Continuous-Microbial-Air-Monitoring-Application-Note-en-L-Sartorius

Continuous Microbial Air Monitoring in Clean Room Environments Claudia Scherwing017%, Jasmin Bunke2??5 1. Product Development Lab Consumables Microbiology, Sartorius Stedim Biotech, Göttingen, Germany 2. Product Management Lab Essentials Microbiology, Sartorius Lab Instruments, Göttingen, Germany * Correspondence E-Mail: kai.nesemann@sartorius.com Abstract Environmental monitoring is an important part of quality assurance for the production environments of sterile pharma­ ceutical products. Especially for aseptic filling lines where products are filled without a terminal sterilization step it is of utmost importance for product safety and thus an essential part of the quality control strategy. Such ISO 5 graded manufacturing environments are required to have < 1 colony­forming unit (?`#BCFU?pR?@ per m³ of air. A typical method for monitoring contamination of air is to actively draw air and filter it through special gelatin filters. According to Annex 1 to the EU GMP guide a minimum sample volume of 1 m³ of air should be taken per sample location. Considering an 8 hours work shift 1 m?w? is a too low sample volume to reliably judge the air quality of the manufacturing environ­ ment. One approach to improve product safety would be the implementation of a continuous air monitoring covering the complete production process (at multiple sampling points). Unlike agar plates, which would dry out during long­term sampling, the Gelatin membrane filters can be used for the whole 8 h period. Human intervention, such as change of agar plates, could then be avoided, thus lowering the risk of secondary contaminations to nearly zero. February 2019 Keywords: continuous air monitoring, clean room, gelatin filter, MD8 Airscan ®, EU GMP Guideline Annex 1 Find out more: www.sartorius.com Application Note 2 Introduction The following study aimed to establish whether a continu­ ous sampling (and multisampling point assay) provides effective monitoring for the entire production process (?`#B8 h?pR?@ by determining whether trapped organisms can withstand long­term drying stress with unaltered recovery. This study examined the recovery and viability of micro organisms captured on gelatin filters during 8 h of filtration with HEPA­filtered air from a laminar flow hood, using the MD8 Airscan ® system. Stressed and unstressed filters were compared with parallel­run reference filters as controls. The CFU were counted and the genus of the identified micro­ organism populations determined to examine any changes in microbiological flora occurring during continuous long­ term sampling. Compared to the unstressed reference filters, neither total recovery nor recovered bacterial diversity changed. No sta­ tistically significant differences in CFU/m?w? were found be­ tween test filters and reference filters, and no differences in the microbiological flora between test filters and reference filters. CFU populations were comparable. 8 h continuous air sampling on gelatin filters with the MD8 Airscan ® system did not affect total recovery or change the diversity of recovered microorganisms when comparing test filters to reference filters. Monitoring microbiological contamination of air in pro­ duction areas is of major importance because aseptic filling is the step in the production process of the pharmaceuti cal industry that harbors one of the highest risks for contami­ nation2??5 . Aseptic filling lines are increasingly used in the phar­ maceutical industry because increasing numbers of bio­ technology products cannot be sterilized after production without the sterilization process affecting their quality. Fill­ ing lines are defined as ISO 5017%, and air actively sampled in these environments must have less than one colony forming units per cubic meter (?`#BCFU/m?w??pR?@, with a minimum sample volume of one m?w? of air taken per sample location, according to Annex 1 to the EU GMP guide. Considering an 8 hour work shift, one m?w? may be too low a sample volume to reli­ ably judge the air quality of the manufacturing environment. Thus, the development of a continuous production­moni­ toring tool to minimize risks for contamination and increase the overall standard of quality control is required. A method is needed, which continually surveys all cycles of the pro­ duction process and allows sampling at multiple points. To determine if continuous air sampling using gelatin mem­ branes can effectively monitor the entire production pro­ cess over an eight hour shift, the viability of microorganisms on gelatin filters during sterile air long­term filtration, i.e. whether trapped organisms can withstand long­term drying stress and yield unaltered recovery, was examined. Former tests showed that gelatin filters with an inlet velocity of 0.25 m/s had average retention rates of 99.9995 % for Bacillus subtilis varniger spores and 99.94 % for T3 coliphages ` @?. Figure 1: Comparison of mean CFU on test and reference gelatin filters.Comparison of mean CFU on test and reference gelatin filters n = 26 0 130 120 110 100 90 80 70 60 50 40 30 20 10 Ref Test CFU 3 Materials and Methods The study examined whether the viability of microorgan­ isms on gelatin filters was maintained during the long­term filtration of filtered air. The expression “filtered air” describes the ISO 5 HEPA­filtered air of the used Class 2 biological safety cabinet. Test and reference gelatin filters were first exposed to non­ sterile air for 30 minutes. The MD8 Airscan ® air samplers (?`#Bset at an air flow rate of 2.0 m?w?/h (?`#B0,144 m/s?e$? had been located in a non­controlled laboratory environment (auto­ clave room) approx. 30–40 cm apart from each other. This sampling location had been chosen in order to build up special environmental conditions. There, a higher relative humidity (?`#B~ 57 ± 6 % and temperature: ~ 21 ± 1 °C?pR?@ was ex­ pected (thus increased amount of drying stress sensitive, waterborne microorganisms (e.g. gram­, generating a “worst case” scenario). Further, a general higher content of airborne microorganisms per cubic meter was expected than in the “normal” laboratory. Because of that, it was pos­ tulated that the following 8 hours of drying stress would show a clearly visible and statistical detectable effect. Following, the test filters were used to sample filtered air for a further 8 hour period. For the filtration of ISO 5 graded air, the MD8 Airscan ® sampling heads were placed under a laminar flow hood (?`#B relative humidity: ~ 43 ± 3 % and temperature: ~ 23 ± 1 °C?pR?@, thus, there was no additional high relative humidity while the 8 hour stressing. The reference filters were subjected to only 30 minutes filtration of non­sterile air without further aeration. They were placed on soybean­casein­digest agar medium directly after sampling. At the end of the 8 h filtration period under the laminar flow hood, the test filters also were placed on soybean­casein­ digest agar medium plates and incubated at 32 °C for 4 days. The colonies that developed were counted and recorded as CFU/m?w? a total of 26 times. Then, the CFU/m?w? were compared for the test and reference filters. Additionally, the genus of each colony was identified to determine if the microbiological flora had changed during continuous long­ term sampling. Results Figure 1 shows the mean CFU/m?w? on test (gold bar, mean = 69 colonies, sd = 51 colonies) and reference filters (grey bar, mean = 64 colonies, sd = 32 colonies). A mean difference of 5 CFU/m?w? (not statistically significant according to the paired T ­test) was found, but observed no general trend upon comparison of test and reference filters (?`#Bsee Fig. 2?pR?@. In 12 cases, there were more CFU/m?w? on test filters than on reference filters, but the opposite was examined in 13 cases (?`#Bsee Fig. 2?pR?@. The standard deviation in test and reference filters can be attributed to the broad fluctuation of micro­ organisms naturally occurring in the ambient air of non­con ­ trolled environments. CFU 26 25 24 23 22 21 20 Figure 2: Comparison of CFU on the paired test and reference gelatin filters. Comparison of CFU on the paired test and reference gelatin filters 1 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 0 260 240 220 200 180 160 140 120 100 80 60 40 20 Test Ref 4 No statistically significant difference in the growth of micro­ organisms on test versus reference filters could be observed. Figures 3 and 4 show a representative soybean­casein­ digest agar medium plate with microbiological flora grown on the paired test (left) and reference filters (right). This visual impression shows that the microbiological population found on the test and reference filters is comparable. The genus identification data from a macroscopic comparison of the microbiological flora shown in Figure 5A and 5B con­ firms the visual impression that the microbiological popula­ tion on the test and reference filters is comparable. No statistically significant difference in mean CFU/m?w? be­ tween test and reference gelatin filters. The gold bar shows a mean CFU/m?w? of 69 colonies, with a standard deviation (sd) of 51 colonies for the test filters (counted 26 separate times). The grey bar shows a mean CFU/m?w? of 64 colonies, with an sd of 51 colonies for the reference filters (counted 26 separate times). The mean difference of 5 CFU/m?w? be­ tween test and reference gelatin filters was not statistically significant. No general trend of CFU/m?w? upon comparison of test and reference filters. The gold bar shows CFU/m?w? for 26 repli­ cates of the test filters, and the grey bar shows CFU/m?w? for the reference filters. Figure 3: Comparison of the microbiological flora grown on a test filter (left) and its corresponding reference filter (right). The composition of the microbiological population found on the test and reference filters is comparable. Representative soybean-casein-digest agar medium plates showing the microbiological flora grown on a test filter (left) and its corresponding reference filter (right). Figure 4: Comparison of the microbiological flora grown on a test filter (left) and its corresponding reference filter (right). The composition of the microbiological population found on the test and reference filters is comparable. Representative soybean-casein-digest agar medium plates showing the microbiological flora grown on a test filter (left) and its corresponding reference filter (right). Cocci 91 % Cocci 90 % Gram - 3 % Gram - 3 % Yeast 3 % Yeast 4 % Sporeforming 0 % Sporeforming 1 % Mold 3 % Mold 1 % Staphylococci 0 % Staphylococci 1 % A B 5 Conclusion This study aimed to examine if gelatin filters manufactured by Sartorius Stedim Biotech GmbH are qualified for long­term (eight hours [8 h]) air sampling in production environments in the pharmaceutical industry. Specifically, if microorgan­ isms collected on gelatin membranes can survive long­term filtration with filtered air. The 8­hour filtration period is rep­ resentative of a typical work shift on an aseptic filling line. The focus of the study aimed to establish whether long­ term air filtration decreased the number of CFU/m?w? on filters. Therefore, a non­sterile air sampling on test filters for 30 minutes, followed by a filtration of ISO 5 graded air for 8 h. The experiment provided no statistically significant differ­ ences between test (stressed) and reference (unstressed) filters. The test filters had the same number of CFU/m?w? as the reference filters (i.e., no microorganisms died during long­term filtration). The standard deviations in test and reference filters were attributable to the broad fluctuation of microorganisms naturally occurring in the ambient air of non­controlled environments. Moreover, no difference between the bacterial flora grown on the test and reference filters in either visual comparison or macroscopic com­ parison could be detected. Even gram­negative bacteria were found on stressed test filters. No statistical difference between stressed and unstressed gelatin filters. In conclusion, this study showed that there was no statis­ tical difference between stressed and unstressed gelatin filters, thus proving that gelatin membranes manufactured by Sartorius Stedim Biotech GmbH are qualified for contin­ uous air monitoring in industrial pharmaceutical production environments covering a whole 8 h work shift without the need for human intervention. References ­ (?`#BDraft of the Revision of?pR?@ Annex 1 of the EU Guidelines to Good Manufacturing Practice, November 2008 (?`#BDecember 2017?pR?@ ­ USP, Chapter 1116 – Microbiological Evaluation of Cleanrooms ­ C. Scherwing, F. Golin, O. Guenec, K. Pflanz, G. Dalmaso, M. Bini, F. Andone, Continuous microbiological air moni­ toring for aseptic filling lines, PDA J Pharm Sci Technol, March | April 2007 61: 102–109 ­ CAMR Report 1993 “An assessment of the Sartorius MD­8 Microbial Air Sampler” Figure 5: A. Composition of the microbiological population grown on the test gelatin filters. Almost all microbes grown on test filters is Cocci. This figure shows a breakdown of microbes grown on soybean-casein- digest agar medium plates from test filters. B. Composition of the micro- biological population grown on the reference gelatin filters. Almost all microbes grown on reference filters are Cocci. This figure shows a break- down of microbes grown on soybean-casein-digest agar medium plates from reference filters. For further contacts, visit www.sartorius.com Specifications subject to change without notice. Copyright Sartorius Lab Instruments GmbH & Co. KG. Status: 10 | 2020 Germany Sartorius Lab Instruments GmbH & Co. KG Otto-Brenner-Strasse 20 37079 Goettingen Phone +49 551 308 0 USA Sartorius Corporation 565 Johnson Avenue Bohemia, NY 11716 Phone +1 631 254 4249 Toll-free +1 800 635 2906