Air filtration challenges and answers for dry heat sterilisation tunnels

Transcription

1 European Journal of Parenteral & Pharmaceutical Sciences 2015; 20(1): Pharmaceutical and Healthcare Sciences Society Air filtration challenges and answers for dry heat sterilisation tunnels Marc Schmidt* 1, Lothar Gail 2, Hugo Hemel 1 1 AAF International, Oberhausen, Germany; 2 VDI, Düsseldorf, Germany Dry heat sterilisation and depyrogenisation is considered to be one of the more critical process steps in medicine manufacturing, contributing to ensuring the sterility of pharmaceutical sterile (terminal sterilisation) and aseptic preparations. Inside such sterilisation and depyrogenisation tunnels, highefficiency particulate air (HEPA) filtration plays an indispensable role in protecting containers, such as vials or prefilled syringes, from contamination that might result in severe health risks for patients. Where the installed HEPA filter has to withstand frequent temperature cycles between ambient and up to 350 C, operating conditions are challenging. To manage the challenges, and, therefore, not to adversely affect manufacturing throughput and product quality and safety, a careful selection of the high temperature HEPA filter is required. This paper describes the key challenges that have to be accommodated for dry heat sterilisation and depyrogenisation processes, and will present two selection criteria that have been found to be most important for a high temperature HEPA filter. The insights are based on extensive interviews that were conducted with both tunnel equipment manufacturers and pharmaceutical end users. Supported by various test cycles, through which air filter durability and particle shedding was defined, a new filter design will be presented that resolves well-known issues with the traditional high temperature HEPA filter design that has served the market for many years. The performance of the new filter design was proven by a field test in an existing dry heat sterilisation tunnel, of which the results will be presented. This paper will therewith provide new insights into how to mitigate process contamination risk by applying new, but proven, HEPA filtration technology. It will support the pharmaceutical industry to obtain a more stable and reliable dry heat sterilisation and depyrogenisation process. Key words: Dry heat sterilisation, dry heat depyrogenisation, high temperature HEPA filtration, filter design, filter durability, particle shedding, filtration efficiency. Dry heat sterilisation and depyrogenisation Critical process step in sterile manufacturing The pharmaceutical production of sterile medicine is subject to special requirements to minimise risks of particulate and microbial contamination. Stringent US Food and Drug administration (FDA), European Union (EU), and other good manufacturing practice (GMP) guidelines are in place requiring limits and control of exposure to such contamination, ultimately preventing harm or life-threatening health risks to the patient. Dry heat sterilisation and depyrogenisation are applied to ensure sterility of pharmaceutical aseptic preparations, as *Corresponding author: Marc Schmidt, AAF International, Centroallee 263b, Oberhausen, Germany; marc.schmidt@aafeurope.com imposed by FDA regulation 21 CFR and EU Guideline 2003/94/EG, Annex 1. For aseptic preparations, of such presentations as vials, ampoules, cartridges or prefilled syringes, terminal sterilisation of the final container is not possible. The glassware, therefore, has to be rendered free from harmful contaminants that might affect the medicine before filling. Depending on the process, dry heat sterilisation and depyrogenisation may be applied. Sterilisation is applied to render a product free from living organisms, taking place above 180 C. Depyrogenisation aims to remove or inactivate endotoxins for which higher temperatures of C are required, taking place in either static ovens or in tunnels for automated, continuous processes. Because of the increasing demand for pyrogen-free sterile packaging and for fast, safe and efficient processing, dry heat sterilisation and depyrogenisation nowadays represents one of the most 13

2 14 MARC SCHMIDT, LOTHAR GAIL, HUGO HEMEL critical steps in the sterile medicine manufacturing process*. Pyrogen-free packaging material was originally demanded merely for the filling of large volume parenteral containers; but has become a standard for the whole field of sterile filling 1. Modern demands on sterilisation processes are set down by the US FDA and EU, requiring temperature programs which demonstrate, for example, that the endotoxin substance has been inactivated to not more than 1/1000 of the original amount (3 log cycle reduction) 2. This demand contributed decisively to the development of safe, fast and efficient dry heat sterilisation processes including unidirectional airflow with HEPA filtration. The protecting role of HEPA filtration Proper HEPA filtration removes particles and considerably improves the heat distribution. The same measure, however, causes a severe risk of particle contamination due to released particles on the clean side of HEPA filters. This issue has to be considered. The dry heat sterilisation of glassware typically follows a three-step approach (Figure 1): infeed, heating and cooling. Though dry heat sterilisers are typically installed in GMP grade D areas, the industry imposes grade A demands on the whole glassware transportation line between washing and filling. Therefore, in these areas any air admitted has to pass through a HEPA filter 3. In particular, the heating process, taking place in the hot zone in which the hot air primarily transfers heat to the glassware, poses high demands on HEPA filters at temperatures up to 350 C. But even in the cooling zone, the installed HEPA filters have to withstand temperatures of C in case of sterilisable cooling sections. * The remainder of the paper will speak about dry heat sterilisation as an umbrella term for both dry heat sterilisation and dry heat depyrogenisation. Challenges in terms of filter durability and efficiency have to be mitigated to guarantee sterility of the containers that leave the sterilisation tunnel. Challenges to be mitigated Performance improvement issues dominate the priority list of the pharmaceutical industry. Challenges related to reducing time to market, increasing manufacturing throughput, quality requirements on cleanliness, complying with applicable regulations and reducing costs are of high concern. The performance of a dry heat sterilisation tunnel has a direct influence on all these critical issues. The degree to which a sterilisation tunnel is able to remove pyrogens from the washed glassware in an effective, efficient and repeatable manner is, in turn, highly dependent on the function of HEPA filtration 4. Traditional dry heat sterilisation equipment, performing, for example, infrared heating, showed deficits in terms of cleanliness, equipment size, processing time and precisely controlled temperature programs. Unidirectional airflow with HEPA filtration was found to be the far more suitable technique for the various challenges of dry heat sterilisation 5. Final filtration of the circulated air stream should enable a consistent airflow for a faster and more simultaneous heating of the glassware. However, the air filter has to withstand integrity challenges caused by high variations in operating temperature during heating and cooling. Process contamination and resulting unscheduled downtime from bypass of unfiltered air, leaks or shedding of particles has to be prevented. The challenge to limit particle shedding can be particularly critical in cases of temperature fluctuations that arise from emergency shutdowns or interruption of power supply. Controlling the challenges during exposure to high temperature and frequent heating and cooling cycles is no sinecure for a HEPA filter: ISO 5 conditions in operation state are being stipulated and determined to demonstrate that HEPA filters are effective and free from cracks or other defects. An overview with the applicable particle limits per ISO cleanroom class is presented in Table 1. Table 1. Airborne particulate cleanliness classes to ISO :1999. ISO classification number (N) Maximum concentration limits (particles/m 3 of air) for particles equal to and larger than the considered sizes shown below 0.1 µm 0.2 µm 0.3 µm 0.5 µm 1.0 µm 5.0 µm ISO Class ISO Class ISO Class ISO Class 4 10, ISO Class 5 100,000 23,700 10, ISO Class 6 1,000, , ,000 35, ISO Class 7 352,000 83, ISO Class 8 3,520, ,000 29,300 ISO Class 9 35,200,000 8,320, ,000

3 AIR FILTRATION CHALLENGES AND ANSWERS FOR DRY HEAT STERILISATION TUNNELS 15 Figure 1. Schematic representation of a sterilisation tunnel. High temperature HEPA filtration Considerations for selecting the right solution Several characteristic requirements for high temperature HEPA filters can be identified that directly influence the productivity of a sterilisation tunnel. From various indepth interviews conducted with tunnel manufacturers and pharmaceutical end users, two HEPA filter requirements have been found most critical in especially the hot zone of a sterilisation tunnel. 1. High stiffness and durability of construction. 2. Proven efficiency performance during operation. A high stiffness and durability of construction should assure that the integrity of the HEPA filter is retained during elevated temperatures. Filter design and material selection should be such that degradation does not occur and thermal expansion and contraction do not create stress cracks. Integrity breaches, caused by stress cracks, should be avoided at all times as these might result in bypass, particle shedding and process contamination. A compliant efficiency performance of the HEPA filter should not only be confirmed from a factory efficiency test by the filter manufacturer, but confirmed in situ by an installation leak test. A consistent filtration efficiency should be retained during multiple heating and cooling cycles, whereby the particle counts are compliant with the ISO Class 5 limits in operation state as set by standard ISO :1999. Available solutions to the challenges Historically, the number of options available to tunnel manufacturers and pharmaceutical end users for HEPA filters that can withstand temperatures up to 350 C has been limited. This section will compare two high temperature HEPA filter options, with a reflection on the selection considerations as introduced before. The first option (A) is a filter design which has served the pharmaceutical industry for many years. This filter type does, however, possess some generally known deficits, directly attributable to its construction and the components used. The second filter option (B) comes in a new design, which promises better results on the requirements of longterm durability and efficiency during operation. Option A: HEPA filter with ceramic sealant and aluminium separators Traditionally, the most commonly used filter option for sterilisation tunnels comes with a ceramic sealant and fibreglass filter medium (Figure 2). The medium is folded around aluminium separators which, in turn, are placed in parallel in a stainless steel frame. Because ceramic material is used to seal the aluminium separated media pack to the frame, this filter type is sensitive to the formation of stress cracks. The cracks occur between the ceramic glue and the filter frame due to internal stresses created by processdriven temperature cycles (Figure 3). In order to try to absorb the movement of the filter components during elevated temperatures, the filter is equipped with a compensation mat directly under the sealant. The stress cracks can cause leaks that result in bypass of unfiltered air. Particle emission is created when the edges of a sealant crack are rubbing each other due to normal fan and motor induced vibration of the whole construction. Thus, the cracks also generate particles themselves. Such uncontrolled shedding into the tunnel, Figure 2. Filter design A. Fibreglass medium Ceramic sealant Compensation mat Glass paper Parallel aluminium separators

4 16 MARC SCHMIDT, LOTHAR GAIL, HUGO HEMEL Figure 3. Photograph of a stress crack in filter design A. which can be increased during even very small temperature fluctuations, results in contamination of the sterilisation process. The air filter industry has worked together with the tunnel manufacturing industry to look for an alternative solution to mitigate the risks from the occurring stress cracks. This has resulted in a so-called dynamic seal, in which spacers are positioned onto the installed filter at the clean air side. With this counter measure, the filtered air (which can include bypass air and released ceramic particles) is taken out from the overpressure situation in the oven to the underpressure situation outside the oven. Although this countermeasure effectively reduces the problem, it is not a structural solution to the intrinsic issue with the filter design itself: for a more robust result, the issue should be solved at the source. The same is true for installing a fine mesh immediately downstream of the (clean) filter outlet. Such a mesh shall protect clean glass containers on the conveyor belt from being contaminated by larger particles, released from the installed filter and applied gasket. Option B: HEPA filter with elastic fibreglass sealant and stainless steel separators Similar to filter design A, filter design B comes with a stainless steel frame, a fibreglass media pack and is free of silicones. The important differences are in the applied separator and sealant material. Filter design B is equipped with corrugated stainless steel separators and with stainless steel support bars and stays (Figure 4). This construction gives a high overall durability, which is less prone to oxidation. Risk of oxidised particles to shed off the separators on the air leaving side is reduced. The separators are placed in staggered position to increase the media pack stiffness and to prevent the separators from nesting. Applying integrated stainless steel stiffener plates and stays prevent winding of the bottom of the pleats. In addition to stainless steel separators and stiffeners, filter design B includes an elastic fibreglass sealant in contrast to the vulnerable ceramic sealant of filter design A. With the inclusion of an elastic fibreglass sealant, the HEPA filter is better able to compensate the forces from heat stretching of components. Integrity breaches from stress cracks are prevented. Demonstrated performance of new filter design Introduction This section will elaborate on the results of various tests that have been conducted on the performance of high temperature HEPA filter design B. First, the outcome of a heat cycle test, strength test and breach test will be presented that determine the stiffness and durability of construction. Second, results of a laboratory particle shedding test will be presented that compares filter design B with filter design A. The filtration efficiency performance of filter design B was confirmed during a field test in a dry heat sterilisation tunnel. The described tests were conducted with air filters in either 150 mm or 290 mm depth. Stiffness and durability of construction Heat cycle test A heat cycle test was set up to assess sensitivity to damage of the filter construction under high temperature fluctuations. Twenty consecutive test cycles were executed at a nominal airflow rate of 2100 m 3 /h. Each cycle had a duration of 4 hours, during which the filter was exposed to a temperature increase from 40 C to 350 C, at which it was kept for 1 hour before cooling back to 40 C (Figure 5). The results demonstrated insignificant differences between pre-test and post-test values for both filtration efficiency and pressure drop (see Table 2). No damage was observed on the media and no stress cracks were found in the sealant. Although a tempering colour did appear on the stainless steel parts due to heating (Figure 6), it seemed to Table 2. Pressure drop and efficiency before and after heat cycle test. Fibreglass medium Elastic fibreglass sealant Angular stainless steel corrugated separators Figure 4. Filter design B. Test Pre-test Post-test Filter dimensions 610 x 610 x 290 mm Pressure drop (Pa) Efficiency at 0.3 µm (%)

5 AIR FILTRATION CHALLENGES AND ANSWERS FOR DRY HEAT STERILISATION TUNNELS 17 First cycle Second cycle 1.5h About 4h About 6.5 Figure 5. Heat cycle test procedure. have no effect on the actual performance. Filter design B with elastic fibreglass sealant and stainless steel separators proved its durability of construction. Strength test Following the heat cycle test, a strength test was performed to understand how filter design B would Pre-test respond to high pressure drop increases over time due to filter fouling. The test was executed under ambient conditions. For demonstrating the elasticity and the strength of the filter construction, the pressure drop was increased by 250 Pa at intervals of 5 minutes to reach a maximum of 4000 Pa (Figure 7). Similar to the heat cycle test, the results of the strength test showed insignificant differences between the pre-test and post-test situation (see Table 3). After withstanding pressures of up to 4000 Pa during the 75 minutes, both the pressure drop and filtration efficiency values of the tested air filter remained on the same level as measured prior to the test. No visible damage was observed on the air filter and its components after the strength test. The strength test validated the sturdy construction of filter design B and confirmed its integrity during challenging conditions. Table 3. Pressure drop and efficiency before and after strength test. Post-test Test Pre-test Post-test Filter dimensions 610 x 610 x 150 mm Pressure drop (Pa) Efficiency at 0.3 µm (%) Figure 6. Photographs of undamaged filter design B before and after heat cycle test. Breach test Complementary to the strength test, a dry breach test was executed to confirm the robustness of construction at elevated temperatures. At a constant temperature of 300 C and a nominal airflow rate of 1440 m 3 /h, the high temperature HEPA filter in design B was challenged to a pressure drop of 1200 Pa for 1 hour. The air filter

6 18 MARC SCHMIDT, LOTHAR GAIL, HUGO HEMEL Figure 7. Strength test procedure. Table 4. Pressure drop and efficiency before and after breaching test. Test Pre-test Post-test Filter dimensions 610 x 610 x 150 mm Pressure drop (Pa) Efficiency at 0.3 µm (%) long sampling tube, consisting of a stainless steel tube and a PTFE (poly-tetra-fluoro-ethylene) tube, was attached downstream of the air filters. By passing through the tube, the air was cooled. Both air filters were not challenged upstream by an aerosol. The filtration efficiency target, measured downstream, was set on 99.99% for 0.3 μm particles. Although the number of particles counted downstream increased during heat stretching, the number of particles counted for filter design B clearly stayed below the target demonstrated that it retained its key performance characteristics in a high air pressure environment (see Table 4). Proven performance on efficiency Particle shedding test Particle shedding caused by the high temperature HEPA filter in operation should be prevented at all times. Any uncontrolled release of particles can result in process contamination that might negatively affect air quality and could lead to loss of production batches as well as to undesired downtimes and recalls. To determine the filtration efficiency performance of filter design B versus filter design A during high temperature, and therewith indicate presence of particle shedding, a comparative particle measurement test was executed (see Figure 8). Prior to the test, both filter designs were burned in at 400 C for 1 hour. Filter design B (with stainless steel separators and fibreglass sealant) and filter design A (with aluminium separators and ceramic sealant) in dimensions of 610x610x150 mm were then installed in a high recirculation test unit. At a nominal airflow of 1440 m 3 /h, temperature was gradually increased from ambient to 350 C by 3.5 C/min. The temperature was stabilised at 350 C after which it was reduced again to the ambient level. During each phase, particles 0.3 μm were counted downstream by means of a laser particle counter. A 4 m Fan Electro-thermal heater Temperature control point Tested air filter Particle measuring point (laser particle counter) Figure 8. Particle shedding test set-up.

7 AIR FILTRATION CHALLENGES AND ANSWERS FOR DRY HEAT STERILISATION TUNNELS 19. Temperature Filter design A with aluminium separators and ceramic sealant Filter design B with stainless steel separators and fibreglass sealant Figure 9. Efficiency performance during heating and cooling. of 99.99% for 0.3 μm (Figure 9). In contrast, the particles counted downstream for filter design A largely exceeded the target. The peak in counted particles occurs directly after the cooling down starts. This is largely due to the integrity breaches that result from the air filter s construction with aluminium separators casted in ceramic sealant, an issue which was explained in more detail in the Available solutions to the challenges section. Filter design B does not show integrity breaches and therewith limits particle shedding. From the particle shedding test, it can be concluded that a high temperature HEPA filter in design B offers a better and more consistent filtration efficiency performance when exposed to heating and cooling. Field test in dry heat sterilisation tunnel In order to assess how filter design B would perform in practice, and to see if the beneficial outcome from the laboratory particle shedding test would be confirmed, filter design B was installed in the hot zone of an existing sterilisation tunnel at a pharmaceutical equipment manufacturing company for a field test. The primary purpose of the field test was to verify the air cleanliness in the tunnel and to see if the particle concentration would meet the acceptance criteria for ISO Class 5 conditions in operation state as per ISO :1999 (see Table 5). Three different tests were performed. First, the high temperature HEPA filter in design B was installed in the sterilisation tunnel and was tested prior to burn-in. The air filter was then burned in overnight and retested the next day and the particle counts were repeated at the same locations. Finally, the high temperature HEPA filter was challenged with a PAO (poly-alpha-olefin) aerosol of 17 million particles/ft 3 > 0.3 μm/cf of air. A filter scan was performed with a TSI Model 9310 particle counter downstream of the air filter in the hot zone (Figure 10). The surface was scanned 125 mm above the conveyor by Figure 10. Photograph of the hot and cooling zone of the sterilisation tunnel with probe marks. Table 5. ISO Class 5 particle limits according to ISO :1999. ISO classification number (N) Maximum concentration limits for particles equal to and larger than the considered sizes shown below 0.1 µm 0.2 µm 0.3 µm 0.5 µm 1.0 µm 5.0 µm ISO Class 5 (particles/m 3 of air) 100,000 23,700 10, ISO Class 5 (particles/cf of air)

8 20 MARC SCHMIDT, LOTHAR GAIL, HUGO HEMEL Table 6. Post burned-in sample test results. Not challenged Location Measured particle concentration (particles/cf of air) 0.3 µm 0.5 µm 1.0 µm 5.0 µm Hot zone A Hot zone A Hot zone A Hot zone B Hot zone B Hot zone B Hot zone C Hot zone C Hot zone C-3 6 Challenged with 17 million PAO particles/ft3 Location Measured particle concentration (particles/cf of air) 0.3 µm 0.5 µm 1.0 µm 5.0 µm Hot zone A Hot zone A Hot zone A Hot zone B Hot zone B Hot zone B Hot zone C Hot zone C Hot zone C Particle limits according to ISO : µm 0.5 µm 1.0 µm 5.0 µm ISO Class 5 limit Conclusion: Filter design B meets ISO Class 5 conditions. Table 7. Sample locations in the hot zone of the sterilisation tunnel. Conveyor direction A-1 B-1 C-1 A-2 B-2 C-2 A-3 B-3 C-3 an isokinetic probe. The tests were conducted under ambient conditions at 1 minute for each of the 9 sample locations in the hot zone. All three tests passed the requirements. The detailed results of the two tests performed after burn-in are summarised in Tables 6 and 7. With the installation of a high temperature HEPA filter in design B, the hot zone of the sterilisation tunnel convincingly met ISO Class 5 conditions in operation state in all three test cases. In particular, the measurement results for the post burn-in situation confirm a proven efficiency, after exposure to challenging conditions. Filter design A would suffer from stress cracks between the ceramic sealant and the frame. In contrast, filter design B retains its integrity and limits particle shedding by its excellent resistance to high temperatures during heating and cooling. Impact of introducing an alternative filter type For introducing an alternative filter material, pharmaceutical manufacturers require evidence of compliance with current quality standards of dry heat sterilisation. Available documents clearly demonstrate that all critical performance factors have been improved by the HEPA filter in design B. Furthermore,

9 AIR FILTRATION CHALLENGES AND ANSWERS FOR DRY HEAT STERILISATION TUNNELS 21 manufacturers of sterilisation tunnels and sterile filling equipment are well prepared to support this process by contributing their own HEPA filter qualification results, offering a significant quality improvement in dry heat sterilisation. Conclusion The exposure to elevated temperatures during dry heat sterilisation sets stringent requirements on high temperature HEPA filters. With high durability of construction and proven efficiency during operation being identified as the two most critical requirements, any failure in these areas will affect the sterilisation process directly. Risk of process contamination from bypass, leaks or particle shedding needs to be mitigated. ISO Class 5 conditions are to be met, demonstrating that the high temperature HEPA filter is effective and free from cracks or other defects. Filter design A, with ceramic sealant and aluminium separators, has served the pharmaceutical industry for many years. Nevertheless, this design does possess some generally known deficits that lead to a loss of integrity from stress cracks between the sealant and the frame. Increased risk of process contamination and premature filter replacement are the result. Multiple heat cycle, strength and breaching tests have demonstrated that filter design B, with elastic fibreglass sealant and stainless steel separators, offers an improved durability and stiffness of construction. From a comparative particle shedding test between filter designs A and B, it can be concluded that design B offers a better and more consistent filtration efficiency performance during heating and cooling. Because of absence of stress cracks, particle shedding is minimised. The beneficial results were confirmed during a field test in an existing sterilisation tunnel, where the high temperature HEPA filter in design B met the ISO Class 5 particle limit requirements before and after burn-in. A dry heat sterilisation application can be considered as being one of the most critical pharmaceutical process steps in sterile manufacturing. This paper has given an analysis of the main challenges that have to be mitigated from an air filtration perspective. It has provided new insights into how to answer the integrity challenges of high temperature HEPA filters, by which to mitigate process contamination risk and protect product quality and safety. Acknowledgements The authors would like to thank AAF International, Nippon Muki and their customers for their cooperation and insightful contributions. References 1 Forbert R, Gail L and Pflugmacher U. Neues Verfahren zur Bestimmung der Inaktivierung von Endotoxin bei der Trocken- Hitze-Sterilisation. Pharmazeutische Industrie 2005;67(5): United States Pharmacopeial Convention. United States Pharmacopeia General Chapter <1211> Sterilization and Sterility Assurance of Compendial Articles. Rockville, MD: US Pharmacopeial Convention. 3 European Commission. EudraLex The Rules Governing Medicinal Products in the European Union. Volume 4 EU Guidelines to Good Manufacturing Practice Medicinal Products for Human and Veterinary Use Annex 1: Manufacture of Sterile Medicinal Products. Brussels, Belgium: European Commission; Parenteral Drug Association. PDA Technical Report No. 3 (Revised 2013). Validation of Dry Heat Processes Used for Depyrogenation and Sterilization. Bethesda, MD, USA: PDA; Wegel S. Kurzzeit-Sterilisationsverfahren nach dem Laminar-Flow- Prinzip. Pharmazeutische Industrie 1973;35(11a):809. SMi presents their 3rd Annual Conferences on... Lyophilisation - Europe Holiday Inn Regents Park Hotel, London, UK 29 & 30 JUNE 2015 PLUS TWO INTERACTIVE POST-CONFERENCE WORKSHOPS WEDNESDAY 1ST JULY 2015 A: Spray Dry and Formulation 8.30am pm B: Defining and implementing QbD in the design space for Scale up 8.30am pm Register online or fax your registration to +44 (0) or call +44 (0)