Published on IVT Network (http://www.ivtnetwork.

com)

A Risk Matrix Approach for Media Simulation Trials

,
,
Clare Leavy Gareth Needham
Tim Sandle, Ph.D. Jan 8, 2013 6:00 am PST
By

Getty Images
ABSTRACT
This paper examines the adoption of a risk-based matrix approach for the selection of product simulations used when
conducting media filling trials. Media trials are a regulatory requirement for aseptic processing manufacturers. Given that not
every product combination can be assessed on multiproduct filling lines, some assessment criteria is required to select the
"worst-case" filling run parameters. This paper discusses the main criteria suitable for such an exercise and illustrates the
application using a case study.
INTRODUCTION
Aseptic processing is generally regarded as the most "at risk" pharmaceutical process to conduct, primarily due to the
possibility of microbial contamination, and because the product, often due to its inherent nature, cannot be subjected to a
terminal sterilization process. A number of environmental controls, particularly in relation to cleanroom and clean air design,
are built into the process to protect the product (1). Personnel are required to follow strict disciplines in relation to good aseptic
technique. One of the periodic assessments undertaken to measure the likelihood of non-sterility is the media simulation trial
(2).
With media simulation trials, a microbiological growth medium is used in place of the product and filled as if it was product
under the ordinarily processed conditions. Media fills start at the beginning of filling operations (immediately after the line setup), during and after manipulations and interventions, and until the last vial has been filled. For conducting media simulation
trials, regulatory guidance provides an outline of what needs to take place, and the acceptance criteria (ideally zero growth in
any filled container) that needs to be adopted. Furthermore, there are several references in literature that provide best practice
advice (3-5).

What is less clear is where there are filling lines used for multiproduct filling. A media fill can not measure clearly every
combination of vial and container-closure at each required interval; with large facilities, a rotational program could take many
years to complete (which would additionally mean that either a problematic fill or the combination deemed to present the
greatest potential contamination risks to the product would not be assessed very often). To address this issue, many
manufacturers adopt a matrix approach in order to assess the product types deemed to be of the greatest challenge to the
process or that are at a greater risk of microbial contamination.
The objective of using a matrix approach is to assess the range of factors associated with aseptic filling that, when combined,
potentially create conditions or circumstances for product contamination. Some regulators refer to this combination as the
"worst-case". Therefore, the term "worst-case," in the context of media fills, is taken to be the combination of events and
circumstances that could, in theory, expose a product (or, in this case, a product simulation) to the greatest chance of
microbial contamination. This paper goes onto examine these risk factors. The factors include vials that have the widest neck
diameter (thereby providing a larger target for the deposition of contamination), fill speeds or fill volumes (that lead to vials
remaining at the point of fill for the longest time), larger media fill runs (where the longest fills, that over the course of time,
could lead to a higher frequency of contamination events occurring), and the type and nature of personnel interventions
(where the act of a person intervening into a critical zone poses a potentially high contamination risk; some interventions, due
to their complexity or time taken to complete, pose a greater risk than others). No single risk factor necessarily presents a
"worst-case." However, one combination of factors will probably present a greater potential risk to a product than another
combination and thus may be considered as greater risk and "worst-case". It is the argument of this paper that, by placing
these factors into some form of matrix, the "worst-case" is easier to visualize and justify. Once the "worst-case" combinations
have been found, it is possible to draw up a rationale for the selection of these conditions for the execution of scheduled media
fills.
Whilst there is no clear or satisfactory definition of "worst-case," given the common usage of the term, this paper makes
reference to the term in relation to the selection of aseptic filling factors, which leads to the greatest potential chance of a
contamination event occurring.
There is little guidance on how a risk-based matrix should be devised. In going some way to address this, this paper examines
the matrix approach for media trials by first, discussing the different criteria that could be used for this process, and secondly,
by using a case study to examine the optimal combination for two filling lines. The case study is based on liquid media fills.
MEDIA SIMULATION TRIALS
Media simulation trials are an established feature of aseptic processing, and they are a regulatory requirement for
manufacturers of aseptic products (6, 7). Media trials use a microbiological growth medium in place of product to simulate the
product fill process; the media is processed in a manner identical to that in which the product is processed. Media simulations
provide information as to whether the process compromises the sterility of individual components and finished product. As
such, media fills are designed to evaluate the aseptic assembly and operation of the critical equipment, qualify the operators
and assess their technique, and demonstrate that the environmental controls are adequate to meet the basic requirements
necessary to produce a sterile drug by aseptic processing.
Media fills are conducted regularly to verify established aseptic filling processes (typically every six months). In addition, media
fills are also used in validation of aseptic processes as one of the final stages of a performance qualification (where a
minimum of three trails are typically conducted). In order to provide a robust test, a media fill program must incorporate the
contamination risk factors that occur on a production line and accurately assess the state of process control.
Media trials must also be representative of the types of products filled on an aseptic processing line. In order to ensure that
media fills are representative, the fill must replicate the conditions under which product filling takes place, and they must be
undertaken under "worst-case" conditions so as to provide a realistic challenge. To ensure that each of the key factors is
captured, a protocol or rationale should be devised. Such a document should cover:
Identification of the cleanroom
Identification of the filling line and equipment
Type of container/closure to be used
Line speed
Number of units to be filled
Number and type of interventions
Number of personnel to participate

Type of media to be used

Volume of medium to be filled into the containers
Incubator identification
Incubation time and temperature
Batch record details
Acceptance criteria.
For such criteria, a risk-based approach is increasingly common. Risk-based approaches consider an assessment of the steps
and interventions that could potentially compromise the sterility of the product (8), such as the number and type of
interventions to be used during process simulation (9).
These criteria are additionally important for many pharmaceutical facilities that contain multiple filling lines upon which a
variety of different types of products in different presentations are filled.
ARGUMENT FOR A MATRIX APPROACH
Where there are multiple product combinations filled on the same filling line, it is not necessary, or even desirable, to conduct
media fills for every vial size and container-closure combination. Instead, manufacturers select representative vial and closure
combinations designed to be representative and "worst-case" in terms of having, theoretically, a greater risk of contamination
occurring.
The reason for doing so is because one media fill, conducted for each line every six months, is normally sufficient in order to
assess the risks of a consequential failure (a major breach of aseptic operations); whichever combinations are selected to be
representative should be examined at regular intervals. Testing a different arrangement at each scheduled media fill interval
could lead to a different type of media fill taking place at each interval; it would therefore take several years to complete the
media fill cycle.
Therefore, within the media fill program, it is sensible to ensure that, on multi-container filling lines, the selected container
sizes are filled at least once in a reasonable time frame (approximately over two years). Otherwise, the possibility of
unexpected contamination as it relates to a particular size may never be addressed.
This approach implies that certain combinations must be selected. The common way to select combinations for validation is
through the adoption of a matrix. In keeping with current pharmaceutical current good manufacturing practices (cGMPs), the
matrix should incorporate a risk assessment. Therefore, a risk-based matrix (alternatively called a bracketing of family)
approach is recommended whereby the "worst-case" product vial and closure combinations are selected. Such a matrix
approach is often referenced in literature (10-13) and by regulatory bodies (14, 15). However, there is rarely any guidance as
to how such a matrix approach can be established for media simulation trials. This paper attempts to redress this.
FACTORS TO CONSIDER
The most important assessment to be made, prior to beginning to construct a matrix, is selecting the factors. The factors refer
to the key aspects of the product fill required to ensure that the final selection of the media fill is both "representative and
indicative of "worst-case." In considering the most important factors for the matrix approach, the manufacturer should be
mindful that it is important to demonstrate to regulators that the media trials selected do indeed represent "worst-case," are a
sufficient process challenge, and closely simulate aseptic filling (16). For this purpose, this paper recommends consideration
of the following factors.
Different Product Types
A multi-purpose filling line should be used for either:
Filling one product in different sized containers
Filling different products in different sized containers
Filling different products in the same size of containers.
For the latter option, where a filling line is used for filling one product in different dosage forms with different formulations filled
into a particular size of container on the same aseptic filling line, then it is necessary to only perform one set of media fills on
that container size and line. This is because the probability of contamination arises in the process, not the product, and the
probability of non-sterility would be the same for each product filled on the line, provided that the container and container

closure is the same (17).

Process Variations
Most, if not every, aseptic process is unique. Even in the same factory, two lines set up for the simplest process, such as filling
liquid products into ampoules, could differ significantly one from the other. Therefore, the protocol should specify the
requirements for each filling line and product combination (18). This means that, if the same container and container-closure
combination is used on two filling lines, media trials should be considered for both lines; simply filling that combination on one
line is not a substitute for the other.
Vial Sizes
In theory, media fills should only be necessary for the container size that takes longest to fill and has the widest neck diameter.
This is on the basis that this combination presents the greatest potential for contamination and therefore addresses the
contamination potential for all smaller sizes. On this basis, the matrix approach would lead to the automatic selection of the
vial size, which is either of the largest size or has the widest neck diameter (or both).
However, this premise may or may not be correct, and evaluation is required. The neck size alone may be insufficient to use
solely as a factor because the neck/flange mold for different capacity vials varies. Furthermore, wide-necked containers may
turn out to be more stable than narrower-necked ones when each moves along the conveyor thus resulting in narrow-necked
containers being more susceptible to contamination. Moreover, many production lines that fill very small vials at fast speeds
can create the greatest number of jammed containers, over or under filled product, or require a higher frequency of manual
interventions. Thus, the wide-necked filling process could be less susceptible to contamination because there are fewer
personnel intrusions necessary for rectifying fallen containers.
Container-closures
In most cases the vial neck size is the key parameter (as discussed above), and the container-closure type is of less
relevance. However, some types of closures are less suited for aseptic processing or have been manufactured in ways that
cause batch-to-batch variation. This can lead to more stoppages or corrective interventions, and thus the closure type may
need to be considered.
Process Set-up
Some types of products require different set-ups, such as whether the product is passed to the filling machine down a line, or if
a vessel makes the connection. Where set-up manipulations are significantly different, consideration should be given to
different configurations when deciding which process simulations represent "worst-case."
Fill Volumes
In terms of fill volume, there should always be sufficient liquid in each container to "wet" all the surfaces during incubation. The
fill volume of the containers must always be sufficient to enable contact of all the container-closure seal surfaces when the
container is inverted and also be sufficient to allow the detection of microbial growth.
The volume of growth medium filled into each container is either reduced in relation to the volume of product typically filled, or,
alternatively, the exact volumes are replicated. Some manufacturers use identical volumes for small fills. The exact volume is
not replicated, but, with larger fills, the filling speed is adjusted to leave the containers open under the filling heads for the
same time as they would be in routine filling. Other manufacturers consider that only an exact replicate of the product fill
volume with media can achieve a complete test. Care must be taken if the latter approach is adopted since there must be
sufficient head space remaining to support the growth of aerobic microorganisms.
Number of Vials
The minimum number of units, according to some studies of probability, required for a realistic challenge is 3,000 units. This is
an expression of the minimum number of units for which a contamination rate of no more than one contaminated unit in 1000
units (0.1%) can be demonstrated with 95% confidence (19). The Pharmaceutical Inspection Convention (PIC/S) guidance,
which both US Food and Drug Administration and the European Union (EU) GMP support, uses either 5,000 or 10,000 units
as representative minima (20). In practice, for commercial products with lot sizes usually much bigger than 10,000 units, it
could be difficult to justify that a simulation run smaller than 10,000 units be considered representative of the actual process.
The key consideration for the risk matrix is how to represent the media fill in terms of time. Time is a key issue to assess, not

least for attempting to replicate operator fatigue (as employees are the main contamination source within cleanrooms, it
logically follows that tired employees may make more mistakes, and thereby present a greater contamination risk).
Should the run time be replicated, but not the number of vials filled, to assess time? That is having stoppages or slowing the
speed down (be careful that validated filling line parameters are not upset). Or should the maximum number of permitted vials
for a product fill be replicated for the media fill?
Some manufacturers adopt the second approach,and consider that the number of vials filled should both reflect the typical
product batch sizes, and that this, conversely, sets the maximum batch sizes for future product fills (within an upper tolerance).
Here vial numbers, rather than time, are used as the deciding criteria for the operation time (based on the premise that the
time taken to fill a media trial is taken to be the equivalent time to fill a product fill of similar size within normal operating
conditions). Other manufacturers use run time rather than vial numbers, which is discussed below.
If a manufacturer chooses to use the media fill to set the maximum product batch size, then in terms of an upper tolerance, it
would perhaps be over restrictive to abandon a product fill if it reaches the point when one additional vial is filled over the
maximum number of vials filled for a media fill. Equally, it would be unwise to continue filling indefinitely. Here a range of 2-5%
in excess of the media fill maxima would seem appropriate.
Run Times
The running time for the media fill is an important consideration. Whilst running a line for one hour compared to 10 hours does
not necessarily increase the likelihood of contamination, many contamination events have an equal likelihood of occurring at
the start fill as they do at the end of the fill. It is logical to test the time normally taken for the process from start to finish. The
speed and line configuration of the filling process are parameters that are highly dependent on the individual line being
challenged; media fills must be run for long enough to fill a statistically significant minimum number of units.
It is important that media fills last long enough to be able to simulate all of the potentially contaminating events that might
occur; it needs to be run for long enough to address the potential for contamination to build up over time (including replicating
the fatigue to which operators are subjected to in a product fill). To achieve this there are three possible options:
Operate the media fills to replicate the actual product fill in terms of vial numbers and run time.
Fill intermittently, with fill periods punctuated by periods of inactivity, over a period which embraces the maximum fill
time.
Running media fills at the end of a normal production run with the same personnel who have been working in the area
for a time, which enables the minimum number of vials to be filled.
Of these approaches, the former arguably provides a stricter challenge. This approach ensures that all staff shift
changes are captured, and it will detect for the possibility of the concentration of contaminants increasing in a
cleanroom over the time it is occupied and operational (although it would need to be a poorly designed cleanroom for
this to happen). In this scenario, some manufacturers fill all of the containers with media whilst others fill the containers
alternatively with media and sterile water (or just run some empty units in addition to those filled with media).
With the maximum run time option, it should be noted that by running the maximum number of vials the media simulation trial
run time would ordinarily last for longer than an equivalent product fill. This is due to the necessity of undertaking more
interventions that would (hopefully) occur during a product fill and allow for more shift change-overs (as more staff than normal
will need to attend the media trial in order to qualify). Furthermore, some manufacturers elect to encapsulate the maximum
stoppage time allowed for a product fill within the media fill.
Line Speeds
The media simulation program should address the range of line speeds employed during production. However, only one line
speed should be evaluated on each run. In theory, the slowest line speed represents the "worst-case" since the exposure of
the opened containers to the environment is the longest thus increasing the risk of contamination. In some cases, such as
when small containers are filled on high-speed lines, the highest speed can be the worst-case. Unstable containers can
generate a high number of line stoppages. Hence, with multi-use filling line, the most appropriate line speed for media fill
should be considered together with the container size. For example, one option could be to fill the biggest container size at the
slowest available speed, and the smallest container at the highest speed.

Interventions
Given that the primary source of contamination in aseptic processes is personnel, the inclusion of interventions in the media fill
is arguably the most important component (21). An intervention is a personnel activity whereby a task is carried out within the
International Organization for Standardization (ISO) class five critical zone (either within the unidirectional airflow device
protecting the filling machine or through a glove port as part of a restricted access barriers [RABS] or isolator).
Representative personnel interventions, which are typically undertaken during product fills, should be simulated during the
media trial. This is the case even though some of the events may be infrequently undertaken in practice. Whilst the actual
types of interventions will relate to the design of the filling line, some typical interventions include:
Filling machine set-up
Aseptic assembly of equipment and filling equipment preparation
Aseptic sample connections and disconnections
Microbiological environmental monitoring, which involves an invasion into the critical zone, such as a change of a settle
plate
Off-loading of vials
Replenishment of stoppers
Replenishment of containers in the container-feed, such as from a depyrogenation tunnel
Dealing with spillages
Glove change (for glove ports or isolators)
Adjustments of fill head assemblies
Filling machine adjustments, such as weight checks
Removal of containers that have fallen over
Repair or replacement of needles/tubes
Replacement of on-line filters
Removal of containers that have missing stoppers
Unblocking vial jams (22, 23).
In setting interventions, the assessment of "worst-case" should both consider those interventions that have occurred on the
filling line between media trials and rank interventions based on their complexity. The criteria for ranking may include:
The duration of the intervention
The perceived complexity of the activity
The types of implements or tools required
The familiarity of the operators with the task
Whether the filling team or an engineer, who is less familiar with working on the line, undertakes the intervention
The association of an intervention with any previous media fill failure.
Through this exercise, those interventions that have the potential for the greatest chance of product combination should be
included in the media "fill and form" part of the final "worst-case" assessment.
Frequency is important if an intervention is carried out several times during a product fill; the intervention should be
undertaken at a similar rate during the media fill. It is additionally important that all personnel involved with the media trial
undertake at least one intervention.
CASE STUDY
For the case study, a multiproduct filling line was examined. The line was used for filling four different liquid products with only
vials supplied through a depyrogenation tunnel. The vial sizes were 20 mL, 50 mL, 100 mL, 250 mL, and 500 mL. In drawing
upon the above discussion, the following decisions were made:
Media fills were based upon the number of containers filled, rather than time.
The number of vials filled both reflects the typical product batch sizes and sets the maximum batch sizes for future
product fills (within an upper tolerance).
For the operation time, the time taken to fill a media trial is taken to be the equivalent time to fill a product fill of similar
size within normal operating conditions. Thus, by running a media fill for an equivalent time, the fatigue to which
operators are subjected to in a product fill is replicated during a media fill.

Certain parameters were considered to be of significance in terms of increasing or decreasing the contamination risk. These
related to the particular case study and, whilst certainly generalizable, should only be considered within the context of this
case study. The factors used for comparison follow.
Amount Filled in to Vial
This was considered to be a critical parameter as the amount in the vial directly affected the time at point of fill. Therefore, the
largest fill volume was weighted as an important factor. In addition, the smallest fill volume was also weighted as this related to
a faster running machine, and, as discussed above, the highest speed tended to generate more line stoppages.
Container Height, Size, Weight, and Diameter
These were each considered to be indicative of the stability of the vial. The heavier and taller the vial, the more stable it was
considered to be, and the less subject to minor "jolts" from the mechanical operation of the machine. These were considered
in relation to "worst-case" conditions.
Vial Neck Diameter
Vial neck diameter was considered to be the biggest microbiological risk factor; the vial with the largest mouth opening has the
highest risk of collecting microbial contamination from the environment.
Line Speed
The line speed was considered on the basis of the longest time at the point of fill, either through fill volume or due to the
automatic dosing stop (here the largest and smallest fill volumes and automatic dosing stop times coincided with the slowest
and fastest line speeds).
Operational Methods
The operation of each filling line was evaluated in relation to whether manual or automatic set-up was required, the number of
personnel interventions required during set-up, and the line set-up complexity. The outcomes of the review was that the same
level of complexity related to all currently filled products and that this factor was not required in relation to the time when the
case study was reviewed.
Maximum Tunnel Conveyor Speed
Whilst media fills are not designed to validate tunnel speed, this factor is worth considering on the basis that the fastest and
slowest tunnel speeds may affect the speed at which the line fills.
Maximum Fill Speed
This reflects the time in which the filling machine aims to fill the required dose. This is therefore a reflection of the time at point
of fill. For the matrix, the maximum fill speed was examined as an indicator of the time a vial remains at point of fill.
Automatic Dosing Stop
This parameter reflects the maximum time that a bottle or vial can remain at the point of fill whilst the machine fills a particular
fill volume. In addition to the vial neck size, this is arguably a critical risk as it reflects the time when a vial is exposed within the
most critical zone.
Closure Type
For this case study, all closure types were prepared and handled in the same way. This was not considered to be a critical
parameter.
The different factors were considered for the five vial sizes (covering the range 20 mL to 500 mL sizes) and four different
products (one to four) and compiled into a risk table (Table I). In the table, the "worst-case" condition for each factor has been
shaded. At the end of the table, the numbers of high-risk factors for each vial size have been totaled.
The assessment of "worst-case" parameters were further analyzed in a second table.
Tables I and II identified the 20 mL, 50 mL, and 500 mL vials as worst-case. Although 100 mL vial has the fastest speed

through the tunnel, the object of the media simulation is not to carry out a direct process validation of the tunnel. The
performance of the tunnel and its depyrogenation capability is assessed through other validation studies.
The 100 mL and 250 mL vials are also identified as having the largest vial neck diameter; however, the diameter of these vials
is the same as the 500 mL vials (32 mm). In addition, the 500 mL vial, with the largest neck diameter, has the longest
automatic dosing stop (20 seconds) and maximum fill speed (15 seconds); therefore, this combination is considered to be the
"worst-case" scenario.
Therefore, a rolling media fill schedule would consist of the 20 mL, 50 mL, and 500 mL sizes where the pattern is repeated
over an 18 month period (based on media fills being scheduled every six months).
In addition to the combination of risk factors that lead to the identification of the "worst-case" media fills, each media fill should
include a representative number of personnel interventions (with the personnel interventions selected on the basis of being
those that are undertaken most regularly together with those that are the most difficult to perform or present the greatest
contamination risk to the product).
Media Fill Failures
The filling line should be suspended where a media fill failure occurs, and an investigation should be undertaken with the aim
of determining a root cause. Areas to consider for the investigation include:
Microbiological environmental monitoring data
Particulate monitoring data
Personnel monitoring data
Sterilization cycles for media, equipment, and components and filters
High-efficiency particulate air (HEPA) filter evaluation
Area airflow patterns and pressures
Operator training and technique
Unusual events during the fill
Storage conditions of sterile items.
Identification of contaminant
Housekeeping procedures and training
Calibration and validation of sterilizing equipment
Pre and post-filter/housing integrity testing
Product/process defects and limitations of inspection process (24).
Once the investigation has been completed, a corrective action should be put in place, and the media fill should be repeated. It
is customary to repeat the media fill on three occasions. If the repeat media fills are successful, most likely a recommendation
that filling should resume will follow. If a media fill exhibits growth in any vial, filling should remain suspended and the
investigation should continue.
In relation to the matrix approach, for the vial combination that failed, it is good practice to revisit how often this combination is
performed in relation to the matrix (as this might now be the "worst-case," depending upon the investigation outcome) or to run
an additional media fill alongside the next periodic media fill defined by the predetermined matrix.

Table I: Consideration of the different media fill factors to evaluate worst-case combinations.

Table II: Assessment of worst-case parameters.

CONCLUSION

Media filling trials are an important component of the qualification of aseptic processing, and they provide a valuable indication
of the likelihood of product contamination (25). As such, careful thought should go into the design of a program and the
planning of media simulation trials. One key aspect of the program design is the types of fills to be conducted.
This paper has presented an overview of the important factors to consider when designing a media simulation trial program.
The paper has continued to demonstrate how these factors can be used to establish a risk-based matrix to decide what are
the "worst-case" combinations. It is noted that "worst-case" is, in part, a matter of opinion. However, given the range of
different products and product combinations filled on many filling lines, a choice needs to be made for which combinations will
be selected for media simulations. Furthermore, a decision must be made for how to answer regulatory questions regarding
why such combinations have been selected. This paper has shown how various risk factors can be selected, examined, and
then used to construct a risk-based matrix with which to select the most appropriate trials for a multiproduct filling line. This
approach was illustrated through the use of a case study. Each filling line will be different, and each facility will vary slightly in
terms of what is or what is not a greater risk. This paper has not set out to provide a "best practice" guideline but rather to form
a basis for how a regulatory question in relation to the "worst-case" media fill might be answered.
REFERENCES
1. B. Ljungqvist and B. Reinmuller, Clean Room Design: Minimizing Contamination Through Proper Design, Interpharm
Press, Buffalo Grove, IL, 1997.
2. J.P. Agalloco and J.E. Akers, "A Critical Look at Sterility Assurance," European Journal of Parenteral Science 7 (4), 97103, 2002.
3. R.T. Oji, "Regulatory aspects concerning the quality controls of microbiological and non-viable particulate
contamination in pharmaceutical manufacturing," American Pharmaceutical Review 7 (1), 18-22, 68, 2004.
4. E.C. Tidswell and B. McGarvey, "Quantitative risk modeling in aseptic manufacture," PDA Journal of Pharmaceutical
Science Technology 60 (5), 267-283, 2006.
5. K. Kawamura and H. Abe, "A Novel Approach to the Statistical Evaluation of Media Fill Test by the Different From No
Contamination Data," PDA Journal of Pharmaceutical Science Technology 58 (6), 309-320, 2004.
6. FDA, Guideline on Sterile Drug Products Produced by Aseptic Processing (Rockville, MD, Sept. 2004), available at:
http://www.fda.gov/downloads/Drugs/.../Guidances/ucm070342.pdf.
7. EU, "EU GMP Annex 1: Manufacture of Sterile Medicinal Products," European Union GMPs 4, European Commission,
Nov. 25, 2009 (Brussels, Belgium).
8. R. Friedman and S. Mahoney., "Risk factors in aseptic processing," American Pharmaceutical Review 6 (1), 44-46, 92,
2003.
9. R. Friedman, "Routes of contamination: aseptic processing case studies," European Journal of Parenteral &
Pharmaceutical Sciences 10 (1), 3-7, 2005.
10. H. Aleem, Y. Zhao, S. Lord, T. McCarthy, and P. Sharratt, "Pharmaceutical process validation: an overview,"
Journal of Process Mechanical Engineering, 141-151, 2003.
11. S.I. Haider, Pharmaceutical Master Validation Plan: The Ultimate Guide to FDA, GMP, and GLP Compliance, CRC
Press LLC, Boca Raton, Florida, 2001.
12. J. Lambert, Validation Guidelines For Pharmaceutical Dosage Forms, Health Canada/Health Products and Food
Branch Inspectorate, 7-15, 2004.
13. R.A. Nash and A.H. Wachter, "Harmonization, GMPs, and Validation," in Pharmaceutical Process Validation An
International (Drugs and the Pharmaceutical Sciences) Revised and Expanded (Marcel Dekkar, Inc., New York, 129,
3rd ed., 2003), 760-792.
14. EMA, "Guideline on Process Validation," EMA/CHMP/CVMP/QWP/70278/2012, European Medicines Agency, March
29, 2012 (Brussels, Belgium).
15. FDA, FDA Guidance for Industry, Process Validation: General Principles and Practices (Rockville, MD, January 2011)
available at: http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformati....
16. D.W. Vincent, "Regulatory and Validation Considerations for Aseptic Processes," Journal of Validation Technology 12
(2), 107-133, 2006.
17. N.A Halls, Microbiological Media Fills Explained, West Sussex: Sue Horwood Publishing Ltd., 2002.
18. M. Budini and F. Boschi, "Aseptic Process Simulations/Media Fills," in Microbiology and Sterility Assurance in
Pharmaceuticals and Medical Devices, M.R. Saghee, T. Sandle, and E.C. Tidswel (Eds.), (Business Horizons, New
Delhi, India, 2011), pp. 701-732.
19. M. Bernuzzi, N.A. Halls, and P. Raggi, "Application of statistical models to action limits for media fill trials," European
Journal of Parenteral Sciences

2, 3-11, 1997.
20. PIC/S, Guide: Validation of Aseptic Processes PI 007-5, published jointly by the Pharmaceutical Inspection Convention
(PIC) and the Pharmaceutical Inspection Co-operation Scheme (Brussels, Belgium, 2011).
21. T.E. Munson, "Basics of an Aseptic Process Simulation Program," Pharmaceutical Technology 28 (5), 35-39, 2004.
22. J.P. Agalloco and J. Akers, "Revisiting Interventions in Aseptic Processing," Pharmaceutical Technology 35 (4), 69-72,
2011.
23. S.K. Doshna, "Media Fills 101: How to Design, Conduct and Document Media Fills for Aseptic Processes,"
American Pharmaceutical Review 6 (4), 52-56, 2004.
24. J.E. Snyder, "Corrective and preventive action planning to achieve sustainable GMP compliance," Journal of GXP
Compliance 6 (3), 29-39, 2002.
25. H. Van Doorne, B.J. Van Kampen, R.W. Van der Lee, L. Rummenie, A.J. Van der Veen, & W.J. De Vries, "Industrial
manufacture of parenteral products in The Netherlands. A survey of eight years of media fills and sterility testing,"
PDA Journal of Pharmaceutical Science Technology 52 (4), 159-164, 1998. JVT

Source URL: http://www.ivtnetwork.com/article/risk-matrix-approach-media-simulation-trials