EN 17141: Understanding the risks of microbiological contamination in pharmaceutical cleanrooms

Published: 21-Apr-2021

EN 17141 outlines the importance of understanding the risks of microbial contamination. Being able to accurately identify microbes isolated from the controlled environment is vital to assess risk to pharmaceutical products, and ultimately patients, and ensure that these organisms can be eliminated or controlled

The new European standard EN 17141:2020 was issued on 12 August 2020 and supersedes EN ISO 14698-1:2003 and EN ISO 14698-2:2003. It is valid in Europe with the existing standard (14698) applicable elsewhere.

The standard covers clean controlled environments and is relevant to cleanrooms, clean zones, controlled zones, clean areas, and clean spaces. Its aim is to provide updated guidance and further information on best practice for establishing and demonstrating control of airborne and surface microbiological contamination in clean controlled environments. The third route of transfer, liquid, is not included as part of the scope of this document. It intends to help increase the effectiveness of risk management associated with microbial contamination, improve efficiency, and provide users with the methodology and understanding to derive a system of microbiological control. In regulated industries this new standard provides guidance that is consistent with the expectations of regulatory authorities.

To establish microbiological control, it is important to understand the risks of contamination. Risk can be defined as the combination of the probability of occurrence of harm and the severity of that harm.1 This is one of the key purposes of the new standard. Understanding the risk can be achieved by considering the sources of the contamination, the likelihood of transfer, and the impact on product quality and ultimately the patient.

The approach outlined is based on Quality Risk Management (QRM) principles. An effective quality risk management approach can ensure high quality of product to the patient by providing a proactive means to identify and control potential quality issues during manufacturing. The new standard outlines the requirements for microbiological contamination control and guidance on qualification and verification of these environments. The standard is relevant to several different applications with particular use for pharmaceutical, biopharmaceutical, medical device, and other industries that use these environments such as life science (healthcare and hospitals) and food. There are many applicable standards already in place in some of these areas, whilst in others there is less guidance on microbiological control (e.g., medical devices and food). It is important to note that this standard is limited in scope to viable microbiological contamination and excludes any considerations of endotoxin, prion, and viral contamination.

There are also several annexes that provide further information on biocontamination control specific to certain applications. When a contamination control system is established, implemented, and maintained it should consider identification of all microbial contamination sources and routes of contamination, assess the risk from these sources, and introduce control methods to reduce the identified risks. There should also be a monitoring program to monitor the identified sources.

To avoid contamination and establish microbiological control, it is important to understand the risks of contamination. This is achieved by considering the sources and the likelihood of transfer and how this will impact both product quality and the patient or consumer.

Understanding risk

There are many elements to a successful approach in microbiological control, but the evaluation of the risk to quality should be based on scientific knowledge. Having a good understanding of the microflora present is vital. Contamination can come from a variety of sources (e.g., personnel, materials, equipment, air, and the surrounding environment). When there is a risk of product or process contamination from particular types of organisms, these are considered microorganisms of interest (or objectionable organisms). It is important to be able to accurately identify these organisms of interest to assess either the possibility of their survival or if they are likely to produce toxins. As part of the risk assessment, the potential for causing contamination of the product or harm to the patient should be assessed.

There are many ways to characterise or identify microorganisms isolated from cleanroom environments (e.g., Gram stain, phenotypic, genotypic, etc.) and being aware of the limitations of these is vital. Basic characterisation such as Gram reaction gives general information about the likely source of the contaminant, but more detail may be required to fully understand the risk posed. It is also recognised that the Gram stain procedure is one of the most frequent causes of incorrect identification of bacteria.2 The Gram staining reaction observed from a bacterial strain does not always correspond to its Gram type, and the multiple steps, and controlled times required for each step, mean there is potential for analyst error, as well as user interpretation of the result. This has the potential to provide incorrect results and is difficult to align with data integrity requirements. With the Gram stain being the first step in many phenotypic identifications, it can lead to erroneous results from these systems. Different methods of identification have varying levels of accuracy. For example, it has long been recognised that genotypic methods are more accurate and precise than phenotypic methods3 as they are comparing conserved regions of DNA (e.g., 16S for bacteria); however, genotypic methods are more time-consuming and expensive, and the higher level of accuracy may not be required for non-critical samples (e.g., isolates from a grade D environment). But, it is still important to have an identification that is as accurate as possible that will aid in risk assessments, trending, and future investigations. In recent years Matrix Assisted Laser Desorption Ionisation Time of Flight (MALDI-TOF) microbial identifications have allowed for faster, more cost-effective identifications. And whilst this technology is not as accurate as sequencing, it has the potential to be much more accurate than phenotypic methods, due to analysing ribosomal proteins that are closer to the DNA of the organism rather than analysis of biochemical reactions (such as fermentation, acid and salt tolerance metabolism, etc.).

When there is the potential for incorrect identifications any subsequent remedial actions may not address the issue, and the problems will continue to occur. Similarly, high rates of unidentified isolates make assessing the specific risks more difficult.

As outlined in the introduction of the new standard, there must be a formal system of microbiological control that identifies, controls, and monitors contamination on an ongoing basis. This is a process of continual improvement and monitoring plans should be reviewed and revised based on results of the monitoring data. Whilst the method of identifying any contamination is only a small part of the process, it is nonetheless a critically important one as the identification generated will determine the risk assessment and subsequent actions. Simply generating “a name” is not enough.

Whilst the technology utilised for identification methods plays a role in the accuracy of the result, one aspect that is often overlooked is the library or database that the system relies upon to generate the ID. This is outlined in the Pharmacopoeias. For example, European Pharmacopoeia (10.0) chapter 5.1.6 states:

  • Databases are part of the systems and are included in the primary validation. As identification methods depend on the use of databases, the extent of coverage of the database with respect to the range of micro-organisms of interest must be taken into account during validation.

United States Pharmacopeia <1113> similarly states:

  • Microbiological identification systems are based on different analytical methodologies and limitations may be inherent to the method and/or arise from database limitations. Identification is accomplished by matching characteristics (genotypic and/or phenotypic) to an established (reference) organism such as a type strain. If a microorganism is not included in the database, it will not be identified.

It is important to note that when an identification system does not have sufficient coverage it does not always result in a “no identification.” It may also result in an incorrect identification, meaning any further risk assessment will be based on incorrect information, which potentially is worse than no identification.

Many commercially available systems were developed to address clinical applications and have a focus on clinically relevant species in their libraries. Library coverage for the sorts of organisms isolated from cleanroom environments may not be comprehensive enough to generate accurate identifications.

Understanding the types of organisms present is essential to making an accurate assessment of the risk to product or patient and tracking the likely source of the contaminant is made easier. Being able to review a database of previous results allows users to determine how the contaminant may have gotten into a critical area. For example, an organism found in a grade A room may have been seen previously in a surrounding area and would be the likely source. Having the ability to link isolates will aid investigations when out-of-specification results are detected.

The standard requires that microorganisms of interest should be identified, and survival possibility considered. A microorganism of interest is defined as an organism that is harmful to the product, the process, or the patient. It is especially important to be able to identify isolates to determine if there may be some resistance to steps taken to remove contamination, e.g., disinfection regime or sterilisation. Dependent on the process, there are specific organisms that can pose challenges. For example, the risk of spore formers is well understood, and cleaning schedules may require the use of sporicidal agents when these are isolated and identified. However, different species will have different resistance to the agents used. Awareness of the species present will help to understand the possibility of survival of the cleaning and disinfection program and define any changes needed to be implemented, e.g., contact times and type of sporicides used.

When sterilisation steps are used (and dependent on which method is utilised) there may be certain species that are more likely to survive these processes. For example, heat and steam is a common sterilisation step, but many bacterial spores show heat resistance especially when the pharmacopeial reference method can’t be used (i.e., 121 oC for 15 minutes). For those products sensitive to the temperatures employed in this method, lower temperatures may be utilised, meaning that the potential risk to the product from bacterial spores should be considered. When lower heat processing cycles are used it will be important to identify any microbes and assess for heat resistance.

A further example is when irradiation is used as a terminal sterilisation process, which is very common in the medical device industry. There are many species known to have resistance to irradiation (e.g., Methylobacterium spp., Acinetobacter spp.), but Deinococcus spp. are a group of organisms that are recognised as being very resistant to irradiation with some studies demonstrating survival at doses of 30 kGy.4 Obviously isolating a species from this genus should be of concern when irradiation is used as the terminal sterilisation process. Whilst studies into these species highlight their existence in specific environments such as deserts or canned meats5, they have also been found in cleanroom environments. From data generated at Accugenix (a global microbial identification laboratory specialising in cleanroom isolates), species from this genus have been identified from cleanroom environments more than 5,600 times. Being able to isolate and accurately identify organisms like these that may be present in the cleanroom is vital. Manufacturers should be aware of the presence of these species in the clean environment and the possibility of survival of any sterilisation process and potential risk to product and patient.

Many products also contain potential growth substrates that may allow for growth within the final product, affecting product quality. Microorganisms are metabolically diverse, particularly filamentous fungi, which have been shown to utilise nutrient sources as varied as plastic6 and even radiation.7 Presence of certain microorganisms in the product may result in spoilage.

The presence of moulds or other microbiological contamination including objectionable organisms can be indicators of poor cleaning or poor design and increases risk to the product or patient. An accurate understanding of which species is present in the clean environment will help accurately assess the risk to the product and patient.

Demonstrating Control: Trending

Data generated from the Environmental Monitoring program should be analysed and the data trended to help identify any adverse trend that may require corrective or preventative action (CAPA). The data should be reviewed by a competent microbiologist to identify any possible issues such as unexpected quantitative or qualitative shifts in the types of organisms recovered due to changes in season, disinfection program, operator population, raw materials, or facility controls. Single results are often not significant.

When the data generated by the monitoring system is comprehensive and consists of accurate reliable identifications, analysis of the information should allow for proactive action e.g., if a microorganism of interest is isolated from an uncontrolled or lesser controlled area, action can be taken to try and prevent this from being transferred to the critical areas or even product. There are many methods of statistical analysis, but the data should be reviewed regularly, which can indicate a developing problem. It should also promptly identify out-of-specification results. Accurate identifications are vital to ensure that the data analysed for trends is reliable for subsequent decision and action. Inconsistent identification methods that may generate different names for the same isolate or fail to identify at all will make trending ineffective.

The tool used for managing data should allow for easy identification of trends but also allow for alerts when an organism of interest is identified. It should also comply with data integrity requirements. Spreadsheets are still commonly used, and whilst they can be useful and powerful, they can be difficult to validate and comply with data integrity requirements. Microorganism names are often long and complicated, and transcription of these names can sometimes introduce errors, meaning that results may be excluded dependent on how the trending tool is used.

Out-of-Specification Investigation

The objective of out-of-specification (OOS) investigations is to determine if there has been a real change in the occurrence of microbial contamination. Investigations should be prompt, thorough, and scientifically sound to determine the cause and appropriate corrective action. A species-level Identification is an important part of this process and when an OOS is detected the first step should be to identify the species. This will help determine the risk to the product or process, determine the likely source of the contamination (tracing the same species via the historical database), and then help define the appropriate CAPA. Accurate identifications are also important when there is an out-of-expectation (OOE) result. During monitoring there will be typical resident flora but when a new species is detected, this may trigger an OOE investigation to determine the risk of this species and the route of transfer.


The new standard has an emphasis on a Quality Risk Management approach that should be based on scientific knowledge. Identifying contamination and the potential sources of that contamination are essential to demonstrate and maintain control and ultimately link to patient safety. Successful monitoring of clean environments requires not only having a system in place to detect microbiological contamination but also a system to accurately identify, to species level, organisms isolated. This allows for assessment of the risk posed by the species found and provides a way of accurately tracking back to the likely source of the contaminant.

Modern, scientifically sound methods will allow for accurate identification of isolates to determine the risk to the product, process, or patient.


  1. International Conference on Harmonisation. Q9 Quality risk management. 2005.
  2. World Federation for Culture Collections. Technical Information Sheet no. 9: A Standardized Gram Staining Procedure.
  3. US Food and Drug Administration. Guidance for industry: sterile drug products produced by aseptic processing—current good manufacturing practice. Silver Spring, US: FDA; 2004.
  4. Rainey FA, Ray K, Ferreira M, Gatz BZ, Nobre F, Bagaley D, Rash BA, Park M-J, Earl AM, Shank NC, Small AM, Henk MC, Battista JR, Kämpfer P, da Costa MS. Extensive Diversity of Ionizing-Radiation-Resistant Bacteria Recovered from Sonoran Desert Soil and Description of Nine New Species of the Genus Deinococcus Obtained from a Single Soil Sample. Applied and Environmental Microbiology. 2005;71(11):7630-. doi: 10.1128/aem.71.11.7630.2005.
  5. Anderson A. Studies on a radioresistant micrococcus. 1. Isolation, morphology, cultural characteristics, and resistance to γ radiation. Food Technol. 1956;10:575-8.
  6. Russell JR, Huang J, Anand P, Kucera K, Sandoval AG, Dantzler KW, Hickman D, Jee J, Kimovec FM, Koppstein D. Biodegradation of polyester polyurethane by endophytic fungi. Applied and environmental microbiology. 2011;77(17):6076-84.
  7. Dadachova E, Casadevall A. Ionizing radiation: how fungi cope, adapt, and exploit with the help of melanin. Current opinion in microbiology. 2008;11(6):525-31.

You may also like