseptic conditions (laminar flow) and the membrane removed for transfer to the culture medium. However, there are also devices on the market on which the membrane filter is welded to the device and the culture medium is pressed onto the filter in a culture medium cassette.

As you use two culture media per sample, two membrane filters must also be used (1 filter per medium). For each filter, 10 ml or the quantity dilution corresponding to 1 g of the product, is added and immediately filtered. Each membrane filter is washed at least three times by filtering around 100 ml of a suitable liquid through it, such as a sodium chloride peptone buffer solution pH 7.0 (as per validation in the harmonised text). Surface active substances such as polysorbate 80 and inactivators for anti-microbial substances from the product can be added to this solution. Other rinsing quantities may be necessary depending on the validation of the method (chapter 11.6 Validation of the method). For filtration and rinsing, the filter must never be sucked dry, as this can damage the organisms. One membrane filter is then placed on the bacteria medium and the other on the fungi medium for incubation. Some investigators use just one membrane filter for both media, for cost reasons, and split it using sterile scissors before placing it on the media. This should be avoided due to the risk of contamination. When testing transdermal patches, 50 ml each are filtered through two membrane filters.

Plate-count methods

With the pour-plate method the sample is mixed with fluidised culture medium and incubated after the culture agar has solidified. To this end, 1 ml of the prepared sample is either placed in a pipette or directly in a Petri dish (9 cm diameter) and blended with 15 to 20 ml melted "bacteria medium" or "fungi medium". It must be ensured that the fluidised agar is not warmer than 45 °C, as any microorganisms that are present might be damaged at temperatures higher than this. If the sample is added directly to the petri dish and the culture agar added, the culture medium must be added immediately as the microbial count can be reduced through the adsorption of cells on the petri dish surface. According to studies by Berry and associates (1966), the time should not exceed 10 minutes. This effect does not occur if the sample is blended with the culture agar before it is poured onto the dish. For each medium, at least two petri dishes must be created per dilution. If petri dishes of a different size are used, this must be taken into account in terms of the amount of culture medium, as there is a risk of drying out during the incubation period if too little culture medium is used.

If it is necessary to create dilutions to rule out anti-microbial properties of the sample, this is done at the cost of the limit of detection of the test. It may be possible to compensate for this by increasing the number of plates. For example, if you have to dilute a preparation 1:100 due to anti-microbial properties and place 1 ml in the plate, then you are only testing 0.01 g and the limit of detection would be 100 CFU/g (CFU = colony forming units). However, if you create 10 plates, then you are testing 0.1 g and can therefore detect 10 CFU/g.

The harmonised text does not require any changes in the performance of the test, as it does not contain such detailed specifications, but only requires that the samples be prepared and then that two plates be created for each medium and for each dilution stage.

With the surface-spread method, a defined quantity (at least 0.1 ml) of the prepared sample is smeared onto 15-20 ml solidified "bacteria medium" or "fungi medium" in 9 cm petri dishes. The plates must first be dried under a laminar flow, as there is a high risk of the colonies running on wet plates. For each medium, at least two petri dishes must be created per dilution. As the sample volume to be applied cannot be increased to any desired amount, this procedure is not suitable for products with an inherent anti-microbial effect and for which there is no neutralising agent available.

With this method too, the harmonised text does not require any changes in the performance of the test, as it does not contain such detailed specifications. Again, it is only required that the samples be prepared and then that 2 plates be created for each medium and for each dilution stage

Most-probable-number method (MPN method)

The pharmacopoeia indicates that the precision and accuracy of this method are lower than that of the membrane filter method and counting on agar plates. Particularly unreliable results are obtained when counting moulds. This method must only be used for bacteria if no other counting method is available. As this is the exception, no description of the method is given here. Some information is given in chapter 12.1 Detection of specified microorganisms under Clostridia detection.

11.3 Culture media and culture media checks 

The following have so far been named as culture media:

• Casein soya bean digest agar: This medium is mainly suitable for counting bacteria (bacteria medium).
• Sabouraud Glucose medium with antibiotics: This medium is suitable for growing fungi (Fungi medium). The addition of antibiotics (benzylpenicillin sodium and tetracyclin or chloramphenicol) prevents the growth of bacteria on this fungi medium.

The harmonised text now also refers to:

• Casein soya bean digest agar for determining the total aerobic microbial count (TAMC = Total Aerobic Microbial Count)
• Sabouraud Dextrose agar (without the addition of an antibiotic) for determining the total yeast and mould count (TYMC = Total Combined Yeasts/Moulds Count)

Sabouraud Dextrose medium is initially provided without the addition of an antibiotic and if bacteria grow on this medium they are to be taken into account in the total yeasts and moulds count. The addition of an antibiotic is only possible in exceptions; if there is a risk that the growth of bacteria will cause limits to be exceeded, antibiotics may be added to the Sabouraud Glucose medium.

Previously, the culture media were only named as examples. In the context of harmonisation, they are now stipulated, as they are used as a reference method.

Each batch of the culture media used must comply with the following tests, although this can be carried out before or at the same time as the test on the product to be investigated.

Culture media can either be produced in-house (usually from ready mixed dehydrated culture mixtures, which just have to be dissolved and sterilised) or can be purchased ready-made. Even though purchasing ready-made culture media saves human resource capacity, it nonetheless makes you much less flexible. In the event of problems with the results, it is usually extremely difficult to procure corresponding documents about production.

Sterility

Samples of the culture media and also of the dilution liquid are incubated as per their intended usage, i.e. usually Casein soya bean digest agar at 30-35 °C and Sabouraud Glucose medium with antibiotics at 20-25 °C and dilution fluid at 30-35 °C and 20-25 °C. No data is given on the incubation period. As at least five days are prescribed for the incubation of the test, five days should also be used as the incubation period here. Microbial growth should not be detectable.

Growth-promoting properties (testing the effectiveness of the culture medium)

This important step in microbial testing has previously only be described in general, the detailed procedure of the harmonised text should be followed. Either standardised stable suspensions of the test strains or follow the instructions in the pharmacopoeia should be used. The organisms for inoculation should not be more than five passages away from the original culture. The organisms should be tested individually:

• Staphylococcus aureus, Pseudomonas aeruginosa and Bacillus subtilis on Casein soya bean digest agar and Casein soya bean digest broth; Incubation 18-24 hours at 30-35 °C.
• Candida albicans and Aspergillus niger on Casein soya bean digest agar and Casein soya bean digest broth; Incubation for two to three days at 20-25 °C

The microbial suspensions are produced using phosphate buffer pH 7.0 or pH 7.2; for Aspergillus-niger spores, 0.05 % polysorbate 80 can be added to the buffer. The suspension must be used within 2 hours. If it is stored in the refrigerator at 2-8 °C, it can be used within 24 hours.

For Aspergillus niger and Bacillus subtilis, you can also use a stable spore suspension for inoculation instead of a fresh suspension of vegetable organisms. This can be stored in the refrigerator (2-8 °C) for a validated amount of time. With the organism suspension produced in this way, the growth-promoting properties of the culture media are checked.

To this end, tubes/plates of Casein soya bean digest agar and Sabouraud Dextrose medium are inoculated with a low quantity (not more than 100 CFU) organisms.

Figure 14.K-8 summarises the organisms to be used as per Ph. Eur. and the harmonisation draft from the harmonised text. For each organism, a separate plate/tube must be used. The incubation conditions are listed in figure 14.K-9.

The organisms must show growth within the prescribed time, although the count must not differ by a factor of 5 (by a factor of 2 in the harmonised text) from the calculated usage amount.

Previously, it was left to the investigator's discretion how often the culture media were checked. Usually, at least every batch of the dried culture medium was tested. In the harmonised text there are clear regulations for the culture media tests for microbial count determination.

The test must be carried out

• on each batch of ready-prepared medium
• on each batch of medium prepared from dehydrated medium
• on each batch prepared from the ingredients described

As sterilisation of the medium is a critical step in the context of culture media preparation, it is extremely important that the medium is not sterilised for too long or at too high temperatures. Therefore, only testing the dehydrated medium batch is not sufficient. It must be ensured not only that it has been sterilised for long enough at the right temperature (risk of non-sterility), but also that it has not been sterilised for too long or at too high temperatures.

11.4 Incubation

The plates with the agar are placed at the cover of the plate in the incubator so that any condensed water does not impair the evaluation or cause the colonies to run by dripping on them. The incubators must be qualified. Critical parameters are uniform heat distribution and the temperature.

The incubation conditions prescribed in the European pharmacopoeia (Ph. Eur.) for "bacteria medium" are five days at 30-35 °C and for "fungi medium" five days at 20-25 °C. If a shorter incubation time allows a reliable count, the incubation period can also be reduced. However, it remains unclear when use can be made of this exception to the rule. It is also not included in the harmonised text. Here, an incubation period of 3-5 days at 30-35 °C for Casein soya bean digest agar and 5-7 days at 20-25 °C for Sabouraud Dextrose agar is specified.

11.5 Evaluation

At the end of the incubation time, the colonies that have grown on the plates are counted. Previously, with membrane filtration, only plates with the maximum number of colonies (< 100) were to be evaluated and the CFU/g or ml calculated. This limitation of the colony count now no longer applies. This is regrettable, as mutual influences between the colonies are to be expected if the count is too high. The count is then used to calculate the number of colony forming units per gram or millilitre.

The pharmacopoeia makes the following requirements of the evaluation:

• The bacteria count will be considered to be equal to the average count of colony forming units on the "bacteria medium" (Casein soya bean digest agar).
• The fungal count will be considered to be equal to the average number of colony forming units on the "fungi medium" (Sabouraud Glucose medium with antibiotics).
• The total viable aerobic count in the sum of the bacterial count and the fungal count. If there is evidence that the same type of microorganisms grow on both media this may be corrected.

This somewhat complicated evaluation has been simplified in the harmonised text, as it only speaks of the Total Aerobic Microbial Count (TAMC) - growth on Casein soya bean digest agar and Total Combined Yeasts/moulds Count (TYMC) - growth on Sabouraud dextrose agar. All colonies are counted, even if detectable fungi grow on Casein soya bean digest agar or bacteria on Sabouraud dextrose agar. If there is a risk that including the bacteria on Sabouraud dextrose medium in the count will make the count exceed the established limit, an antibiotic, such as benzylpenicillin-sodium and tetracycline or chloramphenicol can be added to the medium.

Interpretation of the result

The previous text of the pharmacopoeia contains an interpretation note stating that residue limits can be interpreted in a monograph:

• 10² microorganisms - maximum acceptable residue limit 5 x 10²
• 10³ microorganisms - maximum acceptable residue limit 5 x 10³ etc.

This interpretation is not accepted by the FDA; here, such factors are not permissible. It is important to note that this interpretation can only be used if the specification in the submission file for marketing authorisation is 10² or 10³. If the count is 100, 1000 etc., this is interpreted by the regulatory authorities to mean that the interpretation buffer has not been used. In this case, a microbial count of 101 would already be in excess of the limit, if the limit was 100.

In the harmonised text, a factor of 2 is specified. This means:

• 10¹ CFU: maximum acceptable count = 20
• 10² CFU: maximum acceptable count = 200
• 10³ CFU: maximum acceptable count = 2000 etc.

But the reaction of the FDA is still outstanding for this factor.

11.6 Validation of the method

The EU-GMP-Guideline requires that all methods used in the quality control of medicinal products must be validated. This of course not only applies to chemical methods, but also to microbiological methods. The pharmacopoeia describes the validation of the methods. To this end, microbial suspensions of the organisms specified in figure 14.K-8 are produced and, for each organism (about 100 CFU/ml - the FDA expects an inoculation of 10 (100) CFU), the microbial count methods are tested individually in the absence and presence of the product.

The harmonised text deliberately no longer uses the term "validation" and this is replaced by "suitability of the test". The ability of the test to detect microorganisms in the presence of the product to be tested must be shown. This suitability test must be confirmed if a change in the test performance or in the product is introduced, which may affect the result of the test is introduced.

The previous text of the European pharmacopoeia does not contain any information on incubation, although some specifications are made in the harmonised method (figure 14.K-10).

According to the previous text of the European pharmacopoeia, the recovery of organisms must differ by a maximum of a factor of 5 from the saturation quantity, while the harmonised version stipulates a factor of 2.

The question of how often validation should be carried out is also important. The FDA has given the following opinion on the matter (see figure 14.K-11).

Accordingly, the FDA expects three independent validations on three different batches. If pharmacopoeia methods also apply as validated, the microbial count determinations must be validated specific to the substance or product. Reference is also made to this in the official papers (such as CFR). Even if the method is mentioned in a monograph in the Pharmacopoeia, it must be validated in accordance with the requirements of CFR in house. The harmonisation draft therefore no longer speaks of method validation, but of the test for the suitability of the counting method.

Of course, we must also ask when this test is to be carried out. This has been clearly stipulated in the harmonisation:

• for every new product
• after changes in the performance of the test
• after changes in the product that could have an influence on the test result

There are always discussions about whether validations are transferable to other active substance concentrations. The validations must in principle be carried out specific to the substance or product. The only conceivable exception to this is if a drug product with an identical composition has several different dosages of the active substance. In this case, the validation of the highest and lowest concentration is transferable to the dosages in between on the condition that the two validated methods are identical.

12 Specified microorganisms

As already discussed (chapter 11 Total microbial count), chapter 5 of the European pharmacopoeia (Ph. Eur. 5) contains specifications on the microbiological quality of various pharmaceutical preparations and also in about 80 starting material monographs. There are requirements regarding the microbial count and regarding the absence of specified microorganisms (see figure 14.K-12, figure 14.K-13 and figure 14.K-14). If none of the pharmacopoeia monographs contain requirements regarding microbial purity, then you should refer to the general requirements of the harmonised chapter 5.1.4.

As when determining the total microbial count, the health authorities expect at least random microbiological testing of the medicinal products, in order to obtain information about their quality. Raw materials too (including those for which there are no requirements in the pharmacopoeia) must be tested at least through random sampling (see chapter 11 Total microbial count).

12.1 Detection of specified microorganisms

The pharmacopoeia contains the tests required for detecting specified microorganisms.

Both the previous and the harmonised texts contain the methods for

• Bile-tolerant gram negative bacteria (previously enterobacteria and certain other gram negative bacteria),
• Escherichia coli,
• Salmonella spp.,
• Pseudomonas aeruginosa and
• Staphylococcus aureus.

In addition also for Clostridia, although, no specific requirement has been named to date for these organisms.

The harmonised text also contains a detection method for Candida albicans. Here, corresponding requirements are provided for vaginal preparations.

When naming the corresponding chapter (2.6.13) in a pharmacopoeia monograph, there is an obligation to use this method for detecting the specified microorganisms (unless you have validated your own method against the pharmacopoeia method with corresponding effort). Likewise, the use of the method is recommended when testing preparations, as reference is also made to 2.6.13 here. But the methods from 2.6.13 should also be used to monitor the microbiological quality of raw materials for which no corresponding requirements are given in a pharmacopoeia monograph.

The harmonised text now has a general requirement that these tests should always be used with the defined specifications for microbiological quality both for starting materials and for preparations.

The following section covers the most important points of the individual test. In respect of a series of clear changes in the harmonised text, indications should also be given of how you can create the test in such a way that greater changes are avoided upon enforcement of the harmonised version.

Environmental conditions

Although the pharmacopoeia does not give any data on the test conditions in chapter 2.6.13 "Microbiological testing of non-sterile products - detection of specified microorganisms -" either in the previous text or in the harmonised text, it is not irrelevant whether you use secondary contaminations or not in these tests. The same conditions should be used as when determining the total microbial count (see chapter 11.1 Determination of the total microbial count). Corresponding monitoring should also be carried out (chapter 14.1 Monitoring of the hygiene status).

Sampling plan/sample number

The harmonised text makes the requirement that, when using the test to check the specifications, the specified number of samples should be checked, but there are no details on the matter. As, other than for a few exceptions for raw materials sodium starch glycolate (Type A, Type B and Type C ) - only testing for salmonella and escherichia coli, sodium polystyrene sulphonate - only testing for enterobacteria and certain other gram-negative bacteria, tests for the total microbial count and absence of specified microorganisms are required or in a few cases (bentonite, calcium gluconate, iron gluconate, hydroxypropylbetadex, medical carbon, sodium hyaluronate) only the determination of the total microbial count is required, the sampling plan described in chapter 11.1 Determination of the total microbial count and the sample numbers specified therein should generally be used. It is recommended to also use this when only checking for the absence of specified microorganisms.

Test quantity

The test quantity is specified as 10 g or 10 ml. With this quantity, the sample must be prepared accordingly. For requirements based on 1 gram or millilitre, this specification must be taken into account during further processing.

Sample preparation

Before you can carry out the test, you must first prepare the sample, as when determining the total microbial count. This sample preparation is the same as for the determination of the total microbial count, but partly a reactivation medium is used instead of the buffer (see figure 14.K-16).

The preparation for Clostridia detection deviates from this schema and is described in the detailed description of Clostridia detection (see figure 14.K-12). Sample preparation is described in great detail in the harmonised text in chapter 11.1 Determination of the total microbial count.

12.2 Detection method for the specified microorganisms

During the test for specified microorganisms, a general description is given that reactivation of the microorganisms should be carried out first so that sublethal damaged microorganisms can also be detected. This is then usually followed by selective enrichment and then exclusion of organisms by means of selective media. If the results show a suspicion that the specified organisms are present, corresponding confirmation reactions are suggested.

The incubation conditions are specified for the detection of individual organisms. For enrichment, 18-48 hours at the corresponding temperature is usually specified, while 18-72 hours are usually suggested for incubation of the selective media. These relatively wide time spans allow the investigator to avoid weekend work without impairing the result.

Bile-tolerant gram negative bacteria (previously enterobacteria and certain other gram negative bacteria)

Although the method is primarily designed for detecting bacteria from the family Enterobacteriaceae, it can also be used to detect other gram negative microorganisms (especially from the pseudomona group). In the harmonised text, this otherwise little changed section is called Detection of bile-tolerant gram negative bacteria.

Detection (see figure 14.K-17): for preliminary enrichment the sample prepared in lactose monohydrate broth is incubated for 2-5 hours at 35-37 °C (in the harmonised method in Casein soya bean digest broth for 2-5 hours at 20-25 °C). This section is used for the reactivation of any damaged organisms. They should not multiply. (This step is actually barely validatable.)

The container is then shaken and a quantity corresponding to 1 g or 1 ml of the product is incubated in 100 ml Enterobacteria enrichment broth Mossel for 18-48 hours at 35-37 °C (in the harmonised text for 24-48 hours at 30-35 °C) for selective enrichment. If turbidity occurs, it is incubated for 18-24 hours at 35-37 °C in crystal violet, neutral red, bile agar with glucose (selective medium) (in the harmonised edition with violet red bile glucose agar at 30-35 °C). If well developed, usually red or reddish colonies of gram negative bacteria can be observed after incubation, this is a positive result. This suspicion should be confirmed by further investigations, such as identification systems. The product complies with the test if no colonies of gram negative bacteria develop on any of the plates.

Quantitative determination: For preliminary enrichment the sample prepared in lactose monohydrate broth is incubated for 2-5 hours at 35-37 °C (in the harmonised method in Casein soya bean digest broth for 2-5 hours at 20-25 °C). This section is used for the reactivation of any damaged organisms. They should not multiply. (This step is actually barely validatable.) The container is then shaken and a dilution series with 0.1 g, 0.01 g and 0.001 g (or 0.1 ml, 0.01 ml and 0.001 ml) of the product is created with Enterobacteria enrichment broth Mossel. These dilutions are incubated in Enterobacteria enrichment broth Mossel for selective enrichment for 24-48 hours at 35-37 °C (in the harmonised method for 24-48 hours at 30-35 °C). If turbidity occurs (note the result for the evaluation), incubation is carried out on Crystal violet, neutral red, bile agar with glucose (selective medium) for 18-24 hours at 35-37 °C (in the harmonised version with violet red bile glucose agar 30-35 °C). If well developed, usually red or reddish colonies of gram negative bacteria can be observed after incubation, this is a positive result. The calculation of the bacteria count is carried out using a probability table (see figure 14.K-18). The lowest volume of the product that shows a positive result, and the highest volume that shows a negative result are noted and the probable number of bacteria determined from figure 14.K-18.

Escherichia coli

For preliminary enrichment, a corresponding 1 g or 1 ml of prepared product is incubated in 100 ml Casein soya bean digest broth for 18-24 hours at 35-37°C (harmonised text for 18-24 hours at 30-35 °C) .

If turbidity occurs, the container is shaken and 1 ml is incubated in MacConkey broth for selective enrichment for 18-24 hours at 43-45 °C (harmonised method for 24-48 hours at 42-44 °C) .

If turbidity is observed thereafter, it is smeared on MacConkey agar (selective medium) and incubated for 18-72 hours at 35-37 °C (harmonisation for 18-72 hours at 30-35 °C) . If red, non-mucous colonies with gram negative, rod-shaped bacteria occur, this indicates the presence of E. coli. The harmonised version makes this more general, with a growth of colonies already indicating E. coli. This procedure is significantly more realistic and should also be applied when using the previous method. For confirmation, you can carry out suitable biochemical reactions, such as the formation of indole. In addition, you should use identification systems or in general, immediately start a corresponding identification system. The product complies with the test if such colonies are not observed, or if the biochemical reactions for confirmation are negative (see figure 14.K-19).

Salmonella spp.

For preliminary enrichment the prepared sample in Casein soya bean digest broth is incubated for 18-24 hours at 35-37 °C (harmonised text: 30-35 °C). If turbidity occurs, selective enrichment is carried out for 18-24 hours at 41-43°C in tetrathionate bile brilliant green broth. Here, the test has been thoroughly revised by harmonisation.

The selective medium now is Rappaport Vassiliadis Salmonella enrichment broth (incubation: 18-24 hours/30-35 °C) If growth occurs, at least two of the suggested selective media must be incubated when carrying out the previous method

• Deoxycholate citrate agar
• Xylose, lisine, deoxycholate agar
• Brilliant green, phenol red, lactose monohydrate, sucrose agar

for 18-72 hours at 35-37 °C.

The occurrence of colonies with the properties described in figure 14.K-21 indicates the presence of Salmonella spp.

If you test according to the harmonised method, you inoculate on xylose, lisine, deoxycholate agar and incubate for 18-24 hours at 30-35 °C.

For confirmation, suspected colonies according to the previous method are inoculated onto and into triple sugar, iron agar. After incubation, the appearance of a culture with the following properties indicates the presence of Salmonella spp.: Colour changes from red to yellow in depth but not on the surface, general gas development, with or without hydrogen sulphide. The presence of Salmonella spp. should be confirmed using suitable biochemical and serological reactions.

If you test using the harmonised method, the occurrence of well developed, red colonies with or without black centres on xylose, lisine, deoxycholate agar indicates a suspicion of the presence of Salmonella. Confirmation is carried out through organism identification.

Pseudomonas aeruginosa

This method has been largely retained in the harmonised version. There were only a few changes in the incubation conditions. For preliminary enrichment, a corresponding 1g or 1 ml of prepared product is incubated in 100 ml Casein soya bean digest broth for 18-48 hours at 35-37 °C (harmonisation: 18-24 hours/30-35 °C). If turbidity occurs, it is sub-cultivated on Cetrimide agar (selective medium) and incubated for 18-72 hours at 35-37 °C (harmonisation: 18-72 hours/30-35 °C). If no growth is observed, the product has complied with the test. If growth of gram negative rods occurs, Casein soya bean digest broth is inoculated and incubated for 18-24 hours at 41-43 °C by way of confirmation or, according to the harmonised text, organism identification is carried out.

The product complies with the test if no growth is observed on Cetrimide agar or in Casein soya bean digest broth at 41-43 °C or if the presence of Pseudomonas aeruginosa could not be confirmed (figure 14.K-22).

Staphylococcus aureus

In the context of harmonisation, a new medium was introduced for this method (mannitol salt agar instead of Baird-Parker agar) and the incubation conditions were changed. For preliminary enrichment, a corresponding 1 g or 1 ml of prepared product is incubated in 100 ml Casein soya bean digest broth for 18-48 hours at 35-37 °C (harmonisation: 18-24 hours at 30-35 °C). If turbidity occurs, it is sub-cultivated on Baird-Parker agar (selective medium) and incubated for 18-72 hours at 35-37 °C. If you use the harmonised method, mannitol salt agar must be inoculated and incubated for 18-72 hours at 30-35 °C. The growth of black colonies of gram positive coccoids, surrounded by a clear zone on Baird-Parker agar or yellow-white colonies with a yellow zone on mannitol salt agar indicates the presence of St. aureus. To confirm, you can carry out suitable biochemical reactions, such as coagulase and desoxyribonuclease reaction or, according to the harmonised text, carry out organism identification. The product complies with the test if no growth occurs on Baird-Parker agar or if the biochemical reactions for confirmation are negative.

Clostridium spp.

Detection: First, two parts, which each correspond to 1 g or 1 ml of the prepared product, are taken. One part is heated for 10 minutes at 80 °C and quickly cooled to room temperature, while the other part remains untreated. For preliminary enrichment, 10 ml of each homogenised part is placed in two suitable culture containers with 100 ml Reinforced medium for Clostridia and incubated in aerobic or anaerobic conditions for 48 hours at 35-37 °C (harmonisation: 48 hours at 30-35 °C). If there is growth, subcultures are incubated on Columbia agar (with gentamycine) for 48 hours at 35 to 37 °C in aerobic or anaerobic conditions (harmonisation: 30-35 °C). If there is growth, subcultures of all the different colonies are incubated on Columbia agar (without gentamycine) in aerobic and anaerobic conditions (selective medium), according to the previous method. A growth in exclusively anaerobic conditions of gram positive rod-shaped bacteria with or without endospores, with negative catalase reaction indicates the presence of Clostridium spp. This step is omitted in the harmonised text.

Quantitative determination:
Using the prepared product, dilutions of 1:10, 1:100 and 1:1000 are produced with sodium chloride peptone buffer pH 7.2. From each dilution, three aliquot quantities of 1 g or 1 ml are placed in suitable culture containers with a Durham tube for inoculation of three tubes with 9 to 10 ml sulphite medium. For three dilution stages, there are therefore a total of nine tubes. After mixing through minimal shaking, they are incubated for 24-48 hours at 45.5- 46.5 °C.

Tubes indicating a blackening through iron sulphide and significant gas development in the Durham tubes of at least 1/10 of the volume, indicate the presence of Clostridium perfringens. The probable amount of Clostridium perfringens can be determined using figure 14.K-25.

This quantitative determination is omitted from the harmonised text. If such a determination is required, you should use the previous test as a guide, as the expert groups will presumably also refer to this text.

Candida albicans

The harmonised (new) text suggests a method for the detection of Candida albicans. For preliminary enrichment a prepared product corresponding to 1 g or 1 ml is incubated in 100 ml Sabouraud dextrose broth for 5-7 days at 20-25 °C. If turbidity occurs, it is sub-cultivated on Sabouraud dextrose agar (selective medium) and incubated for two days at 20-25°C. If growth occurs, this may indicate Candida albicans. For confirmation, suitable identification processes must be used (see figure 14.K-26).

12.3 Evaluation

The evaluation is described in the different methods.

As in all cases, however, there is only a suspicion of the corresponding organism, the specified confirmation reactions should always be carried out, or, as suggested in the harmonised text, the findings should be checked using identification systems.

If there is room for interpretation in the determination of the total microbial count, there should be none in the detection of specified organisms. The only exception is the detection of enterobacteria and certain other bacteria. Here, if other bacteria are found, it must be assessed and decided upon if the product can be accepted or not, taking into account the risk of infection for the patient and the germ species.

12.4 Culture media and culture media tests

For the detection of specified microorganisms, the pharmacopoeia or harmonisation draft specifies the culture media listed in figure 14.K-27.

Previously, the media in the European pharmacopoeia were only suggestions, and therefore others could also be used. The pharmacopoeia makes explicit reference to this. To this end however, it must be proven that the media have the same culture and selective properties for the germ species to be tested. As the harmonised methods use separation analyses, the media are now prescribed. You are therefore advised not to use alternative media given in the previous test, as it will initially be assumed in the case of an inspection that the pharmacopoeia methods have been used and the effort for validation of another culture medium is very high.

Each batch of the culture media used must comply with the following tests, although this can be carried out before or at the same time as the test on the product to be investigated.

Sterility

Even if this test is not explicitly required in the pharmacopoeia, it should be carried out in the same way as described in chapter 11.1 Determination of the total microbial count. Only with a sterile medium can you obtain reliable results.

Growth-promoting, inhibitory and indicative properties (Test of the nutritional, selective and indicative properties of the culture media)

While the testing of growth-promoting and selective properties was already required previously, a test for the indicative properties is now also required.

For these tests, the reference strains (see figure 14.K-28) are first incubated in Casein soya bean digest broth for 18-24 hours at 30-35 °C (exceptions: Clostridium sporogenes in Reinforced medium for clostridia; incubation for 24-48 hours at 30-35 °C in anaerobic conditions and Candida albicans on Casein soya bean digest agar and Casein soya bean digest broth; incubation for 2-3 days at 20-25 °C), in order to produce a reference suspension with around 1000 multiplying microorganisms per millilitre through dilution with sodium chloride peptone buffer solution pH 7.0. To test the culture media, the same volume parts of each suspension are mixed according to the previous description, and 0.4 ml (corresponding to around 100 microorganisms of each strain) used as an inoculum for detection of Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli and Salmonella spp. Reference strains are also specified for tests as part of Clostridia detection, although these should be used individually. In the harmonised method, mixed cultures are no longer permitted, instead each strain must be tested individually. Only stable organism suspensions may be used, with the passage count from the original ampoule limited to 5.

The previous version of the test requires that these tests be carried out for at least every batch of dried culture medium. As already illustrated in detail in chapter 11.1 Determination of the total microbial count, the sterilisation of the medium is an important step in the context of culture media preparation. Therefore, it is extremely important that sterilisation is not carried out for too long or at too high temperatures. In the author's opinion this means testing of the dehydrated medium batch is not sufficient. It must be ensured not only that it has been sterilised for long enough at the right temperature (risk of non-sterility), but also that it has not been sterilised for too long or at too high temperatures.

In the harmonised text, the test frequency is clearly established:

• every batch of ready-prepared medium
• every batch of medium prepared in-house from ingredients
• every batch of medium prepared in-house from dehydrated medium

The use of a mixture of all reference strains to check the growth-promoting properties and the selectivity again and again leads to major difficulties in practice. Therefore, we should welcome the fact, that according to the harmonisation, only two organisms are generally provided for each medium - one organism for testing the growth capacity and one for testing the selectivity (see figure 14.K-29).

The assessment of the test has become much stricter. While it was previously sufficient to obtain positive growth, it is now necessary to compare the results of comparable previously tested batches.

• For the test for growth promoting properties a clearly visible growth must occur at the shortest incubation time in liquid media (comparable to a previously tested and accepted batch), and with solid media, the number of colonies must be comparable to that of a previously tested and accepted batch.
• For the test for inhibitory properties no growth may occur at least after the maximum incubation time specified in the test.
• For the test for indicative properties the colonies, after incubation in the prescribed time frame, must be comparable in appearance and indication reactions to a previously tested and accepted batch.

  Figure 14.K-29 Growth promoting, inhibitory and indicative properties of media
   

  Medium	Property	Test strains
  Test for bile-tolerant gram negative bacteria
  Enterobacteria enrichment broth-Mossel	Growth promoting	E. coli P. aeruginosa
  	Inhibitory	S. aureus
  Violet red bile glucose agar	Growth promoting + indicative	E. coli P. aeruginosa
  Test for Escherichia coli
  MacConkey broth	Growth promoting	E. coli
  	Inhibitory	S. aureus
  MacConkey agar	Growth promoting + indicative	E. coli
  Test for Salmonella
  Rappaport Vassiliadis Salmonella enrichment broth	Growth promoting	Salmonella enterica ssp. enterica serotype typhimurium or Salmonella enterica ssp. enterica serotype abony
  	Inhibitory	S. aureus
  Xylose, lysine, deoxycholate agar	Growth promoting + indicative	Salmonella enterica ssp. enterica serotype typhimurium or Salmonella enterica ssp. enterica serotype abony
  	Indicative	E. coli
  Test for Pseudomonas aeruginosa
  Cetrimide agar	Growth promoting	P. aeruginosa
  	Inhibitory	E. coli
  Test for Staphylococcus aureus
  Mannitol salt agar	Growth promoting + indicative	S. aureus
  	Inhibitory	E. coli
  Test for Clostridia
  Reinforced medium for clostridia	Growth promoting	Cl. sporogenes
  Columbia agar	Growth promoting	Cl. sporogenes
  Test for Candida albicans
  Sabouraud dextrose broth	Growth promoting	C. albicans
  Sabouraud dextrose agar	Growth promoting + indicative	C. albicans

12.5 Suitability test of the method (validation of the methods)

Of course, the methods for detection of specified microorganisms must also be validated. This is required in section Testing the validity of the test of pharmacopoeia chapter 2.6.13. As described for the culture media tests (see chapter 12.4 Culture media and culture media tests), corresponding organism suspensions are prepared with the reference organisms (figure 14.K-28) and mixed and diluted in accordance with the previous instruction so that inoculation is carried out with around 100 microorganisms per strain. The addition of the control suspension can, for example, be carried out with the last rinsing fluid (for filtration) or after inactivation or dilution. The method used is considered to be valid if it enables the detection of the specified microorganism. Often it is extremely difficult to achieve this detection using organism mixtures. The harmonised text only requires that the organisms be added individually with this test, which represents a simplification compared with the previous instructions. The method is considered to be acceptable if the reference strains exhibit the reactions described for the methods:

In terms of the frequency of the repetitions, as for the validation of the methods for determination of the total microbial count, you should follow the USP/FDA requirements (see chapter 11.6 Validation of the method). The pharmacopoeia does not give any information on the requirement for validation.

Here too, we now find clear specifications in the harmonised text:

• for every new product
• for changes in the performance of the test
• for changes in the product which may affect the outcome of the test is introduced

As regards the transferability of validations to comparable preparations, you should also proceed as with the methods for determining the total microbial count (chapter 11.5 Evaluation).

13 Testing frequencies

Given the multitude of non-sterile dosage forms and, in particular, the high number of raw materials, it is only possible in the rarest of cases to test all batches or receipts of goods microbiologically. A system must be found that guarantees adequate safety even with random testing.

13.1 Preparations

The microbiological quality of non-sterile dosage forms is highly dependent on other parameters, such as preservation of the product, sanitising in the unit etc. The tests for microbial purity (microbial count and detection of specified microorganisms) can therefore, depending on the drug product, also be carried out through random sampling. Explicit reference is made to this in the explanatory notes to part 1, A.4 and B of the Annex to the modified Directive 75/318/EEC in respect of authorisation to place medicinal products on the market. In the ICH Draft Specifications, Test Procedures and Acceptance Criteria for New Drug Substances and New Products there is a decision-making tree for defining the frequency of testing in the routine (batch by batch or random sampling) (figure 14.K-31).

It is not enough simply to specify whether you want to carry out random sampling or batch by batch testing. If random sampling is possible in accordance with figure 14.K-31, the frequency must also be established. This information too must be specified in the marketing authorisation. In practice, the drug products are first grouped together according to the contamination risk.

We have established three assignment groups, for example:

A: Insufficiently tested drug products: This group initially contains all new introductions from in-house production and external production as well as drug products with significant changes in the formulation and manufacturing procedure. After five batches, assignment to critical (B) or non-critical (C) is carried out.

B: Critical drug products: This group contains all drug products with the microbial count findings exceeded limits.

C: Non-critical drug products: This group contains all drug products for which at least the last five batches did not exceed any limits in the microbial count findings.

Figure 14.K-30 shows the associated testing frequencies and the release relevance of the result.

Of course, such an assignment cannot be rigid, but must be handled flexibly. If microbial counts that exceed limits occur for non-critical drug products, the environment of the batch is tested (usually the five previous and the five subsequent batches). During this time, the microbial count result should be available for batch release. Then the test is carried out again as described under assignment A (figure 14.K-31).

13.2 Raw materials

With the majority of raw materials which are usually used in pharmaceutical companies, we must ask if every raw material and thus every batch must be tested. The pharmacopoeia contains, as shown, corresponding requirements for just a small number of starting materials. This indicates that corresponding tests are not required for all starting materials.

In the above-mentioned IHC draft Specifications, Test Procedures and Acceptance Criteria for New Drug Substances and New Drug Products, we also find a decision-making aid for establishing the test frequency of microbiological tests on raw materials (figure 14.K-32). We can see that the testing and also the test frequency depend on the contamination risk of the starting material.

As with non-sterile dosage forms, we have split the raw materials into different classes depending on the contamination risk: Class 1 contains all materials for which there are microbiological requirements in the pharmacopoeia (obligatory testing) and starting materials from a plant-based or animal origin. With such materials, impurities must always be reckoned on. Synthetic substances are different. These also include natural materials that are reliably disinfected through extraction, etc. during their manufacturing. They are assigned to class 2 (random sampling).

If an impurity can be ruled out, as with nitric acid for example, the material is assigned to class 3. The associated testing frequency is shown in figure 14.K-33.

14 Miscellaneous tests

In addition to the actual tests on the test sample, further tests also have to be carried out to guarantee the results.

14.1 Monitoring of the hygiene status

The work conditions specified for carrying out the tests must be regularly monitored with suitable environmental checks in the working area and using checks of the aseptic work methods.

As already illustrated, the microbial count must be tested in conditions that avoid accidental contamination of the test product during the test. This ultimately means working under aseptic conditions. Similar conditions apply for the detection of specified microorganisms. The same conditions should be used as when determining the total microbial count. Corresponding monitoring should also be carried out. It is, of course, not sufficient to assume the corresponding suitable environmental conditions, they must instead be monitored through corresponding tests. Figure 14.K-34 shows suggested methods and frequencies for monitoring in production rooms.

Figure 14.K-35 shows examples for levels in the monitoring of production rooms. They are based on those for production.

It is also not sufficient to carry out corresponding monitoring and simply collect findings. If levels are exceeded, suitable measures must be taken (figure 14.K-36). If the action level is exceeded, a deviation record must also be compiled in which the influence on the result is considered.

You should use the same monitoring methods as are used to monitor the production areas (see chapter 12.G Microbiological monitoring). This will save additional expense for the maintenance of standard operating procedures, calibration of devices etc. In addition, the results are comparable. As disinfection is often carried out in a microbiological laboratory, it is of particular importance, especially for surface testing, to add inactivating substances to the culture medium. You must always expect to find disinfectant residues on surfaces. If you purchase ready-made contact plates, these usually contain polysorbate and lecithin as neutralising substances. If you are working with swabs, you should be aware that disinfectant residue can be picked up during the swabbing procedure, which can make it necessary to add suitable substances to the medium.

Figure 14.K-37 shows a selection of neutralising substance combinations.

Particular importance is attributed to the detection of the growth-potential using control organisms during environmental monitoring in the laboratory. At the end of the incubation time, the growth potential of the plate, and thus the effectiveness of the inactivation agent, is proven by dropping about 100 Staphylococcus aureus ATCC 6538 on a plate without growth. If this test is negative, the measurement is invalid.

14.2 Aseptic working conditions

For microbial count determinations too the pharmacopoeia requires that the aseptic working condition for sample preparation and, for example, for transfer of the membrane filter to the culture medium be tested. This test should show that aseptic working methods prevail and that no secondary contamination occurs during the test manipulations. The pharmacopoeia prescribes that, to test the efficacy of the aseptic measure used, the method should be carried out as a test preparation using a sterile sodium chloride peptone buffer solution pH 7.0 (previous test) or the selected dilution solution (harmonised test). No growth of microorganisms should be ascertained.

These tests are, of course, to be carried out on people and methods. That is, for example, every person who carries out microbial count determinations on raw materials or preparations must pass this test successfully once a month (FDA investigators sometimes expect controls on test days). If the requirements are not fulfilled, the test must be repeated. If contamination occurs again, personnel training (theory and practice) must be carried out.

Such controls are not considered necessary for tests when detecting specified organisms.