MA CEng MIMechE MSEE
High Meadow Innings Gate
Henley on Thames
The well-attended Eighth Pharmaceutical Isolator Conference, held at Warwick University (UK) in December 2004, signposted rapid gassing as one of the directions in which isolator systems are likely to develop over the next few years. The preoccupation with the subject of biological indicators (BIs) during the conference was an indication of the importance of gassing cycle development and validation. Liquid sanitisation with alcohol will still have its place in smaller hospital pharmacies.
Isolators with rapid gassing chambers
Rapid gassing chambers can achieve cycle times of as low as 15–20 minutes. A typical installation might comprise one rapid gassing chamber directly connected to and serving two process isolators. These are more likely to be four-glove isolators than half-suit isolators, which are considered less convenient and less easy to validate. Product handling is optimised by the use of special mobile racks. These can be loaded with product in the preparation area, transferred into the isolator cleanroom, placed into the rapid gassing chamber and then, when the gassing cycle is complete, transferred through into either of the two process isolators. The two glove ports nearest to the rapid gassing chamber are used for unloading the product ready for processing by a technician, who uses the two furthest ports. Completed bags, bottles or syringes as well as waste are then removed through a conventional air-purged transfer chamber or into a sealable continuous polythene liner at the far end of the process isolator. With this arrangement process flows are simpler and freer of risk than in traditional systems, which comprise sterilising isolators, bank or storage isolators and process isolators, with product transfers between these by means of double-door docking containers.
Programmable logic controller (PLC) technology has enabled complete integration of the control systems of isolators and gas generators so that gassing cycles can be carried out automatically from start to finish. Other procedures, such as leak testing, can also be automated, thus removing the possibility of operator error and bringing great simplification of standard operating procedures (SOPs). Even the simplest isolators can benefit from PLC controls, and they are essential in more complex isolator systems, such as those utilising rapid gassing chambers, for effective integration.
Standards and guidelines
The last two or three years have seen a spate of relevant publications, including ISO 14644-7 (the generic isolator standard),(1) European Community good manufacturing practice (EC GMP) Annex 1 2003 revision,(2) PIC/S recommendation Isolators used for aseptic processing and sterility testing(3) and the new Yellow Book Pharmaceutical Isolators.(4) It is unfortunate that the EC GMP is not harmonised with the international classification of air cleanliness set out in ISO 14644-1.(5) This leads to problems in measuring air cleanliness at the 5m-particle size in Grade A zones, the “local zones for high-risk operations”. A typical particle counter, sampling air at the rate of 28.4l/min (1 cubic foot/min), will require nearly 12 hours to draw a statistically meaningful sample! The EC GMP also specifies continuous particle monitoring in a Grade A zone. This raises questions about the most appropriate position for the probe. Should it be placed where it can pick up process-generated contamination, or should it be placed where the air is supposed to be uncontaminated? In the latter case, is there any particular position that is truly representative? These questions need to be resolved at the design stage, so that the continuous monitoring is meaningful in relation to product quality as well as meeting regulatory requirements.
There remains much confusion and controversy over the relative merits of “laminar” and “turbulent” airflow in isolators. In fact, these terms are used incorrectly and have an entirely different meaning in other branches of engineering and science, which is why they are not used in current clean air standards. The correct terms, “unidirectional airflow” and “non-unidirectional airflow”, are defined in ISO 14644-4 as follows:(6)
- Unidirectional airflow: controlled airflow through the entire cross-section of a clean zone with a steady velocity and approximately parallel streamlines. Note: this type of airflow results in a directed transport of particles from the clean zone.
- Non-unidirectional airflow: air distribution where the supply air entering the clean zone mixes with the internal air by means of induction.
These definitions are self-explanatory apart from the misuse of the term “induction”. The definition for non-unidirectional airflow would be adequate without the last four words. The relative merits of the two airflow systems are well rehearsed. The important thing for designers is to demonstrate that Grade A air is achieved in whichever airflow system is chosen and that there are no “dead” zones.
High-efficiency particulate air (HEPA) filters are essential components in isolators. Care must be taken in specifying HEPA filters because, on its own, the filter manufacturer’s standard EN 1822 does not provide enough information.(7) Specifiers must also include the pass/fail criteria for the in-situ installed filter test method to be used and, in the case of unidirectional airflow isolators, the airflow uniformity requirement. EN 1822 classifies filters according to the penetration of the most penetrating particle size (MPPS). Due to the way HEPA filters are made, particles both larger and smaller than the MPPS penetrate less easily and, therefore, the filtration efficiency at MPPS is the very worst case. For a typical HEPA filter, the MPPS is approximately 0.12micron. All of this is comprehensively explained in Pharmaceutical Isolators and in a Technical Monograph from the Parenteral Society.(8)
HEPA filters are not designed to be effective against vapours. Therefore, some isolator exhausts include activated carbon filters to prevent the release of cytotoxics in the vapour phase. A recent research project at the University of Bath (UK), as yet unpublished, has shown that the main “suspects” – cyclophosphamide and 5-fluorouracil, traces of which have been detected in pharmacy cleanrooms – do not exist in vapour phase at or near room temperatures, nor do they sublimate. Furthermore, cisplatinum, which was found to vapourise at room temperature, was not adsorbed by the particular activated carbon filter under test. Therefore, carbon filters do not have any part to play in preventing the release of cytotoxics, and there must be other causes for the traces of cytotoxics that have been found.
The EC GMP specifies that the background environment for an isolator should be at least Grade D. This is especially important for negative-pressure isolators. Some isolator room layouts place just the operator access in the isolator cleanroom (see Figure 1), with the rest of the isolator backing into a plant room. There may appear to be good logic in such an arrangement, but in this case the plant room would also have to be at least Grade D in case of leaks!
Construction and materials of construction
The essential requirement for an isolator design is that it should be easy to clean inside and out and that the materials of construction should be capable of withstanding regular use of cleaning agents. 316L stainless steel, as specified for product contact areas, is now widely used in isolator construction. There may also be a future for plastic materials that are resistant to chlorine-based disinfectants, especially for isolators to be used with genetically modified organisms (GMOs).
Design for validation
Validation is “the accumulation of documentary evidence to show that a system, equipment or process will consistently perform as expected to a pre-determined specification, and will continue to do so throughout its life cycle”. Design to the user requirement specification (URS) is a key stage of the validation process from which much else flows. For gassed isolators, cycle development, using biological indicators (BIs), is another key stage. Performance qualification (PQ) is the ultimate test of design and cycle development.
- ISO 14644-7: Cleanrooms and associated controlled environments. Part 7: Separative devices (clean air hoods, gloveboxes, isolators, minienvironments). 2004.
- EC guide to good manufacturing practice, revision to Annex 1: Manufacture of sterile medicinal products. 2003. Available at: pharmacos.eudra.org/F2/eudralex/vol-4/home.htm
- Pharmaceutical inspection convention/pharmaceutical inspection co-operation scheme, PI 014-1, 24 June 2002, recommendation: Isolators used for asepticprocessing and sterility testing. Available at: www.picscheme.org/index.htm
- Midcalf BM, Phillips WM, Neiger JS, Coles TJ. Pharmaceutical isolators. A guide to their application, design and control. London: Pharmaceutical Press; 2004.
- ISO 14644-1: Cleanrooms and associated controlled environments. Part 1: Classification of air cleanliness. 1999.
- ISO 14644-4: Cleanrooms and associated controlled environments. Part 4: Design, construction and start-up. 2001.
- EN 1822-1: High efficiency air filters (HEPA and ULPA) – Part 1: Classification, performance testing, marking. 1998.
- Parenteral Society.Technical Monograph No. 2: Environmental contamination control practice. Revised 2002.