The removal of micro-organisms from fluids by passage through filters is a complex process, and careful attention must be paid to the selection and validation of the filters, according to Dr Tim Sandle.
Filtration is a means of sterilising fluids (liquids or gases) through the removal, rather than the destruction or inactivation, of micro-organisms. Membrane filtration is a technology commonly used in pharmaceutical cleanrooms, particularly for liquids that would be damaged by heat irradiation or chemical sterilisation. The sterilisation of liquids by filtration is a key step for aseptic manufacturing prior to the dispensing of the product.
The characteristics of a sterilising grade filter are that it must be compatible with the process; be non-toxic; pass an integrity test; and be sterilisable (or provided sterilised). Furthermore, filters should not adsorb formula components or add extractables to the process and, most importantly, they must remove the bioburden associated with the product.
The removal of micro-organisms from fluids by passage through filters is a very complex process and is dependent on inter-actions relating to the chemistry and surface characteristics of the membrane, the micro-organisms, and the suspending fluid. Careful attention must therefore be paid to the selection and validation of filters.
Membrane filters
The sterile filtration of liquids, and gases, in pharmaceutical cleanroom manufacture is almost always performed using membrane filters. These are thin, uniform, porous sheets that act as sieves to trap particles, like bacteria, larger in size than the pores in the membranes.
Sieving is not the only means by which filters retain particles. Other factors that prevent the passage of particles include inertial impaction to the walls or surfaces of the pores and lodgement in crevices within the depth of the membrane.
There are two types of membrane: hydrophobic, more often used with gas filtration; and hydrophilic, for use with liquid filtration. This article focuses on liquid (hydrophilic) filters.
Liquid filters generally come in two forms: the less widely used disc (sheet or plate) filters and the more commonly used cartridge filters. Cartridge filters are located within cylindrical stainless steel housings or in disposable plastic housings. Since membranes are fragile and do not fold easily without tearing, the membrane is sandwiched between two support layers of a non-woven fabric. All bacteria retentive filters are required to have a maximum porosity of 0.22μm.
Selection of filter
The selection of a membrane filter for a particular product or process is an important choice and one that requires an assessment of the filter, the chemical nature of the product and the physical demands that will be placed on the filter.
The six key considerations are:1
a) Flow rates: the filter must be tested to verify that it can filter the product for the required volume and at the required flow rate (the time taken to filter). The flow rate is important since many products have a maximum time within which the filtration must be completed.
b) Pressure and temperature resistance: the membrane support layers and the housing of the filter must be sufficiently rugged to withstand the pressures and temperatures associated with the process.
c) Hydrophilic or hydrophobic properties: as noted above, sterilising-grade filters for aqueous solutions are normally hydrophilic. However, with certain solvent or chemical liquids, hydrophobic filters are sometimes used.
d) Membrane composition: the filter system must be assessed to ensure that all product-contact surfaces of the filter can withstand the hydraulic, thermal and chemical challenges of the sterilisation and production processes.
e) Compatibility: the composition of the membrane must be compatible with the chemicals intended to be passed through the filter.
f) Sterilisation of the membrane filter: the filter must be able to withstand the sterilisation process without the process damaging the filter or leading to the shedding of fibres or the release of toxic substances. Sterilisation methods include steam sterilisation and gamma radiation.
The purpose of sterile filtration validation is to prove that a particular filtration process generates a sterile filtrate.2 Filters must be qualified by the user to demonstrate that the performance in processing will meet the process requirements.
These tests are:
a) Physical and chemical compatibility: these tests have been discussed in the section on filter selection above; nonetheless they require demonstrating through validation.
b) Binding and adsorption filter characteristics: it is important to show that the filter does not remove active ingredients or formulation components; otherwise the properties of the product will be affected.3
c) Bacterial retentive efficiency: this is arguably the most important test as the objective of filtration is to produce a sterile filtrate. The validation of bacterial retention requires the complete removal of a minimum challenge level of 107 colony-forming units (cfu) of the diminutive bacterium Brevundimonas diminuta (ATCC 19146) per square centimetre of membrane surface area.4 The test must provide a sufficient challenge to the membrane so that every pore is challenged and given the opportunity to allow passage of the test micro-organism.5
The reason for selecting Brevundimonas diminuta is because the micro-organism produces very small cells with a narrow size distribution and it represents the ‘worst-case’ challenge.6
For the bacterial test, the test must simulate the actual pharmaceutical manufacturing process for a particular drug product, including the physical conditions of the process and the physico-chemical characteristics of the pharmaceutical product solution. This requires the micro-organism to be challenged into the product (provided that the product is not bactericidal, in which case a surrogate material should be used).7
The test method is based on the principles described in PDA Technical Report No. 26.1 For this it is necessary to prove the viability of the Brevundimonas diminuta bacterial challenge, by direct inoculation into the product. When carrying out the challenge, the test exposure time should equal or exceed the actual process filtration time. This means, if the batch requires eight hours to filter, the challenge must be run for at least eight hours.
Further factors to take into account are the maximum process differential pressure and flow rates.8 If the product is normally filtered at a high temperature and where this temperature may kill the challenge micro-organism, then the temperature in the validation should be adjusted downwards so that the test micro-organisms survive the process challenge in sufficient numbers.9
Should the filter fail to retain micro-organisms, an investigation is required. The retention of micro-organisms by the filter is a combination of different factors. These include the filter polymer; the filter structure; the properties of the aqueous product, including pH, viscosity, osmolarity and ionic strength; and the process conditions, including temperature, differential pressure and flow rate. The investigation may lead to process modifications or the selection of an alternative filter.
d) Integrity testing: the porosity for filters requires confirmatory testing, based on the pressure required to displace liquid from the pores (a bubble point value) to show that the filter remains intact.10
e) Leachables and extractables: the possibility of products extruding harmful “extractables” out of the plastic membrane must be addressed.11 In addition to the potential adverse effect of extractables on the filtered product, the presence of extractables may relate to the degradation of the filter, ultimately affecting its ability to perform as intended. Extractables are chemical entities, both organic and inorganic, that will extract from the filter into the product under controlled conditions. Consideration must also be given to leachables – chemical entities, both organic and inorganic – that could migrate into a drug product when it comes into contact with the filter.
Summary
This article has presented an overview of the important points to consider for the selection and validation of membrane filters used for the sterilisation of liquids. It has emphasised the importance of assessing the chemical and physical features of filters and the liquids passed through them, together with the most important aspect: the bacterial challenge study, which is used to show whether a sterile filtrate can be produced.
References
1. PDA Technical Report #26, “Sterilizing Filtration of Liquids,” J.Pharm. Sci. and Technol. 52 (1 Supp) (1998).
2. Aranha H, Meeker J. (1995). Microbial retention characteristics of 0.2-microns-rated nylon membrane filters during filtration of high viscosity fluids at high differential pressure and varied temperatures, PDA J Pharm Sci Technol. 49(2):67–70
3. McBurnie, L. and Bardo, B. (2004). Validation of Sterile Filtration, Pharmaceutical Technology, Filtration Supplement, 10 (8): S13–S22
4. Carter J. (1996). Evaluation of recovery filters for use in bacterial retention testing of sterilizing-grade filters, PDA J Pharm Sci Technol. 50(3):147–53
5. Lee SH, Lee SS, Kim CW. (2002). Changes in the cell size of Brevundimonas diminuta using different growth agitation rates, PDA J Pharm Sci Technol. 56(2):99–108
6. Hunter, D. (2006): Bacterial Challenge & Correlation to Integrity Test Data For Sterilising Grade Pharmaceutical Air & Liquid Filters, Dominic Hunter Group
7. American Society for Testing and Materials (ASTM), Standard Test Method for Determining Bacterial Retention of Membrane Filters Utilized for Liquid Filtration, ASTM Standard F838–05, (ASTM, Philadelphia, PA, 1993).
8. Schroeder, H.G. & DeLuca, P.P. 1980. Theoretical aspects of filtration and integrity testing. Pharmaceutical Technology 4 80–85
9. Lee SH, Kim CW. (2002). Microbial retention characteristics of sterilizing-grade membrane filters with alginate substituted for oil-based products, PDA J Pharm Sci Technol. 56(5):248–54.
10. Jornitz MW and Meltzer TH. Filtration handbook – Integrity testing. Parenteral Drug Association. Bethesda Md. 2003 January
11. Kao YH, Bender J, Hagewiesche A, Wong P, Huang Y, Vanderlaan M. (2001). Characterization of filter extractables by proton NMR spectroscopy: studies on intact filters with process buffers, PDA J Pharm Sci Technol. 55(5):268–77
