Sterility Testing

STERILITY TESTING    

A sterility test is essentially a test which assesses whether a sterilized pharmaceutical or medical product is free from contaminating microorganisms by incubation of either the whole or a part of that product with a nutrient medium. It thus becomes a destructive test and is of questionable suitability for testing large, expensive or delicate products or equipment. Furthermore, by its very nature such a test is a statistical process in which part of a batch is sampled and the chance of the batch being passed for use then depends on the sample passing the sterility test. Random sampling should be applied to products that have been processed and filled aseptically. With products sterilized in their final containers, samples should be taken from the potentially coolest or least sterilant-accessible part of the load.

A further limitation is that which is inherent in a procedure intended to demonstrate a negative. A sterility test is intended to demonstrate that no viable organisms are present, but failure to detect them could simply be a consequence of the use of unsuitable media or inappropriate cultural conditions. To be certain that no organisms are present it would be necessary to use a universal culture medium suitable for the growth of any possible contaminant and to incubate the sample under an infinite variety of conditions. Clearly, no such medium or combination of media are available and, in practice, only media capable of supporting non-fastidious bacteria, yeasts and moulds are employed. Furthermore, in pharmacopoeial tests, no attempt is made to detect viruses, which on a size basis, are the organisms most likely to pass through a sterilizing filter. Nevertheless, the sterility test does have an important application in monitoring the microbiological quality of filter-sterilized, aseptically filled products and does offer a final check on terminally sterilized articles. In the UK, test procedures laid down by the European Pharmacopoeia must be followed; this provides details of the sample sizes to be adopted in particular cases. The principles of these tests are discussed below.

a)  Methods

Three alternative methods are available when conducting sterility tests:

•                              The direct inoculation method involves introducing test samples directly into nutrient media. The European Pharmacopoeia recommends two media: (1) fluid mercaptoacetate medium (also known as fluid thioglycollate medium), which contains glucose and sodium mercaptoacetate (sodium thioglycollate) and is particularly suitable for the cultivation of anaerobic organisms (incubation temperature 30–35 °C); and (2) soyabean casein digest medium (also known as tryptone soya broth), which will support the growth of both aerobic bacteria (incubation temperature 30–35 °C) and fungi (incubation temperature 20–25 °C). Other media may be used provided that they can be shown to be suitable alternatives. Limits are placed upon the ratio of the weight or volume of added sample relative to the volume of culture medium so as to avoid reducing the nutrient properties of the medium or creating unfavourably high osmotic pressures within it.

•                              Membrane filtration is the technique recommended by most pharmacopoeias and, consequently, the method by which the great majority of products are examined. It involves filtration of fluids through a sterile membrane filter (pore size ≤0.45 μm), any microorganism present being retained on the surface of the filter. After washing in situ, the filter is divided aseptically and portions are transferred to suitable culture media which are then incubated at the appropriate temperature for the required period of time. Water-soluble solids can be dissolved in a suitable diluent and processed in this way and oil-soluble products may be dissolved in a suitable solvent, e.g. isopropyl myristate.

•              A sensitive method for detecting low levels of contamination in intravenous infusion fluids involves the addition of a concentrated culture medium to the fluid in its original container, such that the resultant mixture is equivalent to single strength culture medium. In this way, sampling of the entire volume is achieved.

With the techniques discussed above, the media employed should previously have been assessed for nutritive (growth-supporting) properties and a lack of toxicity using specified organisms. It must be remembered that any survivors of a sterilization process may be damaged and thus must be given the best possible conditions for growth.

As a precaution against accidental contamination, product testing must be carried out under conditions of strict asepsis using, for example, a laminar airflow cabinet to provide a suitable environment.

The European Pharmacopoeia indicates that it is necessary to conduct control tests that confirm the adequacy of the facilities by sampling of air and surfaces and carrying out tests using samples ‘known’ to be sterile (negative controls). In reality, this means samples that have been subjected to a very reliable sterilization process, e.g. radiation, or samples that have been subjected to a sterilization procedure more than once. In order to minimize the risk of introducing contaminants from the surroundings or from the operator during the test itself, isolators are often employed which physically separate the operator from the materials under test.

b) Antimicrobial Agents

Where an antimicrobial agent comprises the product or forms part of the product, for example as a preservative, its activity must be nullified in some way during sterility testing so that an inhibitory action in preventing the growth of any contaminating microorganisms is overcome. This is achieved by the following methods.

i)    Specific inactivation

An appropriate inactivating (neutralizing) agent (Table 21.8) is incorporated into the culture media. The inactivating agent must be non-toxic to microorganisms, as must any product resulting from an interaction of the inactivator and the antimicrobial agent.

Although Table 21.8 lists only benzylpenicillin and ampicillin as being inactivated by β-lactamase (from B. cereus), other β-lactams may also be hydrolysed by βlactamases. Other antibiotic-inactivating enzymes are also known  and have been considered as possible inactivating agents, e.g. chloramphenicol acetyltransferase (inactivates chloramphenicol) and enzymes that modify aminoglycoside antibiotics.

ii) Dilution

The antimicrobial agent is diluted in the culture medium to a level at which it ceases to have any activity, for example phenols, cresols and alcohols. This method applies to substances with a high dilution coefficient, η.

iii) Membrane filtration

This method has traditionally been used to overcome the activity of antibiotics for which there are no inactivating agents, although it could be extended to cover other products if necessary, e.g. those containing preservatives for which no specific or effective inactivators are available. Basically, a solution of the product is filtered through a hydrophobic-edged membrane filter that will retain any contaminating microorganisms. The membrane is washed in situ to remove any traces of antibiotic adhering to the membrane and is then transferred to appropriate culture media.

c)   Positive Controls

It is essential to show that microorganisms will actually grow under the conditions of the test. For this reason positive controls have to be carried out; in these, the ability of small numbers of suitable microorganisms to grow in media in the presence of the sample is assessed. The microorganism used for positive control tests with a product containing or comprising an antimicrobial agent must, if at all possible, be sensitive to that agent, so that growth of the organism indicates a satisfactory inactivation, dilution or removal of the agent. The European Pharmacopoeia suggests the use of designated strains of Staphylococcus aureus, Bacillus subtilis and Pseudomonas aeruginosa as appropriate aerobic organisms, Clostridium sporogenes as an anaerobe and Candida albicans or Aspergillus niger as fungi.

In practice, a positive control (medium with added test sample) and a negative control (medium without it) are inoculated simultaneously, and the rate and extent of growth arising in each should be similar. However, the negative control without the test sample, is, in effect, exactly the same as the growth promotion control that is also described in the test procedure, so, for the organisms concerned, it is not necessary to do both.

All the controls may be conducted either before, or in parallel with, the test itself, providing that the same batches of media are used for both. If the controls are carried out in parallel with the tests and one of the controls gives an unexpected result, the test for sterility may be declared invalid, and, when the problem is resolved, the test may be repeated.

d) Specific Cases

Specific details of the sterility testing of parenteral products, ophthalmic and other non-injectable preparations, and surgical sutures will be found in the European Pharmacopoeia. These procedures cannot conveniently be applied to items like surgical dressings and medical devices because they are too big. In such cases the most convenient approach is to immerse the whole object in culture medium in a sterile flexible bag, but care must be taken to ensure that the liquid penetrates to all parts and surfaces of the material.

e)  Sampling

A sterility test attempts to infer the state (sterile or nonsterile) of a batch from the results of an examination of part of a batch, and is thus a statistical operation. Suppose that p represents the proportion of infected containers in a batch and q the proportion of non-infected containers,

then, p + q = 1 or q = 1 − p.

Suppose also that a sample of two items is taken from a large batch containing 10% contaminated containers. The probability of a single item taken at random being contaminated is p = 0.1 (10% = 0.1), whereas the probability of such an item being non-contaminated is given by q = 1 − p = 0.9. The probability of both items being contaminated is p² = 0.01, and of both items being non-contaminated, q² = (1 − p)² = 0.81. The probability of obtaining one contaminated item and one non contaminated item is 1 − (0.01 + 0.81) = 0.18 = 2pq.

In a sterility test involving a sample size of n containers, the probability p of obtaining n consecutive ‘steriles’ is given by qⁿ = (1 − p)ⁿ. Values for various levels of p (i.e. proportion of infected containers in a batch) with a constant sample size are given in Table 21.9, which shows that the test cannot detect low levels of contamination. Similarly, if different sample sizes are employed (also based on (1 − p)ⁿ) it can be shown that as the sample size increases, the probability of the batch being passed as sterile decreases.

It can be seen from the above that a sterility test can only show that a proportion of the products in a batch is sterile. Thus, the correct conclusion to be drawn from a satisfactory test result is that the batch has passed the sterility test not that the batch is sterile.

f)    Retests

Under certain circumstances a sterility test may be repeated, but the only justification for repeating the test is unequivocal evidence that the first test was invalid; a retest cannot be viewed as a second opportunity for the batch to pass when it has failed the first time. Circumstances that may justify a retest would include, for example, failure of the air filtration system in the testing facility which might have permitted airborne contaminants to enter the product or media during testing, non-sterility of the media used for testing, or evidence that contamination arose during testing from the operating personnel or a source other than the sample under test.