Awarded the Journal of GXP Compliance

2012 Article of the Year

MICRO BIO LO GY TO PIC S

Contamination Control in
the Compliance Program
Scott Sutton

Microbiology Topics discusses various topics in microbiology of practi-

cal use in validation and compliance. We intend this column to be a useful
resource for daily work applications.
Reader comments, questions, and suggestions are needed to help us fulfill
MEDICALRF.COM/GETTY IMAGES

our objective for this column. Please send your comments and suggestions
to column coordinator Scott Sutton at scott.sutton@microbiol.org or journal
managing editor Susan Haigney at shaigney@advanstar.com.

IMPORTANT
PORTA POINTS
Thee follow
following impor
important point
points are discus
discussed:
MEDIC

t
The

The contamination
co mina control plann is central
cen to goo
good manufacturing
an cturing
IMAGE SOURCE,

practice, describing
actic , des ib g the
th procedures
procedu es andand policies
poli s designed
s to create
products
p odu s ununder controlled
der contro led conditions
ond on
O

t
The contamination control planp covers all p

phases of a facilitys
y
IM

status
tatus
a s
tt
The contamination
nt control plan
lan iss a comprehensive document,
describing control of incoming contamination and measures
designed to minimize their impact
t
Consistency and application are key to successful execution of the
contamination control plan.

INTRODUCTION
Development of a contamination control plan is critical to the success of
aseptic, terminal sterilization, and non-sterile manufacturing facilities. This
is most obvious in the aseptic arena, where the US Food and Drug Admin-
istration has issued clear regulatory guidance on the need for control of
contamination at all stages of the process. What is less obvious is the even
greater need for a plan to address contamination control in non-sterile man-
ufacturing. This need only becomes obvious after a problem has arisen
frequently a problem in contamination control that requires product recall.
The contamination challenge in non-sterile production is different from that
in aseptic production. The objective in aseptic production is to exclude all
microorganisms from the finished product; in the non-sterile environment
it is to control the types and numbers of microorganisms in the finished

58 Journal of GXP Compliance

 Scott Sutton

Figure 1:
Example Fishbone diagram for contaminated product.

Cause & Effect Diagram for Potential Causes of Contaminated Product

Personnel
Training
Raw Materials
SOP not
Inadequate Hand Correct No Evidence Proficiency
Raw Material/Components Sanitization Not Reviewed
of Training
Received Contaminated SOP Program Regularly
Raw Materials/Components Poor Personal Hygiene Operators Ineffectively
Contaminated on Sampling (piercings, jewelry, SOP not Relevant SOP Trained
makeup, illness) Followed not A3&))"
Raw Materials/Components Inadequate PPE
Contaminated on Storage Training Program Proficiency not
(foot, coat, face, hair) not in place Demonstrated

Contaminated Product
Sample Hold Compromised
Inappropriate Materials of
Construction (difficult to sanitize, Inadequate Sampling Error
Pest Control Inadequate Removal
particulate generation, etc.) ,#
&0&)"
"/&0

Organic Load Transfer/Sampling
Lack of Air
Ineffective Issues
Quality Control
Sanitization Agents
Air Quality
Compromised + ,*-1&&)&15
,#
Sample Transport Compromised
Cleaning-Sanitation
Compressed Agents Shift of Environmental
Gases
&,2/!"+
#/,*

H
HVAC-HEPA Compromised Compromised Product Contact Sensitive to Hold times not validated
Resistant Species (process as well as
Water Generation Surfaces
equipment clean
System Compromised ty h
and dirty holds)
ot-to-Lot Cleaning
Lot-to-Lot C g Ineffective
Ine ve
Delivery (Hoses)
Water Delivery (H //53"r
9
//53"r
Comp promised
Compromised "0&!2)0
9
"0&!2)0

&,2/!"n
9

&,2/!"n qu
Inadequate ration
Separation
W
Water Q
Quality n-Prod
Non-Product tact
Contact off C
Clean and
Comp
Compromised SSurfaces Contamin
Contamination of rty Operations
Dirty atio
Cleaning Implements
P
Piping Issues de
Inadequate Protection
(d
(dead-legs, reverse om Uncontrolled
From
inclines, etc.) Source
e Water
W O er Water
Other Wa r Issues Cleaning
Clean and
ng a Sanitization
d Sa itiza n Env
Environment
Compromised 9 +!&+$ Wa
9
1+!&+$
Water
9
/&--&+$
&-"0
Facility
Fac
acili y
products. Regulatory action has extended this consid-
eration even into the realm of personal care products
(e.g., cosmetics, toiletries, and soaps).
Contamination control is central to compliance. The
whole point of current good manufacturing practice
(CGMP) is to produce safe and effective medications
(1). As an indication, the word contamination or its
variants appears no fewer than 24 times in 21 CFR
211. One of the most frequent cause of drug recalls,
lack of sterility assurance, relates directly to the inabil-
ity of the manufacturer to document adequate protec-
tion against contamination by adventitious bioburden.
See the Reference section for a historical review (2) and
a biotech case study on this topic (3).
Diverse markets such as pharmaceuticals, medical
devices, diagnostics, and personal care products have
operated historically under different CGMP. However,
Process
the considerations for contamination control are simi-
lar and can be approached from the perspective of root
cause analysis (4). Use of a Cause and Effect diagram
can be an excellent tool in the determination of likely
routes of contamination after the fact during an inves-
tigation. It can also be used as a proactive learning tool
for the development of the contamination control plan
(see Figure 1). For those more comfortable with Six
Sigma procedures, this can be revisited as an Ishikawa
Diagram (see Figure 2). The main point here is to
identify, and come to agreement, on the likely causes of
potential problems.
A more traditional proactive approach to risk
management might be through use of failure mode
and effects analysis (FMEA), which can be extremely
useful in determining the most important aspects of
control for your process (5).
Autumn 2011 Volume 15 Number 4 59
MICROBIOLO GY TO PIC S

Figure 2:
Example Ishikawa diagram for contaminated product.

Cause & Ef$#!2
'%0+
$-0
-2#,2'*
31#1
-$
-,2+',2#"
0-"3!2
: 

Manpower
Methods 
,-2
Materials 0-8!'#,!6

,-2

Followed
Not Reviewed Correct
Inadequate Hand No Evidence Regularly
Raw Material/Components Sanitization of Training
Received Contaminated 
0-%0m
Raw Materials/Components --0
#01-,*
6%'#,#
Operators
Contaminated on Sampling (piercings, jewelry, Ineffectively #*#4,2


makeup, illness) Trained not A4'* *# '*32'-,1
 
Raw Materials/Components ,"#/32#

Tr',',%
0-%0+

Contaminated on Storage 0-8!'#,!6
,-2 0-!#"30#
(foot, coat, face, hair) #+-,1202#" not in place
Inappropriate to Specs
Unvalidated
-,2+',2#"
0-"3!299
Sample Hold Compromised
Inappropriate Materials of
Construction (difficult to sanitize, Inadequate Sampling Error
#12
-,20-l Inadequate Removal
particulate generation, etc.) -$
'1' *#
# 0'1

Organic Load Transfer/Sampling
Lack of Air
Ineffective Issues
Quality Control
Sanitization Agents
Air Quality
Compromised ,!-+.2' '*'26
-$ Sample Transport Compromised
Cleaning-Sanitation
Compressed Agents Shift of Environmental
Gases '- 30"#,
$0-+

HV
A Compromised Compromised Sensitive to Hold times not validated
0-"3!2
-,2!2

Resistant Species (process as well as
Water Generation Surfaces
equipment clean
System Compromised and dirty holds)
Lot-to-Lot Cleaning Ineffective
W2#0
#*'4#06
-1#1

#*'4 ;
0064#r
Compromised
mpro ;
#1'"3*1
;
#1'"3
;
'- 30"#n
30"# Inadequate Separation
equate Sepa
Water Quality -,0-"3!2
-,2!2
-,0-"3 !2 of Clean
Clea and
Compromised
Compro Surfaces
Sur Contamination
Contaminat of : '026<
.#02'-,1
<

Cleaning Implements
Imp
'.',%
113#1

'. ,"#/32#
0-2#!2'-,

 #
0 ,
(dead-legs,
(de reversee From Uncontrolled
inclines, etc.) Source Water Other Water Issues Management Environment
Compromised
ompro se 2,"  er
;
2,"',%
ater
;
0'..',%
'.#1
0'..' '.#1
Machinery
Mach ery
Mother
h Nature
N t

COMPONENTS
OMPONENTS OF A STRONG G At a minimum,
in these include the following:
CONTAMINATION CONTROL PLAN t
Commissioning and initial start up
The contamination control plan and the protocol t
Ongoing operations
governing the program are essential documents t
Shut-down for regular maintenance
useful in providing the rationale and methods used t
Start-up after scheduled shut down.
to accomplish three tasks:
t
Minimizing the bioburden throughout the These phases will not have the same level of
manufacturing processes contamination control. A good plan will discuss the
t
Minimizing the level of batch residual cross- concerns specific to each of these phases.
over contamination
t
Minimizing the level of cleaning material resid- Validated Methods
ual contamination. All measures of bioburden in a facility will be
This article examines the components of the indirect. Bacterial cells cannot be counted on a
plan in terms of specific areas of interest in meeting surface or in the air. The microorganisms must be
these three key objectives. transferred to an agar plate (or some other mecha-
nism) and colony-forming units (CFU) counted. If
Phases of Manufacturing Operation the assumption is made that the transfer of micro-
The contamination control plan must take into ac- organisms from the air or from a surface to agar
count different stages of facility operational status. is consistent, then these numbers can be used to

60 Journal of GXP Compliance

 Scott Sutton

estimate trends over time. This assumes that the ing fiscal responsibility. The FDA aseptic processing
nutrient agar is capable of growing the microorgan- guidance document recommends genetic identi-
isms to visible colonies. As residual disinfectant on fication of all organisms isolated from the manu-
a surface may impede the growth of microorgan- facturing environment on a regular basis. This is a
isms, neutralizers are frequently incorporated into laudable goal, but few of us have anything near the
the growth media (e.g., Dey-Engley Agar, MCTA). required budget to accomplish this task. In all hon-
The contamination control plan should describe the esty, it is reasonable to wonder if the effort is really
sampling methods used, and how these methods necessary. The numbers of CFU from validated sites
are to be validated for the conditions of use (6). (i.e., viable air and surface, non-viable) is sufficient
The facility should be disinfected regularly using to provide a measure of the state of control of the
validated sanitizers and sporicides. The contamina- facility. However, periodic cataloging of the resident
tion control plan should describe the methods for microflora (at least to the species level) will provide
testing and rationale for acceptance of materials to a good check on the continued effectiveness of the
be used in the ongoing program of disinfection. The disinfectants in use (7). Shifts of bioburden to spore
plan should ideally describe the in vitro or labora- forming microorganisms will be strong evidence of
tory tests to evaluate the sanitizers, including the the need for use of a sporicidal agent. Occasionally,
identification of the most resistant microorganisms this effort will also pick up shifts among non-spore-
found in the h facility as well as the most difficult-to- forming organismsthis is not due to resistance
disinfect materials
m in the facility. This is also where but rather ecological shifts towards species more
the method hod for on-going
-goi evaluation
aluation of the san sanitiz- naturally
rally resistant
re ant to thee disinf
disinfectant
nt iin use.
erss base
based ed on environmental
ronm monitoring
monito g data will The
he considerations
cons ations in thee non-sterile
sterile facility
ity
be rec
recorded.
corde Thee choice of disinfec
disinfection regimens
egim s are
a similar, as we need nee proofof thatt the sanitization
san ation
shouldu be reevaluated
uld uated annually,
an al and the ccontamina-nt mina program
progr m u used
e is eeffective
ct e and the regular use of
tionn control plan should
n sh uld describe
escr be hhoww th thiss eva
evalua-
ua sporicidal
poric a agents
g nts appropriate.
appro riat The contamination
tion will occur. control plan
p must address how the sanitization
Itt iiss alsoo critical
cr ticaal too have ropriaattee microbio-
av appropriate
appro microbio
microb bio- program
program w
prog willl b
bee m
monitored
on
nito
ored efficacy.
d for ef
fficaacyy.
logical
ogical methods for the relevant v specifications
cif
plate counts less than 20 CFU are notoriously Control Incoming Bioburden
unreliable. Traditionally the countable range for The first step in any control program is to control
a standard sized petri dish is 25-250 (or 30-300) contamination at the very beginning of the process.
CFU/plate. These lower limits (limits of quantifica- This includes raw materials (e.g., excipients, API,
tion) are routinely ignored when setting product water) and the primary containers. All materials
and environmental specifications, thereby sacrific- should be tested for incoming bioburden against
ing accuracy and precision. Similarly, the upper documented acceptance criteria. Part of the incom-
limits (250 or 300 CFU/plate) are set for particular ing bioburden will also include water used as an
microorganisms on standard-sized plates, smaller excipient to the process. The GMPs for pharmaceu-
plates, or organisms that produce larger colonies ticals, biologics, medical devices, and diagnostics
will have lower numbers as their upper limit. These all provide instruction on this point.
will require dilutions as appropriate to measure the
same specification. Appropriate Gowning
The gowning methods and materials are of critical
Know the Enemy importance to minimization of contamination. The
A successful contamination control plan is intended primary source of contamination in most controlled
to provide the most useful information on the mi- areas will be the personnel. The contamination
croorganisms present while at the same time show- control plan must address whether it is designed for

Autumn 2011 Volume 15 Number 4 61

MICROBIOLO GY TO PIC S

aseptic gowning procedures or protection of personnel Validated Sanitization and

in non-sterile manufacturing facilities. All personnel Sporicidal Procedures
should be well trained in appropriate gowning practice The efficacy of the sanitizers and sporicides used
and behavior. The contamination control plan should in the program must be demonstrated in a study
describe the rationale for the level of gowning chosen, designed to test their efficacy on the materials of
the frequency of gown cleaning, personnel behavior, construction and against resident microorganisms
and the acceptable gown materials for the type of found in the facilities governed by the contamina-
manufacturing process. tion control plan. This can be done optimally in the
following four-step process:
Training 1. Suspension test of efficacy. This is a screen-
Operator training is critical to contamination con- ing effort using your candidate agents against
trol. No supervisor can be present at all locations at lab strains of indicator organisms as well as a
all times. Each operator must be aware of his or her variety of the microorganism species found in
role in contamination control and how to minimize your facility. The goal of this assay is to deter-
the risk to batch integrity. Minimal skill sets in mine the most resistant microorganism(s) for
relation to product protection and sampling should the next step.
be described by job function in the contamination 2. Coupon study. Using the representative organ-
control plan.
an This is most effectively accomplished isms (e.g., gram positive, gram negative, spore
using a wel
well-designed and enforced standard oper- former, yeast, and mold) and organism(s) iden-
atingg p
procedure
oce (SOP)
OP) systemm (8). tified in
i thee previous
prev us study,
stud testst the
t efficacycacy of
o
the san
sanitizing
ing agents
ag on coupons
pons ofo materials
terials
Controlled
ntroolled Environments
ronment found in the facility. The
he faci Th purpose
urpose oof thisis test iis
Control
nttro and monitoring
nito ing of th
the envi
environment
onm t iis an-
n to demonstrate
t dem mo stra eefficacyca on these materials us-
other
he critical element
men of the
the contamination
contam nation control
control ing
i g thee appropriate
ppropriat application
app procedures.
plan. Large portions of this can be addressed byy the
g p 3. Mock sanitization study. y This studyy
corporate
poratte environmental
orporate
rp enviroonmentalal monitoring
monito
tor ngg (EM)
(EM
M master
mast
masster provides rreal-world
prroviidess reeal-w
world evevidence
vviidence of efficacy.
a y. Let
effficcacy Leet
plan
lan (which provides rationale
na and consistency
o for a re
representative room go untouched for a
a single EM philosophy across the different facilities period of time to become contaminated.
of the corporation) or the site environmental master Take bioburden samples throughout the room,
plan (which provides consistency and detailed and then sanitize the surfaces and repeat the
instruction for the various manufacturing build- bioburden sampling. The samples taken after
ings at a given site). The contamination control plan cleaning should be far less contaminated than
should cite the relevant documents and their role in the first set.
contamination control. 4. Confirm efficacy from environmental moni-
The appropriate EM plan for non-sterile manufac- toring. The final step in validating the saniti-
tures and for active pharmaceutical ingredient (API) zation program will be ongoing evidence that
manufacturers is not well defined from a regulatory the program allows the facility to operate in a
perspective. There are no strong recommendations state of control.
such as those seen for the environmental monitor-
ing of aseptic facilities; however, the absence of The sanitization program will ideally consist of
regulatory guidance is not the same thing as the a qualified disinfectant and sanitizing agent, used
absence of need for the activity. EM is useful for appropriately (concentration and contact dwell time
determining the state of control of the facility and observed) as the primary agent. This will then be
so is an important part of the monitoring program periodically rotated with the use of a sporicidal
for all manufacturers (6). agent (also validated for effectiveness) (9, 10).

62 Journal of GXP Compliance

 Scott Sutton

Well-Defined and Understood manufactured at a site. This concern has little to do

Manufacturing Processes with the sterility of the finished product and is rel-
The manufacturing process should be evaluated evant to sterile and non-sterile manufacture alike.
for its potential to limit or eliminate bioburden.
The two common methods for performing this are Minimization of Cleaning
either a HACCP-type or a FMEA approach. The Material Residual Contamination
use of organic solvents, heat, or other inhospitable Validation of cleaning procedures is essential to
activities can greatly reduce bioburden of a process. demonstrate not only that the cleaning procedure
For example, the contribution of compression (and effectively cleans and sanitizes the manufacturing
associated shear) should be evaluated for a potential equipment, but also that residual cleaning material
reduction in risk of excessive microbial contamina- is removed to prevent contamination of the next
tion. The contribution of the finished product water batch manufactured.
activity should also contribute to this analysis. Pro-
cess bioburden reduction steps should be factored DANGER SIGNS IN CONTAMINATION CONTROL
into the contamination control plan on a process- Compliance professionals must be vigilant in terms
by-process basis. of contamination control. Contamination control is
Of particular importance in this evaluation for a sure area of regulatory enforcement interest and
the potential ia for microbial contamination of the one where the short-sighted might be tempted to
process are re cleaning steps, equipment hold times, institute cost-savings. Frequent manifestations of
HVAC,, con control levelel oof environments
ronmen for critical
critic cost savings
saving mentality
ntalit deal
eal with
wit validation
lida as-
tasks, open-system
sks, open-s
o m versus
vers closed-system
losed-sy m opera- ope pectsthe
sthe sanitizer
itizer aand cleaning
lea qualifications
quali ions in
tions, bioburden
ns, biob monitoring,
n monito g, and oothers spespecificc particular
p icular canc be expensive.
exp e. Thee temptat
temptation is real
to theh p
he process. Ass an exa example
pl off th
the imp
importance
ta e of tto pull
u bac backk from
om a fu
full pr
program. Some examples
thee bioburden controlontr l po point issue,
nt is ue, tthere
here iis a ststrong
ong off these
the aapproaches include
pr aches inc ude accepting without con-
regulatory
g y expectation
p in Europep that p products firmation bioburden information on incomingg raw
sterilized
terilized
err ilized
i d by filtration
f ltraationn should
sho
hoould have
haave a pre-filtration
pre-filtratio
pree-f ltrration
n materials,
m
matatterriials, assuming
a su
su
umminng efficacy
efficacccyy of sanitizers,
san
nitizzers,, neglect-
neeglecct
bioburden
ioburden of not more than n 10 CFU/100 /1 mL imme- ing tot cl clean floors and other non-product contact
diately before the sterilizing filter (or be subjected surfaces (especially in non-sterile production), and
to dual filtration in series). other relatively high-cost overhead activities.
Finally the contamination control plan should These are not optional safeguards to be ignored
cite the need for clear SOPs on all aspects of manu- with impunity, as several companies have recently
facturing, monitoring, and control. These SOPs come to realize (10-15).
are critical for training, documentation, and batch A second major area to be constantly examining
release. is in the general topic of consistency. A well-con-
structed contamination control plan will include
Minimization of Batch Residual reasons for various activities. For example, if you
Crossover Contamination require facility scrubs in a non-sterile production,
The contamination control plan should also address there is a reason for this that should be described.
the potential for a batch to be contaminated by Requiring all operators to be in scrubs and boo-
material from the previous batch manufactured us- ties is not consistent with allowing maintenance
ing the same equipment. The contamination control personnel to enter the area in boots, denims, and
plan should describe the methods by which this flannel shirt for a quick signature on a work order.
likelihood is minimized. Likewise, requiring shoe covers (booties) to protect
The concern over batch residual crossover is against tracking soil organisms through the facility
most relevant when there is more than one product is not effective if operators fail to change the boo-

Autumn 2011 Volume 15 Number 4 63

MICROBIOLO GY TO PIC S

ties during the day (i.e., entering, leaving, going to 13. Macdonald, G., Tabletting delay caused lethal allopurinol
lunch, using the restroom, walking in them until contamination, say HK chiefs, Decision News Media, 2009.
they hang in tatters from their ankles). Consistency http://www.in-pharmatechnologist.com/content/view/
is key. print/239189
14. Yang, C., et al., Nosocomial Outbreak of Two Strains of
REFERENCES Burkholderia cepacia Caused by Contaminated Heparin
1. Oji, RT, Regulatory Aspects Concerning the Quality (Letters to the Editor), Hosp Infect Soc. 69(4):398 400,
Controls of Microbiological and Non-viable Particulate 2008.
Contamination in Pharmaceutical Manufacturing, Amer 15. FDA. FDA acts to prevent contamination problems with
Pharm Rev. 7(1):18,20,22,68, 2004. Triad antiseptic products, 2011. http://www.fda.gov/News-
2. Prince, H. and Prince, D., Contamination Control: The Life Events/Newsroom/PressAnnouncements/ucm249706.htm
and Death of Bacteria and Other Germs, Contract Pharma, GXP
pp.2 7, June 2005.
3. Schleh, M., et al., Effectiveness of Upstream Barrier Tech- ARTICLE ACRONYM LISTING
nologies for Inactivation of Adventitious Contaminants of API Active Pharmaceutical Ingredient
Cell Culture, Amer Pharm Rev 13(7):72-76, 2010. CFU Colony Forming Units
4. Sartain, E. & J. Polarine, Prevention of Microbial Contami- CGMP Current Good Manufacturing Practice
nation, Pharm
P Technol. 35(6):62-65, 2011. EM Environmental Monitoring
5. Sutton, S.,
S Successful Microbiological Investigations, Amer FDA US Food and Drug Administration
Pharm Rev. 74(2):34-42,
m Re 34-4 2011.. FMEAA Failure Mode Effects
Fa Analysis
fects A sis
6. Sutton,
Suttoon, S., The
Environmental
Environ al Monitoring
Monito Program In a GMPss Good Manufacturing
Go Manufa ng Practices
tices
GMP
GMMP Environment,
En men J GX ompliance 14(3):22-30,
GXP Compliance, :22- 2010.
010. MCTA
M A Microbial
Mi Content Te
ial Con Test Agar (Soybean
gar (Soyb Casein
n Casei
7. Sutton,
Suut SVW and d AM
A Cundell,
Cun l, Microbial
icrob al Identification
Ide ic n in Digest
D ge AgAgar wwith Lecithin)
the Pharmaceuticalical Industry,
ndust y, Pharm F rum. 30(5):1884-
P rm Forum. 0(5): 884- SOP
OP Standard
S Operating
an ard Opera ing Procedure
1894,, 2004.
8. Sutton,
Sutton, S., Thee Importance
Su nc of a Strong
Impoortance St SOP
SOP System
Sy temm in thee ABOUT
AB
A BO
B OUTT THE
OU THE AUTHOR
TH AUTH
U HO R
QC Microbiology Lab, J GXP C Compliance. 14(2):44-52, Scott
ott Sutton
tto is the
h founder
d and principal consultant
l off Micro-
2010. biology Network, Inc (http://bit.ly/n62yFG). He is a recognized
consultant and trainer with emphasis in GMP, investigations,
9. Martinez, J., The Rotation of Disinfectants Principle: True
environmental monitoring, and contamination control as
or False? Pharm Technol. 33(2):58 71, 2009. well as microbiology laboratory audits and operations. He
10. Sutton, SVW., Disinfectant Rotation-A Microbiologists may be reached at scott.sutton@microbiol.org or by phone at
View, Controlled Environ., pp. 9-14, July 2005. 1.585.594.8273.
11. Donald, T., Vaccines Pulled Due to Potential Contamina-
tion, Contamination Control. 4(1):12 13, 2008.
12. Suvarna, K., et al., Case Studies of Microbial Contamina-
tion in Biologic Product Manufacturing, Amer Pharm Rev.
14(1):50-56, 2011.

64 Journal of GXP Compliance