Environmental Monitoring Risk Assessment
Environmental Monitoring Risk Assessment
Environmental Monitoring Risk Assessment
INTRODUCTION
Over the past decade, environmental monitoring has become more sophisticated in moving from
random sampling, using an imaginary grid over the room and testing in each grid, to the current
focus on risk assessment and the use of risk assessment tools to determine the most
appropriate methods for environmental monitoring.
This paper explores current trends in the application of risk assessment to the practice of
environmental monitoring by examining the following key areas:
Determining the Frequency of Monitoring: Using the concept of risk assessment to decide
how often to monitor different types of cleanrooms
Risk Assessment Tools: Applying risk assessment tools to establish methods for
environmental monitoring
The examples used are from a sterile drug manufacturing facility and focus mostly on aseptic
filling; however, the concepts and tools are applicable to the environmental monitoring of other
types of manufacturing and packaging operations.
Using a criticality factor is a means of assigning a monitoring frequency based on the risk
assessment of each critical area. The risk assessment relates to the potential product impact
from any risk. For example, an area of open processing at an ambient temperature, a long
exposure time, and the presence of water, would constitute a high risk and would attract a
higher risk rating. In contrast, an area of closed processing, in a cold area, would carry a
substantially lower risk and associated risk rating.
Using a range of 1 to 6, with ‘1’ being the most critical and ‘6’ the least critical, a score of 1
would be assigned to an aseptic filling operation; a score of 2 to final formulation, a score of 3
to open processing, and so on. Each user must adapt such a scheme to his or her particular area
and defend it by way of supportable rationale. An example of monitoring frequencies under such
a scheme can be seen in Figure 1, and an example of its application is seen in Figure 2.
Each controlled area would be evaluated against set criteria and, with the use of a series of
guiding questions, the monitoring frequency would be determined. Decision criteria include
considerations in two category areas: areas of higher weighting and areas of higher monitoring
frequency. Examples of these categories follow:
‘Dirtier’ activity performed in a room adjacent to a clean activity, even if the clean
activity represents later processing
Areas that have a higher level of personnel transit (given that people are the main
microbiological contamination source). This may include corridors and changing rooms.
Routes of transfer
Component preparation activities and sites Duration of activity (such as a lower criticality
for a 30-minute process compared to a six-hour operation)
Once the monitoring frequency for each controlled area is determined, it should be reviewed at
regular intervals. This review may invoke changes to a room’s status, and hence, its monitoring
frequency, or to changes for different sample types within the room. For example, it may be
that after reviewing data for one year, surface samples produce higher results than air samples
for a series of rooms. In this event, the microbiologist may opt to vary the frequency of
monitoring and take surface samples more often than air samples. There would also be an
increased focus on cleaning and disinfection practices, and their frequencies, based on such data
(Sandle, 2004b).
When both types of monitoring are producing low level counts, the balance of risk would be
towards air samples. This is because air samples are direct indicators of the quality of the
process and assign a level of control to the process, whereas surface samples are indicators of
cleaning and disinfection. If the results of surface samples are generally satisfactory, as
indicated by trend analysis, then either the number of samples or the frequency at which they
are taken can be reduced. If subsequent data showed an increase in counts, the monitoring
frequency could easily be restored. Indeed, all types of monitoring frequencies may increase as
part of an investigation, as appropriate. Therefore, the criticality factor approach not only sets
the requirement for a room, it can also be used to vary the sample types within a room
(Ljungqvist and Reinmuller, 1996).
Once the status for each room has been selected, a risk assessment procedure is required to
determine locations for environmental monitoring. Such risk-based approaches are
recommended in ISO 14698 and regulatory authorities are increasingly asking drug
manufacturers about this subject.
Risk-based approaches include Failure Mode and Effects Analysis (FMEA), Fault Tree Analysis
(FTA), and Hazard Analysis and Critical Control Points (HACCP), all of which employ a scoring
approach. (Other approaches include: Failure Mode, Effects, and Criticality Analysis (FMECA);
Hazard Operability Analysis (HAZOP); Quantitative Microbiological Risk Assessment (QMRA);
Modular Process Risk Model (MPRM); System Risk Analysis (SRA); Method for Limitation of
Risks; and Risk Profiling.)
At present, no definitive method exists, and the various approaches differ in their process and in
the degree of complexity involved. However, the two most commonly used methods appear to
be HACCP, which originated in the food industry, and FMEA, which was developed for the
engineering industry (Whyte and Eaton, 2004a).
Taking into account changes to the work process and seasonal activities
These risk assessment approaches are not only concerned with selecting environmental
monitoring locations. They integrate the environmental monitoring system with a complete
review of operations within the cleanroom to ensure those facilities, operations, and practices
are also satisfactory. The approaches recognise a risk, rate the level of the risk, and then set out
a plan to minimise, control, and monitor the risk. The monitoring of the risk will help to
determine the frequency, locations for, and level of environmental monitoring (for example,
refer to an article by Sandle [2003a], for a more detailed example).
This paper explores an example from three different techniques:
FMEA
Tabular Approach
An example using a simple table for analyzing risk in environmental monitoring situations
appears in Figure 3.
HACCP
Each of these seven key points is a vital step in developing the risk assessment.
Examples include:
Areas adjacent to Cleanroom or Isolator (e.g.: airlocks, changing rooms)
Air supply and Room air
Surfaces
2. Assess the importance of these sources and determine whether or not they are
hazards that should be controlled.
Examples include:
Amounts of contamination on, or in, the source that is available for
transfer
Ease by which the contamination is dispersed or transferred
Ease with which the contamination can pass through the control
method
The use of a scoring method can greatly help in assessing the
relative importance of these contamination sources.
For example:
Air Supply: High Efficiency Particulate Air (HEPA) filters
Dirty Areas adjacent to Cleanroom or Isolator: differential pressures,
airflow movement
Room Air: air change rates, use of barriers
Surfaces: sterilisation, effectiveness of cleaning and disinfection
procedures
People: cleanroom clothing and gloves, room ventilation, training
Machines and Equipment: sterilisation, effectiveness of cleaning, exhaust
systems
4. Determine valid sampling methods to monitor either the hazards or their control
methods or both.
For example:
HEPA filter integrity tests
Air supply velocity, air change rates
Room pressure differentials
Particle counts
Air samplers, settle plates, contact plates, etc
5. Establish a monitoring schedule with ‘alert’ and ‘action’ levels and the corrective
measures to be taken when these levels are exceeded.
For example:
The greater the hazard, the greater the amount of monitoring required
Trend analysis for alert and action levels, in or out of control
For example:
System for data review
Examine filling trials
Audits
Reassess - hazards, effectiveness of control systems, frequency of
monitoring, appropriateness of alert and action levels
For example:
Describe the steps being taken
Describe the monitoring procedures
Describe the reporting and review procedures
Before implementing HACCP, it is important to train all staff involved in the process and to use a
multi-disciplinary team. For example, the team may be comprised of personnel from Production,
Engineering, Quality Control (QC), Quality Assurance (QA), Validation, and so on.
FMEA
FMEA schemes vary in their approach, scoring, and categorisation. All methods share a
numerical approach. The example presented here, based on a sterility testing isolator, assigns a
score (from 1 to 5) to each of the following categories:
Severity
Occurrence
Detection
Where:
o Detection is based on the monitoring systems in place and on how likely a failure
can be detected
By asking a series of questions, each main part of the cleanroom or isolator system can be
grouped or classified into key parts.
What is the function of the equipment? What are its performance requirements?
How can it fail to fulfil these functions?
What can cause each failure?
How much does each failure matter? What are its consequences?
The scoring is 1 (very good) to 5 (very bad). Therefore, a likelihood of high severity would be
rated 5; high occurrence rated 5; but a good detection system would be rated 1.
A third component of the risk assessment approach is to evaluate a risk once an activity has
taken place. Then, by using a largely numerically-driven set of tools, repeatability and
reproducibility can be ensured. Examples of individual out-of-limits results and data-sets relating
to an operation are examined below using examples from an aseptic filling process. Following
this, an example of an overall assessment of different processes over time is explored.
Numerical approaches are useful in applying a level of consistency between one decision and
another.
Individual Assessments
The section below details some methods that can be used to quantify the risk of contamination
in pharmaceutical cleanrooms. The models outlined are based on the work performed by Whyte
and Eaton (2003a and b).
Estimating the Risk to Product Using Settle Plate Counts
The method applies to the assessment of settle plates at the point-of-fill, under the
Grade A zone. It allows an estimate of the probable contamination rate to the product as
derived from the following equation:
The fixed value is the area of the petri dish, which for a 90mm plate, is 64 cm2.
The formula can also be applied to the monitoring of product filtration activities when ‘1’ is
entered as a constant for neck area of product.
There is no available guide as to what percentage constitutes which level of risk. The 0.03%
figure has been used by some practitioners. This is based on the Parenteral Drug Association
Survey of Aseptic Filling Practices (2002), where it is common in the pharmaceutical industry to
allow 0.03% of broth bottles in a media simulation trial to exhibit growth at a ‘warning level’
(where 0.03% = 1/3,000, with 3,000 being the average size of a media fill). An ‘action level’ is
often set as 3/3,000 bottles or 0.1%. This would constitute a high risk. Logically, the range
between 0.03 and 0.1 would be a medium risk (Whyte and Eaton, 2004c).
Therefore, where the ‘risk’ is that of micro-organisms detected on a settle plate, with a
probability of <0.1% depositing in the neck of a bottle when bottles are exposed in a
unidirectional air flow, risk categories would be as shown in Figure 5.
Finger Plate Assessment
The formula can readily be applied to operations that relate to Grade A operations, for example:
filtration connection, vessel to filling machine connection, the filling activity, and loading a
freeze-dryer. Where the operator is only present in the Grade B room and has no impact on the
Grade A operation, this is automatically considered to be low risk if there are no other special
factors. (Low risk does not imply lack of action or assessment. However, it aims
conceptualisation of the result in terms of probable risk to the batch.)
In this example of a finger plate assessment, the location, activities, and duration require
weighting. Examples of logic that apply to the rating of the location, activities, and duration
categories can be seen in Figures 6, 7, and 8, respectively
Finger Plate Assessment Worked Example
A finger plate with a count of 1 cfu for an activity at point-of-fill, using forceps, that lasts for one
minute.
The score produced would be rated according to standard risk assessment categories:
These risk ratings are based, in part, on the worked example. Based on historical data over the
past six-months, the highest record example of a Grade A intervention finger plate is a count of
2 cfu: using forceps to retrieve a fallen vial and lasting for more than 120 seconds. This would
have given a score of 7.5, which falls within the medium risk category. The user should develop
a scheme that fits his or her facility (Whyte and Eaton, 2004b).
Where:
Where a count of 2 is detected from a conveyor belt (a filling machine non-product contact
location)
This scoring scheme is based on contamination of a product contact site being high risk by
virtue of its direct proximity to the critical area or the product.
A count of 1 cfu on one of these product contact site locations would give a score of 9.4. In most
filling zones and clean zones, sample results from product contact sites would be expected to
record zero counts for 999 samples out of every 1000. Whereas, a count of 3 from a non-
product contact site would result in a medium risk category.
Approaches are available for the risk assessment of active air samples that use a numerical
system. However, the formulae associated with these are difficult to calculate in practice
because often all information is not available and assessment of variables, such as impaction
speed, are not readily calculable. Therefore, a qualitative assessment, such as the one included
in the example of the numerical approach, may be more suitable.
Alternatively, non-numerical risk assessment can be used based on the proximity and the
operation. See Figure 12 for this example.
The table shown in Figure 13 is proposed as a tool for risk assessment and to aid investigations.
It supplements the risk assessment tools that have been previously examined.
An Overall Assessment
The approach taken for an overall assessment involves the historical examination of a number of
operations and assigning a value above which the operation is considered to be atypical. A 95%
cut-off is considered to be the most suitable cut-off point.
CRITICALITY SCORING
Criticality scoring is a way of assessing the totality of results from an environmental monitoring
session. This may be, for example, a batch fill. The data from the session is examined and
points are awarded for each result above a pre-set warning or action level. The total score is
then summed and the results obtained are compared to a set level at which atypical sessions
are indicated.
The pre-set level would be assessed from historical data over a reasonable time period (such as
one year). An example of such a scheme follows:
For Grade A
The results from a filling operation are examined (for individual viable counts and for the mean
particle counts taken during the fill). Each result, which equals or exceeds a warning or action
level, is scored according to the criteria in Figure 15 and Figure 16. Using the criteria presented
in Figures 15 and 16 produces the Grade A score.
For Grade B
The results from a filling operation are examined (for the individual viable counts and the mean
particle counts taken during the fill). Each result that equals or exceeds a warning level is scored
according to the criteria in Figure 17 and Figure 18.
If a warning level or action level is of the same count (cfu) as a value in the count (cfu) column,
the warning or action level score should be selected. This produces the Grade B score. To
produce the total criticality score, the two scores are added together (Grade A + Grade B).
Once the data has been generated, the score at which approximately 95% of the fills would be
below (and 5% would be above) can be calculated. That figure would then be used as the cut-
off value with which to assess the ‘atypical’ filling operations.
Figure 19 displays a simple representation of this assessment. For the data set, the criticality
score was calculated at 25, which corresponded with the 95th percentile for a set of data from
the filling of an example drug.
The graph in Figure 19 indicates that some fills exceeded the cut-off criticality value during a
particular time period (see fill numbers 12 through 15). After some corrective action, the scores
for the fills were reduced (see fill numbers 16 through 29) and the situation returned to a state
of control.
CONCLUSION
The use of risk assessment approaches is an important current Good Manufacturing Practice
(cGMP) topic in microbiological environmental monitoring. This paper has outlined some possible
tools for such a risk assessment approach; however, each suite of cleanrooms or isolator will be
subtly different. The microbiologist must consider each aspect of the environment and decide
what level of monitoring best suits his or her system, and then must justify the techniques used
and the locations selected.
The approach adopted should be detailed in a written rationale and approved by senior
management. After this, a rigorous and defensible system will be in place to satisfy regulatory
expectations, and to aid the user in assessing the risk of problematic environmental monitoring
situations or results.
Article Acronym Listing
REFERENCES
BS EN ISO 14698 – 1:2003: ‘Cleanrooms and Associated Controlled Environments – Bio-
Contamination Control – Part 1: General Principles and Methods.’
Code of Federal Regulations, 1998, Title 21, Part 210, Current Good Manufacturing Practice in
Manufacturing, Processing, Packaging, or Holding of Drugs – General, 210:3.
'Guidelines on Sterile Drug Products Produced by Aseptic Processing,' FDA, 1987 (draft produced for
review in August 2003).
ISO 14644-1 Cleanrooms and Associated Controlled Environments – Classification of Air Cleanliness.
ISO 14644-2 Cleanrooms and Associated Controlled Environments – Specifications for Testing and
Monitoring to prove continued compliance with ISO 14644-1.
ISO 14698-1 Cleanrooms and Associated Controlled Environments Biocontamination Control – Part 1:
General Principles.
Jahnke, M. ‘Introduction to Environmental Monitoring in Pharmaceutical Areas,’ PDA, 2001.
Ljungqvist, B. and Reinmuller, B. ‘Some Observations on Environmental Monitoring of Cleanrooms,’
European Journal of Parenteral Science, 1996 1: 9 –13.
PDA Technical Report No. 13 (revised): 'Fundamentals of an Environmental Monitoring Programme,'
September/October 2001.
Sandle, T. ‘The Use of a Risk Assessment in the Pharmaceutical Industry – the Application of FMEA to
a Sterility Testing Isolator: a Case Study,’ European Journal of Parenteral and Pharmaceutical
Sciences, 2003 8(2): 43-49.
Sandle, T. ‘Selection and Use of Cleaning and Disinfection Agents in Pharmaceutical Manufacturing’ in
Hodges, N and Hanlon, G. (2003): ‘Industrial Pharmaceutical Microbiology Standards and Controls,’
Euromed Communications, England.
Sandle, T. ‘General Considerations for the Risk Assessment of Isolators Used for Aseptic Processes,’
Pharmaceutical Manufacturing and Packaging Sourcer, Samedan Ltd, Winter 2004, pp: 43-47.
USPNF#25 <1116>
Whyte, W. ‘Cleanroom Technology: Fundamentals of Design, Testing and Operation, 2001.’
Whyte, W. and Eaton, T. ‘Microbiological Contamination Models for Use in Risk Assessment during
Pharmaceutical Production,’ European Journal of Parenteral and Pharmaceutical Sciences, 2004 Vol.
9, No.1, pp: 11-15.
Whyte, W. and Eaton, T. ‘Microbiological Risk Assessment in Pharmaceutical Cleanrooms,’ European
Journal of Parenteral and Pharmaceutical Sciences, 2004, Vol. 9, No.1, pp: 16-23.
Whyte, W. and Eaton, T. ‘Assessing Microbial Risk to Patients from Aseptically Manufactured
Pharmaceuticals,’ European Journal of Parenteral and Pharmaceutical Sciences, 2004, Vol. 9, No.3,
pp: 71-79.
SUGGESTED READING
Anon. (1998), The Gold Sheet, Vol. 32, No.10, October 1998.
De Abreu, C., Pinto, T. and Oliveira, D. ‘Environmental Monitoring: A Correlation Study between Viable
and Nonviable Particles in Cleanrooms,’ Journal of Pharmaceutical Science and Technology, Vol. 58,
No.1, January-February 2004, pp: 45-53.
Kaye, S. ‘Efficiency of Biotest RCS as a Sampler of Airborne Bacteria,’ Journal of Parenteral Science
and Technology, Vol. 42, No.5, September-October 1986, pp: 147-152.
Meir, R. and Zingre, H. ‘Qualification of Air Sampler Systems: MAS-100,’ Swiss Pharma, 22 (2000),
pp: 15 – 21.
Ohresser, S., Griveau, S. and Schann, C. ‘Validation of Microbial Recovery from Hydrogen Peroxide-
Sterilised Air,’ Journal of Pharmaceutical Science and Technology, Vol. 58, No. 2, March-April 2004,
pp: 75-80.
PhRMA Environmental Monitoring Work Group ‘Microbiological Monitoring of Environmental
Conditions for Nonsterile Pharmaceutical Manufacturing,’ Pharm. Technol., March 1997, pp: 58-74.
Reich, et al. ‘Developing a Viable Microbiological Environmental Monitoring Program for Nonsterile
Pharmaceutical Operations.’ Pharm. Technol., March 2003, pp: 92-100.
'Rules and Guidance for Pharmaceutical Manufacturers and Distributors' ('EU GMP Guide'), MHRA,
2002.
Sandle, T. ‘Environmental Monitoring in a Sterility Testing Isolator,’ PharMIG News No. 1, March 2000.
Sandle, T. ‘Microbiological Culture Media: Designing a Testing Scheme,’ PharMIG News No. 2, August
2000.