OR WAIT null SECS
The author reviews test methods for microbiological cleaning processes and suggests ways to improve microbial bioburden method suitability studies.
Surface microbial bioburden monitoring methods are described in Standard Methods for the Examination of Dairy Products (1). A review of the literature shows that few studies have been conducted to examine actual swab-recovery efficiencies and limits of detection (2–6). Niskanen and Pohja indicated that the contact-plate method is suitable for flat, firm surfaces, (considering both recovery and repeatability), whereas swabbing is better for flexible and uneven surfaces and for heavily contaminated surfaces (7, 8).
Previous studies reported a poor correlation with the amount of microbial contamination on surfaces and the recovery obtained. For example, a collaborative study comparing surface monitoring methods showed that artificially contaminated stainless steel at a theoretical level of 1.4 colony forming units (cfu) per cm2 (about 35 cfu per sample) gave recoveries of 25 to 30% when bacterial spores were employed (9). As reported by Kang, Eifert, and Williams, recovery of Listeria monocytogenes from stainless-steel inoculated test areas, or coupons, using a sonicating brush head (contact or noncontact) yielded a recovery level of about 60% compared with swab and direct agar contact methods (about 20%) (10). Many factors may contribute to this poor correlation, including differences in materials used (e.g., cotton, polyester, rayon, calcium alginate), the organisms targeted for culture, variations in surface, and differences in the personnel collecting and processing samples (11). Additional sources of error are the potential for nonhomogenous surface deposition of test microorganisms resulting in unequal or incomplete removal of microorganisms from the test surface (4). Based on these studies it is widely accepted that positive swab samples are indicative of high surface concentration of microbes, whereas negative swab samples do not assure that microorganisms are absent from the surface sampled (4).
As stated by Moore and Griffith, in the use of swabs for recovery of microorganisms from surfaces, there is a lack of standardization of both the swabbing pattern and the pressure applied to the swab during sampling. This means that technician-to-technician variation in the sampling procedure may potentially play a significant role in the recovery and enumeration of the sampled surface. This can lead to biased results for the initial cleanliness of the surface or the effectiveness of the cleaning procedure used. Although many surface-sampling techniques are available, their effectiveness as surface cleanliness indicators may vary. The bioburden recovery methods using swabs can be also influenced by the experimental design. Bioburden recovery results using swabs may not be as sensitive as other sampling techniques (i.e., replicate organism detection and counting [RODAC]). In addition, results may be invalidated due to the presence of excessive, uncontrolled variability and data scattering due to technician-to-technician variability (12, 13).
In microbiological studies evaluating cleaning and/or sanitization in the pharmaceutical industry, the removal and enumeration of viable microorganisms from material surfaces is a critical point for environmental control. The data obtained pertaining to survival and enumeration of microorganisms would affect validation (11). Variables affecting the accuracy of the detection and enumeration using the swabbing technique initially include the ability of the swab to remove the microflora from the surface as well as its effectiveness to release removed microorganisms from the swab and their subsequent recovery and cultivation (12, 13). The proportion of attached microflora on surfaces that are trapped or tenaciously bound to the interwoven fibers of a swab head are unknown. Sampling techniques that preserve the underlying surface as well as the viability of the detached microflora will detach only a portion of the total population. Adherent bacteria on surfaces become increasingly difficult to remove by use of swabs, especially if they become associated with a biofilm (11, 12–14). Studies conducted under controlled conditions have demonstrated that recovery is low. Kusumaningrum, et al. reported that in evaluating the survival and recovery of Bacillus cereus, Salmonella enteriditis, Campylobacter jejuni, and Staphylococcus aureus on stainless-steel surfaces, the direct contact method using solidified agars recovered 18% of Bacillus cereus, 23% of Salmonella enteriditis, 7% of Campylobacter jejuni, and 46% of Staphylococcus aureus from the initial concentration applied to the surface (15). A validation and comparative study on recovery of microorganisms using swabs, Hygicult TPC dipslide, and contact agar plates yielded similar results and did not differ in precision, with recoveries ranging from 16 to 30% of the microbial load applied to the surface (9).
Inadequate cleaning procedures may result in a number of contaminants present in future batches manufactured on the contaminated equipment such as:
Process validation is a series of interrelated functions and activities using a variety of specified actions that are designed to produce a predetermined result. This activity examines a process under normal operating conditions to prove that the process is in control. Cleaning in process validation is the documented study that demonstrates the reproducibility and consistency of the cleaning process to reduce undesired material from contact surfaces to a safe level. These studies apply appropriate analytical recovery methods designed to determine the presence of selected materials or entities (16, 17).
Cleaning validation should include studies using appropriate bioburden recovery methods where the possibility of microbial contamination of subsequent product is deemed possible and presents a product-quality risk. Validation studies are performed using coupons of the representative surfaces inoculated with the test microorganisms. The test microorganisms, which are usually known laboratory-adapted strains, are spread onto a space that is approximately 25 cm2 and allowed to air dry. After air drying, the test microorganisms are recovered by either swabs or contact plates. The test samples, along with positive and negative controls, are treated and/or incubated. Results are analyzed based on the percent of test microorganisms that grow after recovery compared with an inoculation control (1).
The validation of surface-recovery methods (i.e., chemical and microbiological) is a pre-requisite for residual determination of cleaning effectiveness in process validation studies. These methods should be challenged in the laboratory using pilot-scale controlled conditions in order to evaluate the suitability for their intended use. For this purpose, validation specialists select representative surfaces identified within the production area and potentially in contact with ingredients, product intermediates, and bulk products. Surfaces selected for method validation commonly include stainless steel 316L, glass, plastic (such as polyvinyl-chloride and polyethylene), and some metal alloys. However, surface selections for challenging studies are not justified based on what really matters: demonstrating the effectiveness of the monitoring method (16, 17).
There are many types of surfaces in the pharmaceutical production areas and current-good-manufacturing-practice (CGMP) equipment, all with distinct physico-chemical properties. Most of these surfaces are well defined. When microorganisms are released into the manufacturing area, they will be deposited onto these surfaces as either aerosol particles or as liquid droplets. The type of surface greatly influences their ability to survive and their possibility to contaminate other materials.
Surfaces can be split into several groups, for example, porous and nonporous, inert or active, rough or smooth, and hydrophobic or hydrophilic. Glass and stainless steel are examples of nonporous inert surfaces, and galvanized steel, brass, and copper are example of nonporous active surfaces (18, 19). Stainless steel has been extensively studied, in part because it is the principal material of construction of good manufacturing practice (GMP) equipment (29). Microscopically, stainless steel may show grooves and crevices that can trap bacteria but glass does not (20). Some bacteria have been found to be able to adhere to stainless-steel surfaces after short contact times if the conditions are appropriate (i.e., adequate temperature and humidity) (21).
The porosity of a surface is a major factor affecting bacteria adherence. Highly porous surfaces facilitate adherence of bacteria. However, the adherence of bacteria depends on the number of cells—the higher the number of cells, the higher the probability those cells remain attached on surface after rinse (22). It was demonstrated that gram-negative bacteria adhesion could be decreased with the addition of silicone on porous material such as plastics, Teflon, and Dacron (19). Additionally, it was reported that rubber and plastic coupons were significantly more accessible to bacteria than glass coupons, as revealed by the high population of bacteria recovered from their surfaces (23). Porous materials such as plastics, Teflon, Dacron, and their combination are used less often as materials of construction in GMP equipment. Rijnaarts and colleagues reported that bacteria deposition on Teflon is faster than on glass.
Silicone rubber has found widespread use in medical, aerospace, electrical, construction, and industrial applications (24). Silicone rubbers are synthetic polymers with a giant backbone of alternating silicon and oxygen atoms. The nonporous nature of silicone's surface does not allow the adhesion of bacteria (25, 26). Nevertheless, studies of bacterial adhesion with laboratory strains of bacteria (i.e., type culture collection strains), many of which had been transferred thousands of times and lost their ability to adhere, first indicated that very smooth surfaces might escape bacterial colonization. Subsequent studies with "wild" and fully adherent bacterial strains showed that smooth surfaces are colonized as easily as rough surfaces and that the physical characteristics of a surface influence bacterial adhesion to only a minor extent (27). This point is important to remember when selecting test microorganisms for suitability testing.
Porosity may prevent water evaporation. The lethal effect of desiccation was found to be the most important death mechanism in bacteria (26). Similar studies performed on a Teflon surface using Escherichia coli, Acinetobacter sp., Pseudomonas oleovorans, and Staphylococcus aureus demonstrated that all four species survived well during the droplet evaporation process, but died mostly at the time when droplets were dried out at 40 to 45 min (26). However, some non-spore-forming bacteria might be able to withstand dry conditions on surfaces for an extended period of time (15). Kusumaningrum and colleagues concluded that the survival of microbes under these conditions is dependent on the contamination levels and type of pathogen. High-density polyethylene (HDPE) is prepared from ethylene by a catalytic process (15). The absence of branching results in a more closely packed structure with a higher density and somewhat higher chemical resistance than low-density polyethylene (LDPE). HDPE is chemically inert and the surface is more porous than LDPE. HDPE is used in products and packaging detergent bottles, garbage containers and water pipes (28).
Fabrics are porous surfaces (e.g., cotton, polyester, polyethylene, polyurethane and their combinations, etc.) that demonstrate survival of gram-negative and gram-positive microorganisms even longer than plastics (29, 30). It has also been observed that gram-positive bacteria survive slightly longer than gram-negatives. Because of their porous nature, it is recommended to rinse fabrics and other porous surfaces in order to detach microbes from them. Swab and plate contact methods are not suitable for fabrics.
It is well known that charged molecules in a solution are able to kill bacteria (31–33). However, it has been realized more recently that charges attached to surfaces can kill bacteria upon contact (34). Certain surfaces such as brass, copper and galvanized steel can be toxic to bacteria because the presence of water and air allows the release of metal ions from metal surface (19), and these metal ions exert an antimicrobial effect by interfering with biological pathways and enzymes (18, 33). Copper releases Cu2+ ions, galvanized steel releases Zn2+ ions, and brass releases both Cu2+ and Zn2+ ions. These metal ions are, in fact, essential micronutrients of bacterial cells at very low concentrations. Above these levels they may exert a toxic effect on aerobic bacteria.
Different types of plastics have different surface properties, which influence their effect on bacteria. Polyvinyl chloride (PVC) and polypropylene (PP) are two similar plastics, but have different properties. PP is more stable and less reactive than PVC. PVC surfaces show high mortality rates for bacteria while PP surfaces show no significant levels of mortality. Studies with Enterococcus faecalis aerosol on PVC and PP demonstrated that PVC had a significant effect on the survival of bacteria due to oxidation reactions with the walls of gram-negative bacteria (30).
Wildführ and Seidel reported that bacteria die faster on stainless-steel and glass surfaces than on plastic surfaces. These comparative studies tested Pseudomonas aeruginosa, Staphylococcus aureus, and Candida albicans demonstrated that the survival rate on plastic was almost double (50% more) than on stainless-steel or glass coupons (35). In this study test microorganisms' suspensions were transferred onto stainless-steel, glass, and plastic coupons and then dried. After 90 minutes, it was evident that only a very small quantity of bacteria was present on the stainless-steel and glass surfaces, but the quantity of viable bacteria on the plastic was still high up to 120 min later.
Tiller et al. reported that plastic coupons (i.e. polypropylene and Polystyrene) kept bacteria more viable than aluminum, steel, and glass (34). In this experiment, suspensions (106 cells per mL) of Staphylococcus aureus in distilled water were sprayed over various surface materials, air dried for 2 min, and were incubated in a 0.7% agar bacterial growth medium overnight, after which the colonies counted. Bacterial adherence in the presence of oral liquid pharmaceuticals on different coupons showed that rubber and plastic coupons were significantly more accessible to the bacteria than glass coupons (36, 37).
The literature cited herein supports the argument that porous nonactive carbon polymer surfaces such as plastic and rubber harbor viable bacteria for longer periods than steel, glass, metals, and metal alloys surfaces. Table I summarizes the interaction of the different surfaces with microbes.
As stated previously, validation studies are usually performed using representative surfaces identified on the production areas but no other justifications are currently included. Because of this, results are subject to variability and low reproducibility at high costs and efforts. Even though useful information arises during these processes, method suitability must be seen in terms of the capacity of the method to recover viable microbes. For method suitability purposes, the author highly recommends selecting surfaces that may potentially harbor and allow the survival of microbes for the longest time possible. Surfaces that minimize survival may jeopardize method-validation results. Surface(s) must be allowed to verify the method effectiveness for its intended use without the possibility of harming the microbes housed there.
One porous and one nonporous surface may be chosen to test methods and may support the method suitability for sampling all other drug-contact surface materials. A porous inert surface such as plastic and a nonporous inert surface such as as polished stainless steel are represented in most food, medical device, and pharmaceutical production areas. The author highly suggests performing test method suitability studies using wild-type isolates from production surfaces instead of laboratory-adapted strains of bacteria. Wild-type strains are a better representation of the organisms encountered on production areas than those strains that lack wild characteristics. The method will be ready for use in monitoring studies once it is optimized and its accuracy is established.
Angel L. Salaman-Byron, PhD, is a microbiology consultant at Pharma-Bio Serv, tel. 787.384.7216, firstname.lastname@example.org.
1. R.L. Dyer et al., "Microbiological Tests for Equipment, Containers, Water and Air" in Standard Methods for the Examination of Dairy Products, R.T. Marshall, Ed. 17th ed. (American Public Health Association, Washington, USA, 17th ed., 2004), pp. 328.
2. R. Angelotti et al., Health. Lab. Sci. 1, 289–96 (1964).
3. J.M. Barnes, Proc. Soc. Appl. Bacteriol., 15, 34–40 (1952).
4. G.S. Brown et al., J. of Appl. Microbiol. 103 (4), 1074–1080 (2007).
5. M.S. Favero et al., J. Appl. Bacteriol., 31 (3), 336–43 (1968).
6. L.B. Rose et al., Emerg. Infect. Diseas., 10 (6),1023–1029 (2004).
7. A. Niskanen and M.S. Pohja, J. of Appl. Bacteriol. 42, 53–63 (1977).
8. V. Marshall, S. Poulson-Cook, and J. Moldenhauer, PDA J. Pharm. Sci Technol. 52 (4), 165–169 (1998).
9. S.A. Salo et al., J. of AOAC Intl. 83, 1357–1365 (2000).
10. D. Kang et al., J. of AOAC Intl. 90 (3), 810–816 (2007).
11. J.F. Frank, "Microbial attachment to food and food contact surfaces" in "Advances in Food and Nutrition Research," Taylor, S.L., Ed. (CA. Academic Press, San Diego, Vol. 43, 2001) pp. 320–370.
12. G. Moore and C. Griffith, Dairy, Food, and Environmental Sanitation 22 (6), 410–421 (2002a).
13. G. Moore and C. Griffith, Food Microbiol. 19, 65–73 (2002b).
14. J.W. Costerton et al., Annu. Rev. Microbiol. 49, 711–745 (1995).
15. H.D. Kusumaningrum et al., Intl. J. of Food Microbiol. 85 (3), 227–236 (2003).
16. PDA Technical Report No. 29, "Points to Consider for Cleaning Validation," PDA J. Pharm.Sci. and Tech., 52 (6), 1–23 (1998).
17. FDA, Guide to Inspections of Validation of Cleaning Processes, (Rockville, MD, July 1993).
18. P. Tandon, S. Chhibber, and R.H. Reed, J. of Gen. and Mol. Microbiol., 88 (1), 35–48 (2005).
19. A.P. Williams et al., J. of Appl. Microbiol. 98 (5), 1075–1083 (2005).
20. A.A. Mafu et al., J. of Food Protection, 53 (9), 742–746 (1990).
21. L. B. Rose et al., Emerg. Infect. Diseas., 10 (6), 1023–1029 (2004).
22. D.R. Absolom et al., Appl. and Env. Microbiol., 46 (1), 90–97 (1983).
23. L.O. Egwari, and M.A. Taiwo, West Indian Med. J. 53 (3), 164–169 (2004).
24. H.H.M. Rijnaarts et al., Environ. Sci. Technol. 30 (10), 2877–2883 (1996).
25. W. Lynch, Handbook of Silicone Rubber Fabrication, (Van Nostrand Reinhold, New York, 1978).
26. X. Xie et al., Appl. Microbiol. and Biotech. 73 (3), (2006).
27. G.M. Chudzik, J. Validation Technol. 5 (1), 77–81 (1998).
28. J. Kahovec, R.B. Fox, and K. Hatada, Pure and Applied Chem. 74 (10), 1921–1956 (2002).
29. A.N. Neely and M.P. Maley, J. of Clinic Microbiol., 38 (2), 724–726 (2000).
30. A.N. Neely and M.P. Maley, Am. J. of Infect. Cont. 29 (2), 131–132 (2001).
31. Y. Endo, T. Tani, and M. Kodama, Appl. Environ. Microbiol. 53 (9), 2050–2055 (1987).
32. S. Fidai, S.W. Farer, and R.E. Hancock, Methods Mol. Biol. 78, 187–204 (1997).
33. S.J. Kim, Marine Ecology – Progress Series 26, 203–206 (1985).
34. J.C. Tiller et al., Proc. Natl. Acad. Sci. USA 98 (11), 5981–5985 (2001).
35. G. Reiser, Using Technology to Break the Contamination Chain, Business Briefing: Hospital Engineering and Facilities Management, Issue No. 2 (2005).
36. A. Winn, Medical Plastics and Biomaterials 3 (2), 16–19 (1996).
37. E. Robine, D. Derangere, and D. Robin, Inter. J. of Food Microbiol. 55 (1–3), 229–234 (2000).