- L'humanitaire au cœur de l'entreprise : enjeux & opportunités, retours d'expérience !
- Congrès International A3P 2018 : restitution des messages clés de la table ronde sur la révision de l'Annexe 1.
- Impact of Annex 1 revision on new vial filling line at Sanofi Pasteur Marcy-l'Étoile.
- Quantitative evaluation of microorganisms recovery from surfaces using contact plates.
- Investigation by molecular method of potentially non-compliant aseptic process simulation-media fill test (APS-MFT) and sterility test.
- Aseptic Processing Overview.
- L'industrie pharmaceutique : renforcer la culture de la qualité.
- Sauvegarde, réplication et archivage... Quelles mesures prendre pour préserver l'intégrité de vos données ?
Environment sampling for microbiological contaminants is a key component of environmental monitoring and risk characterization practices used across diverse fields of application. Many microbiological techniques are reported to validate sanitization procedures, to control surface / air cleanliness, or to evaluate the levels of microbial contamination(1).
They are based on a transfer of the microorganisms present on the surface or in the air to a culture medium. Surface sampling is assessed through 2 major- techniques: swabbing methods and culture media contact application. However, confidence in surface sampling results, both in the field and in controlled laboratory studies, has been undermined by large variation in sampling performance results(2).
Sources of variation include controlled parameters, such as sampling materials and processing methods(3–5), nature of the solid surface(6-7), nature of the contamination(8), organic or inorganic material contaminating the surface, as well as random and systematic errors. However, relative contributions of these factors remain unclear and evaluating the efficiency of commonly used microbiological techniques remains a concern(9). In this study, we investigated how the nature of the surface, the contact plate supplier and the nature of microorganisms could influence the recovery of microorganisms, using a standardized protocol. Three industry-relevant surface were tested: crystal polystyrene plastic, glass and 316L stainless steel with three contact plates suppliers : bioMérieux, BD and Merck Millipore. The bacteria S. aureus was used as a model in the first part of the study, and extend to five more strains in the second part of the study.
1. Material and methods
1.1 Tested microorganisms
Test microorganisms included in the study were sourced from culture collection and commercial calibrated strains:
- BioBall Multishot (bioMérieux)
Aspergillus brasiliensis ATCC 16404, ref. 56011
Candida albicans ATCC 10231, ref. 56013
Escherichia coli ATCC 8739, ref. 56006
Pseudomonas aeruginosa ATCC 9027, ref.56017
Bacillus subtilis ATCC 6633, ref. 56012
Staphylococcus aureus ATCC 6538, ref. 56019
- Culture collection and in-house strains (from pharma industries collection)
Comamonas aquatica ATCC 11330
Acinetobacter baumanii ATCC 19606
Enterococcus faecalis ATCC 29212
Streptococcus pyogenes ATCC 19615
Klebsiella pneumoniae ATCC 13883
The recommended dilutions of each microorganism were between 20 to 70 CFU / inoculum.
1.2 Surface testing
Different surfaces were included in the study: crystal polystyrene plastic, 316L stainless steel and glass. The crystal polystyrene plastic used is gamma irradiation sterilized. Glass and stainless steel surfaces coupons were pretreated using a standardized and reproducible procedure. The surface was first cleaned using a detergent and rinse thoroughly with distilled water. Surface coupons were then soaked in a distillated water bath at 60°C during 2 hours, rinsed with distilled water, wiped with non-woven paper and sterilized by autoclaving at 121° C for 25 minutes. This procedure was shown to be the most adequate to homogenize surface finishing.
Microorganism inoculum tested was spotted on the sterile surface (spiking volume of 50 µl), uniformly spread with a sterile loop and left to dry in a ventilated incubator of 37°C for 30 minutes. This allows obtaining a dried inoculum on the surface coupon.
Immediately after inoculum drying, a first contact plate (CT1) was applied on the spotted inoculum during 10 seconds with a pressing force equivalent to the weight of a 500 g mass. Successive second (CT2) and third (CT3) contact plates were applied on the same area with the same force and time application. The surface sampled is always 25 cm2. In parallel, an identical spiking volume is tested on 90 mm TSA plate (bioMérieux, ref. 43011). The CFU mean number enumerated on this 90 mm TSA is the inoculum control referred as CFUcontrol.
Contact and 90 mm TSA plates are incubated at 30-35°C for 3 to 5 days for bacteria and yeast. Molds are incubated at 20-25°C for 3 to 7 days.
1.3 Culture media
Contact plates used in the study were supplied by:
- bioMérieux, Irradiated Count-Tact® 3P™ Agar (CT3P), ref. 43699, batch. 1005960800;
- BD, BBL™ IC-XT Pack Trypticase™ Soy Agar with Lecithin and Polysorbate 80, RODAC™ Locking Lid (LL), ref. 257637, batch. 7222078;
- Merck Millipore, Tr. Soy Cont. A w. LTHTh-ICR, ref. 1462310200, batch. 147550.
90 mm plates used were supplied by bioMérieux, Trypcase Soy Agar (TSA), ref. 43011, several batches.
1.4 Recovery rates
Individual contact plates recovery rates were calculated as follow:
Recovery rate CT1 = CFUCT1 / CFUcontrol
Recovery rate CT2 = CFUCT2 / (CFUcontrol – CFUCT1)
Recovery rate CT3 = CFUCT3 / (CFUcontrol – CFUCT1 – CFUCT2)
Cumulative recovery rates on contact plates were calculated as follow :
Cumulative CT2 recovery rate: Cumul rate CT 2 = (CFUCT1 + CFUCT2) / CFUcontrol
Cumulative CT3 recovery rate: Cumul rate CT 3 = (CFUCT1 + CFUCT2 + CFUCT3) / CFUcontrol
1.5 Statistical analysis
Experiments were performed randomly to ensure statistical independence. The comparison of contact plates recovery rates to determine factor influence was performed using a two-way nested ANOVA test (with a risk alpha = 5%). Statistical significant differences were considered for p-value < 0,05.
The data were represented as box plot which allows to visualize:
- 1st and 3rd quartiles
- Mean and median
- Minimum and maximum value.
In this paragraph contact plates supplier names were replaced by supplier A, B & C. for non-competition reasons.
2.1 Microorganisms recovery screening
We first screen the recovery of different microorganisms, from different family, using a single contact plate application, on stainless steel surface. A known inoculum of microorganism is spotted on stainless steel surface and then dried under heat and ventilation to obtain a simulated contaminated surface. A contact plate is applied to recover microorganisms spotted. The number of CFU recovered on the contact plate is compared to the number of CFU spotted to calculate the recovery rate (recovery rate CT1), as represented on the graph below.
It is observed that recovery rates for yeast and mold vary between 15% to 40%. The best recovery is obtained with P. commune.
Gram negative bacteria show really low recovery, from 0% to 15% for P. aeruginosa. As recovery rate is reflects a cumulative effect of microorganisms survival to desiccation and microorganisms removal from the surface, it is possible that gram negative bacteria badly tolerate the drying step.
Gram positive bacilli are better recovered, with recovery rates from 15% to 40% for B. subtilis. As B. subtilis was used as spore form, it is probably more resistant to desiccation and in consequence is better recovered.
Gram positive cocci is the strain family with the best recovery rates, from 10% to 80%, particularly S. aureus is highly recovered. As gram positive bacteria can live in hash environments like human teguments, soil, water and air they are highly resistant to desiccation.
In the light of the results above, we decided to extend our study with the strain Staphylococcus aureus, which shows the best recovery, to make the comparison of different media suppliers, different type of surfaces and successive contact plate applications.
Then we selected the best recovered strain from all microorganisms families, meaning P. commune, P. aeruginosa, C. striatum, B. subtilis, M. luteus, to deeply analyze their recovery rates differences.
2.2 Surface and media suppliers influence statistical analysis on Staphylococcus aureus ATCC 6538 recovery rate
The recovery rates of Staphylococcus aureus ATCC 6538 were compared on 3 different surfaces (crystal polystyrene plastic, glass and 316L stainless steel) for 3 different contact plates suppliers. For each surface and media supplier, five replicates of 3 successive contact plates on the same inoculum were performed.
Data were analyzed for the first contact plate recovery rates, as pharma industries usually performed only one contact plate for a defined surface. The comparison of first contact plates recovery rates was performed using a two-way nested ANOVA test (with a risk alpha = 5%).
For the factor « surface », the p-value of the test is higher than the risk alpha (0,05) indicating that there is no statistically significant difference between the surfaces tested. In other words, the surface (plastic, glass or 316L stainless steel) is not influencing the first recovery rate of the bacteria S. aureus.
For the factor « supplier », the p-value of the test is lower than the risk alpha (0,05) showing that there is statistically significant difference between the different media supplier tested. Expressly, depending on the media supplier used the recovery of S. aureus may varies significantly.
It is interesting to note that, for the first contact plate recovery rate, the main part of the variability (72,5%) comes from the repeatability error. Indeed, it is well known that in a microbiological experiment the repeatability error is important because all steps can add variability (pipetting, inoculum deposit, peel off, culture conditions, reading…).
The « media supplier » factor contributes to 27,5 % of the variability of the first contact plate recovery rate. The « surface » factor does not add any additional variability.
We also analyzed the total contact plate recovery rates, corresponding to the total of bacteria recovered after 3 successive contact plates on the exact same sampling area (Cumulative recovery rate CT3). The comparison of total contact plates recovery rates was performed using a nested ANOVA test (with a risk alpha = 5%).
For the cumulative CT 3 recovery rates, all the p-value are higher than the risk alpha (0,05) indicating that there no statistically significant difference between surfaces or media supplier tested. Thus, the surface (plastic, glass, 316 L stainless steel) and the media supplier are not influencing the total (cumulative CT 3) recovery rate of the bacteria S. aureus.
Considering that the factor « media supplier » was influencing the first contact plate recovery rate, these results indicate that a second and third contact plate application is counterbalancing the first contact plate recovery rate differences.
Once again, the biggest contributor to the variability of the results is the microbiological experiment repeatability itself (82,2%). For the cumulative recovery rate CT3 the supplier factor do not convey anymore variability but the surface factor contributes to 17,8 % of the variability of the result.
2.3 Staphylococcus aureus ATCC 6538 mean and variability recovery rates, by supplier
Data analysis for recovery rates variability were performed for each supplier for all surfaces (plastic, glass and stainless steel), as the ANOVA demonstrated no influence of the surface.
The boxplot of figure 2 shows the recovery rates of S. aureus obtained with a first contact plate application, for all surfaces, for the 3 different media suppliers.
Two media suppliers (Supplier A & C) give a high first plate recovery rate, with a median around 70 %. The third supplier (Supplier B) shows lower first plate recovery rate, with a median around 60 %. For all media suppliers, the mean and median of the recovery rate are similar, suggesting that data follow a normal distribution.
Regarding the variability, the supplier B shows a higher distribution of data, with a large « box » (1st and 3rd quartile) and maximum / minimum value. This means that the first supplier B contact plate could give highly variable results, with a minimum as small as 25% recovery rate of S. aureus.
Supplier A & C show less variable results, with a smaller « box » distribution. The smaller distribution of value, including minimum and maximum values, is obtained with supplier A contact plates. Depending on the media supplier, a first contact plate application allows to obtain a recovery rate between 40 to 75 % for S. aureus inoculum.
The following boxplot shows the recovery rates of S. aureus obtained with a second contact plate application (Cumul rate CT2), for all surfaces (plastic, glass and stainless steel), for the 3 different media suppliers. Recovery rates are calculated as cumulative recoveries (total of first + second contact plate).
The median recovery rates of S. aureus on surfaces are similar between suppliers, around 85%. As for the first recovery rates, the supplier B contact plates recovery rates show higher variability with slightly smaller recovery. Supplier A & C show less variable results, as well as for the « box » distribution and for minimum / maximum value. Depending on the supplier, a second plate application allows to obtain a cumulative recovery rate between 65 to 95 % for S. aureus inoculum. The following boxplot shows the recovery rates of S. aureus obtained with a third contact plate application, for all surfaces (plastic, glass and stainless steel), for the 3 different media suppliers. Recovery rates are calculated as cumulative recoveries (total of first + second + third contact plates).
The mean recovery rates of S. aureus on surfaces are similar between suppliers, around 95%. With a successive third contact plate, differences between suppliers are smoothed out. Indeed, nested ANOVA analysis do not conclude to a statistically significant recovery rates between supplier (see chapter 2). Recovery rates variability is slightly lower with the supplier C contact plate and slightly higher with the supplier B contact plates.
Depending on the media supplier, a third plate application allows to obtain a cumulative recovery rate between 80 to 100 % for S. aureus inoculum. In other words, it is possible to recover almost all the bacteria S. aureus initially inoculated on a surface after 3 successive contact plates.
Finally, this study demonstrates that a first contact plate application allow to recover on average (depending on the supplier) 60 to 70% of the S. aureus initially inoculated. The surface sampled (plastic, glass or stainless steel) do not influence this recovery rate. A second contact and third plate application increase the recovery rate by around 20% and 6% respectively. After 3 successive contact plate applications on the same inoculum it is then possible to recovered almost all the bacteria attached to a surface (around 95%).
For the second part of the study, we evaluated the recovery rates of different strains (mold, gram positive and gram negative bacteria) on one surface using different media suppliers. As the previous results showed no influence of the surface on the microorganism recovery, we decided to use the 316 L stainless steel, because it is the most representative material in the pharma industries.
2.4 Strains mean and variability recovery rates by supplier
In this second study, we tested 5 other strains (mold, gram positive and gram negative bacteria), following the same protocol (3 successive contact plate applications on the same inoculum) on 316L stainless steel surface, as it is the more relevant for pharma industries.
Strain survival on stainless steel during the drying step could be different than S. aureus, so it is important to remember that the recovery rates obtained are a combined effect of the strain survival plus strain recovery on contact plates.
It is well known that a mold contamination on surface could be difficult to eliminate as mold spores are highly resistant to external variations(10). Penicillium contamination could occur in controlled area as it can be carried by clean room ventilation, pallets, bags, boxes, pallet jacks, scrubbers, cart wheels, carts, shoes, pens, cellphones … etc.
The following box plots show the recovery rates of P. commune spores obtained with a single (top), 2 successive (middle) and 3 successive (bottom) contact plate applications on stainless steel surface, for the 3 different media suppliers.
The first contact plate application allows recovering about 40 % of the P. commune initially inoculated. The supplier B contact plate recovery rates show a higher distribution of data, with both a « box » and maximum / minimum value enlarged. The second and third contact plate cumulative recovery rates are very similar and quite equivalent for all the suppliers. These successive contact plate applications allow obtaining about 60 % of P. commune recovery rate.
Bacillus subtilis is a gram positive bacteria capable of growth within diverse environments including the gastrointestinal tracts of animals, soil and vegetation. It is best known for its ability to respond to adverse changes in its environment by developing into a dormant endospore, highly resistant to environment variations (11).
The figure 6 shows the recovery rates of B. subtilis spores obtained with a single (top), 2 successive (middle) and 3 successive (bottom) contact plate applications on stainless steel surface, for the 3 different media suppliers.
With a first contact plate application, Bacillus subtilis spores could be recovered with a rate around 65% for the supplier A contact plate and 55% for the supplier B & C, respectively. A second contact plate application allows increasing the recovery rate to 70-85%, depending on the supplier. A third contact plate could recover almost all the B. subtilis spores initially inoculated, as the recovery rate reaches around 90%.
Corynebacterium striatum is a gram positive bacillus, commensal of mucosa and tegument(12). Thus, it can be recovered in pharma environment through operator contamination.
The following box plots show the recovery rates of C. striatum obtained with a single (top), 2 successive (middle) and 3 successive (bottom) contact plate applications on stainless steel surface, for the 3 different media suppliers. Attention must be payed to the axe’s scaling (maximum of the scale is 40 %).
A first contact plate on stainless steel allows recovering only 10 % of C.striatum inoculated, for all supplier tested. Depending on the supplier, a second and third contact plate applications increase the recovery rate to 15-20%. The maximum mean recovery rate for this bacteria is around 20%, even with 3 contact plate applications. As mentioned in the introduction, it should be remember that the recovery rate reflects both drying resistance of the strain (which is spotted and then dried 30 minutes at 37°C before contact plate application) and the contact plate removal of strain.
Micrococcus luteus is a gram positive coccus, commensal of mammals skin. Thus, this strain is frequently recovered in pharma environment through operator contamination(13).
The following box plots show the recovery rates of M. luteus obtained with a single (top), 2 successive (middle) and 3 successive (bottom) contact plate applications on stainless steel surface, for the 3 different media suppliers.
Micrococcus luteus recovery rate with a first contact plate application is between 45 % to 50 %, depending on the supplier. A second and third contact plate application increases this recovery to 60 % and 70 % respectively.
Pseudomonas aeruginosa is a gram negative bacteria which has the ability to survive within diverse environments including soil, marshes, water and skin flora. P. aeruginosa may also be found in a biofilm on surfaces, highly resistant to cleaning and disinfection procedures(14). The following box plots show the number of CFU of P. aeruginosa obtained with a single (top), 2 successive (middle) and 3 successive (bottom) contact plate applications on stainless steel surface, for the 3 different media suppliers. Attention must be payed to the axe’s scaling in CFU (and not in recovery rate), because P. aeruginosa recovery was really low.
Recovery of P. aeruginosa on stainless steel with contact plate is really low. The first contact plate application allows recovering barely 1 to 3 CFU for a 30 CFU inoculum (around 10%). A second and third contact plate applications increase the number of CFU to 2 to 5 CFU. As for C. striatum bacteria, it should be kept in mind that the recovery rate reflects both drying resistance of the strain (which is spotted and then dried 30 minutes at 37°C before contact plate application) and the strain contact plate removal.
Surface control with swabbing methods and culture media contact application is a large part of environmental monitoring strategy of pharmaceutical industries. However, surface recovery of microorganisms is still an open debate, as laboratory or in house studies are rare and performed under different conditions(4,9). In this study, we determined surface recovery of microorganisms with a reproducible and standardized protocol. The results we obtained showed no influence of surface type sampled: the recovery rates are equivalent between crystal polystyrene plastic, glass and 316L stainless steel surfaces. This result is interesting because different of what we can anticipate: different surfaces as different as plastic (passive surface), glass (charged surface) and stainless steel (charged surface) may have different microorganisms attachment(15–17). In our study it was not the case, possibly because the contact period between the microorganism and the surface is short(18). However, we showed that the variability brought by the surface is increasing as the contact plate application increases. It is then possible that the surface differences (inert / active, rough / smooth) add variability in more stringent conditions, i.e second and third contact plate application (low number of microorganisms, microorganisms hidden in less accessible area, strongly attached microorganisms ..).
In addition, media suppliers of contact plates have been found to influence S. aureus first contact recovery rates. It is possible that composition of the culture media, presence of surfactant, surface water contents, plastic petri dish design differ between suppliers and influence this first microorganisms recovery(3).
A first contact plate on dried inoculum of S. aureus shows a recovery rates around 60-70%, depending on the supplier. This data are consistent with other studies performed on Staphylococcus strain(3,4). This indicates that a single contact plate application on a surface do not allowed to remove all the microorganisms present. It’s only after 3 successive contact plate applications on the same area that it was possible to recover around 90% of the S. aureus inoculated on the surface. It is understandable that 3 successive contact plate applications are highly constraining for environment monitoring practices, however the proportion of microorganisms removed during surface sampling must be remembered.
Is surface monitoring the exact reflect of the microbial contamination present on the surface ? It is not certain, especially because our results show that the different microorganisms are not recovered the same. Also because they are probably less resistant to desiccation, the maximum recovery of P. aeruginosa or C. striatum is only 20 %. For more desiccation-resistant microorganisms or spores, after 3 successive contact plate on the same dried inoculum, it’s possible to recover at least two-thirds of the strains present on the surface. A single contact plate application allows to recover between 40 to 65 % of the inoculum, depending on the microorganism.
Finally this experimental design provides clear set of data on 6 different microorganisms surface sampling recovery, on 3 different surfaces, using 3 different contact plate suppliers, testing until 3 successive contact plate applications on the same inoculum. We showed that various parameters could influence the surface sampling results, pointing out that in-house validation studies, using the specific contaminant flora of the controlled zone must be performed.
Jeanne GROSSELIN – BIOMERIEUX INDUSTRY
Laurent LEBLANC – BIOMERIEUX INDUSTRY
(1) Ismaïl, R. et al. Methods for recovering microorganisms from solid surfaces used in the food industry: a review of the literature. Int. J. Environ. Res. Public. Health 10, 6169–6183 (2013).
(2) Edmonds, J. M. Efficient methods for large-area surface sampling of sites contaminated with pathogenic microorganisms and other hazardous agents: current state, needs, and perspectives. Appl. Microbiol. Biotechnol. 84, 811–816 (2009).
(3) Deckers, S. M., Sindic, M., Anceau, C., Brostaux, Y. & Detry, J. G. Possible influence of surfactants and proteins on the efficiency of contact agar microbiological surface sampling. J. Food Prot. 73, 2116–2122 (2010).
(4) Obee, P., Griffith, C. J., Cooper, R. A. & Bennion, N. E. An evaluation of different methods for the recovery of meticillin-resistant Staphylococcus aureus from environmental surfaces. J. Hosp. Infect. 65, 35–41 (2007).
(5) Dalmaso, G., Bini, M., Paroni, R. & Ferrari, M. Qualification of high-recovery, flocked swabs as compared to tradi-tional rayon swabs for microbiological environmental monitoring of surfaces. PDA J. Pharm. Sci. Technol. 62, 191–199 (2008).
(6) Valentine, N. B. et al. Evaluation of sampling tools for environmental sampling of bacterial endospores from porous and nonporous surfaces. J. Appl. Microbiol. 105, 1107–1113 (2008).
(7) Weir, M. H., Shibata, T., Masago, Y., Cologgi, D. L. & Rose, J. B. Effect of Surface Sampling and Recovery of Viruses and Non-Spore-Forming Bacteria on a Quantitative Microbial Risk Assessment Model for Fomites. Environ. Sci. Technol. 50, 5945–5952 (2016).
(8) Goverde, M., Willrodt, J. & Staerk, A. Evaluation of the Recovery Rate of Different Swabs for Microbial Environmental Monitoring. PDA J. Pharm. Sci. Technol. 71, 33–42 (2017).
(9) Moore, G. & Griffith, C. Problems associated with traditional hygiene swabbing: the need for in-house standardization. J. Appl. Microbiol. 103, 1090–1103 (2007).
(10) Wyatt, T. T., Wösten, H. A. B. & Dijksterhuis, J. Fungal spores for dispersion in space and time. Adv. Appl. Microbiol. 85, 43–91 (2013).
(11) Driks, A. The bacillus spore coat. Phytopathology 94, 1249–1251 (2004).
(12) Martínez-Martínez, L., Suárez, A. I., Rodríguez-Baño, J., Bernard, K. & Muniáin, M. A. Clinical significance of Co-rynebacterium striatum isolated from human samples. Clin. Microbiol. Infect. Off. Publ. Eur. Soc. Clin. Microbiol. Infect. Dis. 3, 634–639 (1997).
(13) Kooken, J. M., Fox, K. F. & Fox, A. Characterization of Micrococcus strains isolated from indoor air. Mol. Cell. Probes 26, 1–5 (2012).
(14) Masák, J., Čejková, A., Schreiberová, O. & Rezanka, T. Pseudomonas biofilms: possibilities of their control. FEMS Microbiol. Ecol. 89, 1–14 (2014).
(15) Jindal, S. & Anand, S. Comparison of adhesion characteristics of common dairy sporeformers and their spores on unmodified and modified stainless steel contact surfaces. J. Dairy Sci. (2018). doi:10.3168/jds.2017-14179
(16) Alam, F. & Balani, K. Adhesion force of staphylococcus aureus on various biomaterial surfaces. J. Mech. Behav. Biomed. Mater. 65, 872–880 (2017).
(17) Simões, L. C., Simões, M., Oliveira, R. & Vieira, M. J. Potential of the adhesion of bacteria isolated from drinking water to materials. J. Basic Microbiol. 47, 174–183 (2007).
(18) Crouzet, M. et al. Pseudomonas aeruginosa cells attached to a surface display a typical proteome early as 20 minutes of incubation. PloS One 12, e0180341 (2017).
ATCC : American Type Culture Collection
CFU : Colony Forming Unit
CT : Contact plate
TSA : Trypcase Soy Agar
ANOVA : Analysis of Variance
BD : Becton Dickinson
Ref. : Reference
Cumul : Cumulative