European guidelines for sterile manufacture

No document relating to the field of cleanroom technology has stirred up as much controversy as EC GMP Annex 1 in its various editions, especially since the start of the phased publication of the EN ISO 14644 'Cleanrooms and associated controlled environments' series of international standards1. Issues that will, it is hoped, be resolved in the new revision, which is due to appear this summer, include:

• full harmonisation with the ISO classification of air cleanliness,
• elimination of the questionable requirement to carry out particle counts for particles of 5 µm,
• adoption and acceptance of the term “unidirectional airflow” in place of the misused and incorrect term “laminar flow” and
• guidance on the specification and in-situ testing of HEPA filters.

EC GMP Annex 12 has much to be said for it. It specifies four grades of environment for the manufacture of sterile medicinal products, depending on the activity, and it gives air cleanliness limits in terms of particle counts for each of these in both the “at rest” state and the “in operation” state with the additional requirement that the “particulate conditions for the "at rest" state should be achieved after a short ‘clean up’ period of 15 - 20 minutes in an unmanned state after completion of operations”.

This is exactly the sort of application-specific guidance that was anticipated by the expert authors of the generic ISO standard, which simply states, “the particulate cleanliness in a cleanroom or clean zone shall be defined in one or more of three occupancy states, viz. ‘as-built’, ‘at rest’, or ‘operational’.”

Annex 1 goes on to give recommended limits for microbial contamination in the form of a table with criteria for four different test methods. This again is extremely helpful as the first two test methods, air samples and settle plates, measure airborne contamination and the second two, contact plates and glove prints, measure surface contamination, thus emphasising that, however well a cleanroom (or a piece of clean air equipment) functions, good aseptic techniques and good cleaning are absolutely essential.

All this should have formed the basis for a really useful guidance document, but unfortunately, until now, the European Agency for the Evaluation of Medicinal Products (EMEA) has chosen neither to harmonise Annex 1 with the ISO classification, test methods and terminology, nor to take on board some of the points that have been strongly argued by technical experts in the marketplace. Hopefully this will all change in the new revision due this summer. Here are some of the contentious issues.

Classification of air cleanliness: Table 1 shows the divergence between the Annex 1 (2003 revision) and the ISO classifications. The differences for 0.5µm particles are barely material and therefore all the more irritating. The big problem arises with 5µm particles where the Annex 1 limits in Grade A areas require excessive and impractical sampling times. (The need for testing at 5 µm at all is questioned in the next section). Is there any valid reason why the maximum permitted particle counts for Grade A “at rest” and “in operation” and Grade B “at rest” and cannot be to ISO Class 5?

Harmonisation would make everything so much simpler for manufacturers and users of particle counters and for everyone concerned with the classification and monitoring of cleanrooms and clean air equipment.

Classification, monitoring and the 5 µm particle

The present inclusion, in Annex 1 of particle count limits at 5µm leads to very lengthy sampling times for classification and even greater problems in continuous monitoring. Two complementary papers in the latest volume of the European Journal of Parenteral and Pharmaceutical Sciences argue a case, based on scientific evidence, for the abandonment of particle counting at 5µm.

The first, by T Eaton3 demonstrates how “adequate and more appropriate particle monitoring can be achieved by recording particles at the smaller (>0.5µm) particle size only” and the second, by W Whyte and M Hejab4, which describes an experiment in which the dispersion of MCPs (microbe carrying particles) from 55 people was measured in a dispersal chamber, showed that “there was no situation where the dispersion of MCPs was not accompanied by substantial numbers of both >0.5µm and >5.0µm particles”.

Furthermore, the US FDA Aseptic Processing Guideline5 specifies a limit for >0.5µm particles only. It is to be hoped that those responsible for new Annex 1 will take account of the evidence and adopt this more practical and realistic approach.

Unidirectional airflow

The term “laminar flow” should not be used in clean air technology and has different meaning in other branches of science and technology, such as aeronautical engineering. The correct term is “unidirectional airflow” - defined in the ISO standards as “controlled airflow through the entire cross-section of a clean zone with a steady velocity and approximately parallel streamlines. It should be noted that this type of airflow results in a directed transport of particles from the clean zone. Likewise the term “uniformity of airflow” should be used in place of “laminarity”, which is equally incorrect. ISO defines “uniformity of airflow” as “unidirectional airflow pattern in which the point to point readings are within a defined percentage of the average airflow velocity”. The average airflow velocity and the percentage are for application-specific guidelines such as Annex 1 to define. The wording in the present Annex 1 could be sharpened here.

A further point is that the present Annex 1 makes a distinction between laminar airflow as used in “laminar air flow workstations” and “uni-directional air flow and lower velocities” which “may be used in isolators and glove boxes”. Apart from the misuse of terminology, this statement itself is questionable. The author has carried out an experiment with airflow inside an isolator which suggests that unidirectional airflow operates more effectively at similar velocities to those specified for open systems, especially with respect to the “directed transport of particles from the clean zone”.

filter leakage tests

One question that is frequently asked of the author, although it may not have been prominent in the extensive public response on Annex 1, is what grade of HEPA filter should be specified for a pharmaceutical cleanroom or isolator and what should be the maximum permissible penetration in an installed filter system leakage test using a DOP (Dispersed Oil Particle) aerosol test.

BS 5295: 1989 used to recommend a maximum permitted penetration of 0.001% for clean zones up to what is now ISO Class 5 and 0.01% for higher (less clean) Classes. BS 5295: 1989 was withdrawn upon the publication of EN ISO 14644-1: 1999, and PD 6609: 20006, which carried forward this requirement pending the publication of EN ISO 14644-3: 2005 - Test methods, will shortly be superseded by PD 6609: 2007, which will not specify a maximum permitted penetration.

The only guidance that will therefore exist is in EN ISO 14644-3: 2005, which states “Designated leaks are deemed to have occurred where a reading greater than 0.01% of the upstream challenge aerosol concentration [sic]. Alternative acceptance criteria may be agreed between the customer and the supplier”.

It would be very helpful if Annex 1 were to include guidance on the permitted limit (just as there is already guidance on airflow velocities for unidirectional airflow) as there are currently two views: -

a) The UK should fall in line with the rest of the world and adopt 0.01%

b) The UK should stay with the tried and tested 0.001%

The author understands that the next edition of Quality Assurance of Aseptic Preparation Services7 may include a recommendation that aseptic units in the UK NHS stay with 0.001% based on experts' evidence.

This short article provides a wish-list of what the industry is hoping for from the new Annex 1 when it is published later this summer.

Technical experts from all the major industrial nations, and from a broad cross-section of industries, have combined their considerable knowledge and experience to produce the EN ISO 14644 series of cleanroom standards, covering classification, testing and many other aspects of cleanroom technology. These carefully developed generic standards, along with their methodology and language, should be recognised, respected and fully utilised when industry-specific guidelines such as Annex 1 are produced.