The Problem of Biofilms and Pharmaceutical Water Systems

• Bio Products Laboratory Ltd

Introduction

Biofilms are problematic to pharmaceutical water systems. If a biofilm develops then an out-of-control situation is likely to emerge. Such a situation can often only be detected from point-of-use samples and since several excursions are required to alert of the probability of a biofilm, the biofilm will most likely have become significantly established at this point in time. This article assesses biofilms and pharmaceutical water systems, considering prevention and control measures; looking at what can be meaningfully gleaned from monitoring; and measures required to address biofilms.

Biofilms

Biofilms are natural communities of bacteria and they are very common in nature. Biofilms are bacterial populations where the bacteria are adherent to each other and/or on surface interfaces. Here the biofilm differs to microorganisms found within the planktonic state. Adherence is enhanced by many species within the community secreting a polysaccharide coating which is ‘slime like’ and very adhesive. The function of the coating is to encourage the attachment of other bacteria; to trap nutrients; and to provide a degree of protection. As discussed later, this protective mechanism makes the biofilm difficult to destroy. The organisms within the biofilm undergo physiological changes and undertake different forms of sensing in relation to the environment and to each other (as well as different forms of communication, bacteria can also share genetic information within the community). Over time, biofilms can amount to a considerable mass.¹

Biofilm formation undergoes several steps. Initially the attracted bacteria are held in a state of reversible attachment. This, over time (minutes rather than hours) becomes an irreversible attachment. Within a few days micro-colonies form, leading to an enlarged mass and the establishment of a mature biofilm community.² In general, the morphological group Gram-negative bacteria form biofilms more readily; one reason is due to appendages on the bacterial cell that allow such bacteria to attach to surfaces more easily.³ A second reason that can led to bacterial adherence in a water system is surface charge.⁴

Most microbial activity in an ecosystem is with microorganisms adhered to a surface, within a biofilm community; including most aquatic systems. Despite the typicality, with well-maintained pharmaceutical water systems, biofilms are very atypical. There are reasons, however, why bacteria have a propensity to form biofilms and why biofilms can sometimes develop.

Formation of Biofilms in Pharmaceutical Water Systems

The requirements for biofilm formation reads like a simple recipe: a given concentration of planktonic bacteria, plus water, plus a surface, plus available nutrients (including a carbon source). The nutrient levels need only be relatively low.⁵ On a standard agar plate the nutrient levels are typically 2,000 microgram per liter; in a purified water system minimal nutrient levels, of say 0.5 microgram per liter, are still sufficient to allow a community to flourish. Moreover, the fact that pharmaceutical water is a low nutrient environment is most likely a driver for the establishment of a biofilm as part of the microbial survival strategy. This is because nutrients tend to gravitate to surfaces and planktonic bacteria find it more difficult to absorb nutrients compared with bacteria in the sessile state. The biofilm community also provides a degree of protection against factors like osmotic shock.

In a well maintained pharmaceutical water system, how might biofilms form? There will be differences between new (and poorly conceived) water systems and established (but compromised) water systems. Biofilm formation in new water systems is often a product of poor design. This may relate to the type of material used for the pipework; the finish of the pipework; the diameter of the pipe; the velocity of the circulating water or the presence of dead-legs (bends in the pipe where the water velocity slows).⁶

In established water systems, biofilm formation is associated with shutdowns or periods of very low water use, which affects the velocity of the water flow; or following repairs and maintenance, such as the need to cut into a water system to replace pipework or to change a valve.

Monitoring of Water Systems

The standard approach for assessing microbial levels in water systems is to pass a portion of water through a membrane filter and to capture any colonies present in the filter membrane. Organisms are then cultured by the addition of liquid culture media or by placing the filter onto an agar plate. The samples are then assessed against pharmacopeia limits (which are advisory in the USP but mandatory in the European Pharmacopoeia). Results generated from water systems rarely give meaningful information when individual values are assessed, and differentiating between similar values is rarely meaningful (such as between 5 and 10 CFU/100mL). Moreover, 5 CFU itself may relate to 5 individual bacteria or 5 clumps of bacteria, with each clump consisting of 50 cells. Beyond this, there is the problem of extrapolation: to what degree is a 100mL sample representative of a water system that, at the time of sampling, contained 10,000 liters? It is more appropriate to assess water systems for trends and it is fluctuations and gradual rises with trends that are most likely to inform about biofilms.

The time taken to detect a biofilm, from its initial formation, is variable. This depends on the type of water system and its design and the size of the biofilm community. The time taken may vary from several days to several months. In addition, and in support of trend analysis, the rate the organisms from the biofilm are detected will vary. Some water samples taken over successive days, for example, will detect counts of different magnitude whereas others will not. The same pattern may be seen with bacterial endotoxin test results. Outside of direct sampling of use outlets, there is no established way of detecting for biofilms in pharmaceutical water systems.

The rate of detection is a factor of the dispersal of bacteria from the community. Bacterial cell clusters are shed from the biofilm at different rates, with the shedding arising from shear forces and pressure variations (pressure ‘shocks’, for example, can occur each time a user valve is opened). The dispersed cells can either move to a new area of pipework to colonize a surface or be eluted out through a user outlet.

Preventing Biofilm Formation

Poorly designed storage and distribution systems create opportunities for re-colonization and ultimately product contamination. However, biofilm formation can be prevented from the outset with good control design principles. Pipes are more likely than any other part of the water system to develop contamination, if they are poorly designed (or improperly maintained). ⁷

With pipes and other parts of the water distribution system, some important design elements include:

1. Ensuring smooth internal surfaces in tanks and in pipework. This is because microorganisms adhere less well to smooth surfaces compared with roughened surfaces. This additionally makes corrosion resistance and regular steps to avoid rouging as important design features (such as electropolishing of stainless steel).
2. Having continuous movement of the water in tanks and rapid flow in pipe-work. Flow rate is necessary for minimizing the opportunities for microorganisms to adhere to surfaces (and hence form biofilms). Where shear forces are sufficient microorganisms adhere poorly to surfaces. In terms of general guidance, typical flow velocities in the range of 1-2 m/s tend to be suitable.⁸
3. As well as continuous movement, areas where water flow can slowdown or stagnant pockets of water can develop need to be avoided. This includes so-termed ‘dead legs’, which are pipe branches that are too long as to cause turbulence to occur with the water flow. The length of branch pipes should always be minimized.
4. Valves must also be properly designed so they do not allow water to remain. This can occur with some user points. This is over come with the use of zero dead-leg valves, together with sloping drainage.
5. Where leakage occurs this must be quickly addressed, since this can allow bacteria to ingress from the external environment.
6. Use of filters on air vents for with water storage tanks; this also helps to minimize the ingress of microorganisms.
7. The use of coated surfaces on pipes and in tanks, where appropriate (as not to pose a risk of leaching toxic substances) can help to address bio-fouling.

It is also important that when maintenance work is performed on the system, the part of the system being worked on should be isolated and, once complete, a full system sanitization should be performed.

Controlling Biofilms

Where biofilm communities are present, these can be controlled by maintaining a water system at an elevated temperature (the temperature required will be somewhere between 65 and 80°C). This option is only available if the system has been designed as a hot water system with user outlets that cool the water down through heat exchangers. Typically this is more common with Water-for-Injection than with purified water systems. Alternatively, ozone can be used to maintain an environment above 10 parts per billion of ozone, with the ozone removed at critical junctures for health and safety reasons. Outside of these conditions, the biofilm may flourish or, if design is inadequate, the alternate conditions may allow biofilms to form.

Eliminating Biofilms

Even when good design and maintenance approaches are used, biofilm formation can still occur. For such occurrences attempts are required to eliminate the biofilm community. There are four methods that can be considered for water systems in order to remove microbial contamination. These are heat, chemical additives, filtration and ultraviolet light.⁹ With heat, either maintaining the circulating loop at 65°C to 80°C for hot water systems may address the biofilm or the system will requiring ‘steaming’ by elevating the temperature within the region of 121°C and ensuring the temperature is held for a sufficient contact time (typically 10-15 minutes).

With chemical treatment there are a range of potential chemicals with the choice being partly dependent upon material compatibility. Chemicals for consideration include: ozone, chlorine, chlorine dioxide, hydrogen peroxide, peracetic acid and sodium hydroxide. The concentration of the chemical will vary depending upon the agent; with chlorine, chlorination can be achieved with minimum levels of 0.2mg/liter of free chlorine.

The third option, membrane filtration can be attempted using a 0.22µm porosity-filter. However, this is generally only effective with early stage biofilms since filtration does not directly remove the source. With ultraviolet radiation this is effective at a wavelength of 254nm for a re-circulating system where water flows over a multiple lamp system. If these treatments do not prove effective, the final recourse is to disassemble pipework and replace with new pieces.¹⁰

Summary

This article has outlined the problem of microbial biofilm developing in pharmaceutical water systems. The article has discussed where the organisms that can potential form biofilms may originate from and the conditions, either relating to poor design or maintenance that can lead to biofilm formation. For the microbiologist, careful attention needs to be paid to trend data in order to ensure that biofilms are detected in a timely manner (noting that the data pattern itself may appear variable). Where biofilms cannot be controlled, there are various treatment outcomes, as discussed. It is often the case that biofilms are difficult to eliminate and a variety of approaches may be required. Where control and remediation are difficult, good design is the overwhelming factor that will keep water systems in a satisfactory state.

References

1. Bhattacharjee, S., Elimelech, M., Borkovec, M. (2003) DLVO interaction between colloidal particles: Beyond Derjaguins approximation, Croatica Chimca Acta 71, 883-903
2. Wingender, J., Neu, T. R. and H-C. Flemming, H. C. (1999) What are bacterial extracellular polymeric substances? In: Wingender, J., Neu, T. R.., Flemming, H-C. (Eds.) Microbial Extracellular Polymeric Substances SpringerVerlag Berlin Heidelberg, pp. 1-19.
3. Sandle, T. (2015) Characterizing the Microbiota of a Pharmaceutical Water System-A Metadata Study. SOJ Microbiol Infect Dis 3(2), pp1-8
4. Jiminez, L. Microorganisms in the Environment and their relevance to Pharmaceutical Processes. In Jiminez, L. (Ed.) Microbial Contamination Control in the Pharmaceutical Industry. New York: Taylor & Francis; 2004. p. 8-9.
5. An, H. Y., and R.J. Friedman, R. J. (1998) Concise review of mechanisms of bacterial adhesion to biomaterial surfaces, Journal of Biomedical Materials Research: Applied Biomaterials 43, 338-48
6. Sandle, T. (2013). Bacterial Adhesion: an Introduction, Journal of Validation Technology, 19 (2): 1- 10
7. Lehtola, M.J., Miettinen, I. T., Keinanen, M.A. et al (2004) Microbiology, chemistry and biofilm development in a pilot drinking water distribution system with copper and plastic pipes, Water Research, 38 (17): 3769-3779
8. Sandle, T. (2013) Avoiding Contamination of Water Systems, The Clinical Services Journal, 12 (9) pp33-36
9. Flemming, H-C. (2002) Biofouling in water systems – cases, causes and countermeasures, Applied Microbiology and Biotechnology, 59 (6): 629-640
10. Sandle, T. (2017) Design and control of pharmaceutical water systems to minimize microbial contamination, Pharmaceutical Engineering, 37 (4): 44-48