Content uploaded by Pascal Bonnabry
Author content
All content in this area was uploaded by Pascal Bonnabry on Jan 04, 2014
Content may be subject to copyright.
PRACTICE REPORT Microbial contamination
2032 Am J Health-Syst Pharm—Vol 66 Nov 15, 2009
Microbial contamination of syringes
during preparation: The direct influence
of environmental cleanliness and risk manipulations
on end-product quality
Cy r i l St u C k i , An n A -MA r i A SA u t t e r , Jo C e ly n e FA v e t , A n d PA S C A l Bo n n A B r y
Cyril StuCki, Pharm.D., is Hospital Pharmacist; an n a -maria
Sa u t t e r , Ph.D., is Hospital Pharmacist, Pharmacy, University
Hospitals of Geneva, Geneva, Switzerland; J
o C e l y n e Fa v e t , Ph.D.,
is Microbiologist and Teacher, Biology Department; a n D PaSCal
Bo n n a B r y , Ph.D., is Professor, Pharmacy Department, University of
Geneva, Geneva, Swtizerland.
Address correspondence to Dr. Stucki at the Pharmacy, Univer-
sity Hospitals of Geneva, Pharmácie des HUG, Rue Gabríelle-Perret-
Gentil 4, CH-1211 Geneva 14, Switzerland (cyril.stucki@hcuge.ch).
The authors have declared no potential conflicts of interest.
Copyright © 2009, American Society of Health-System Pharma-
cists, Inc. All rights reserved. 1079-2082/09/1102-2032$06.00.
DOI 10.2146/ajhp070681
Purpose. The direct influence of environ-
mental cleanliness and risk manipulations
on prepared syringes was evaluated.
Methods. Media-fill testing was used to es-
timate potential microbial contamination.
Syringes were prepared in three different
environments using four different uncon-
trolled high-risk manipulations. The three
environments included an International
Organization for Standardization (ISO) class
5 horizontal laminar-airflow hood in an
ISO class 6 cleanroom (in accordance with
United States Pharmacopeia [USP] chapter
797), an ISO class 7 drug preparation area
of an operating room, and an uncontrolled
decentralized pharmacy in a ward. For
each combination of environment and
manipulation, 100 syringes were filled by
a single operator. The four high-risk ma-
nipulations used included simple filling of
syringes with trypticase soy broth, three-
second contact by the ungloved fingers of
the operator with the hub of the syringe,
three-second contact between an object
and the hub of the syringe, and exposure
of the filled syringes to ambient air for 10
minutes.
Results. Of the 1500 syringes prepared in
three different environments, none pro-
duced within the cleanroom contained
microorganisms, 6% were contaminated
in the operating room, and 16% were
contaminated in the ward (p < 0.0001).
Certain high-risk manipulations were as-
sociated with a significant increase in the
contamination of the surrogate syringes,
including exposure to nonsterile ambient
air and nonsterile objects or fingers (p <
0.0001).
Conclusion. High contamination rates
were measured when the hub of syringes
touched nonsterile environmental sur-
faces and fingers, whereas the drawn-air
manipulation was associated with a lower
risk of contamination. Working within a
properly operating unidirectional airflow
primary engineering control in an ISO class
5 cleanroom in accordance with USP chap-
ter 797 requirements was demonstrated
to be the best way to avoid bacterial or
fungal contamination of injectable drugs
directly resulting in patient infections.
Index terms: Aseptic areas; Compound-
ing; Contamination; Injections; Laminar
flow; Pharmacy, hospital; Standards;
Syringes
Am J Health-Syst Pharm. 2009; 66:2032-6
The preparation of sterile inject-
able preparations frequently
necessitates a drug transfer or
dilution step from a vial to the final
syringe or bag. When the prepara-
tion is not performed in an aseptic
environment, it is recommended that
the drug be administered rapidly af-
ter reconstitution to avoid microbial
contamination.1 However, in some
situations, drugs are prepared ahead
of time and stored until they are
needed. For instance, anesthesiolo-
gists often prepare syringes contain-
ing anesthesia-induction agents,
neuromuscular blockers, and resus-
citative drugs before they are used in
surgical procedures.
The practice of storing drugs in
hospital-filled syringes raises ques-
tions related to drug safety (i.e.,
might such storage adversely affect
drug potency and allow for micro-
bial growth of any contaminants
introduced during preparation?). It
is well-known that the contamina-
tion of syringes may increase the risk
P R A C t i C e R e P O R t
PRACTICE REPORT Microbial contamination
2033
Am J Health-Syst Pharm—Vol 66 Nov 15, 2009
of infection, and several serious cases
of such infection have been reported
in the literature.2-13 To minimize the
risk of end-product contamination,
United States Pharmacopeia (USP)
chapter 797 requirements limit the
storage of drugs in prefilled syringes
to a one-hour period.14
The two main factors that con-
tribute to microbial contamination
of drugs are the cleanliness of the
work environment and the com-
petency and care of the operator;
however, no published data have
quantified their respective impact.
There is growing awareness that
when proper procedures are fol-
lowed, the provision of ready-to-use
syringes can reduce the risk of dilu-
tion errors14-18 and that reduction of
the contamination risks associated
with the preparation of these drugs
by competent personnel in clean-
rooms, as an alternative to ward
environments, is needed.19
During drug preparation, a va-
riety of improper manipulations
may compromise sterility, resulting
in potential contamination of the
end-product. Although both com-
monsense and operator training in
aseptic technique for compounded
sterile preparations (CSPs) clearly
recommend against careless or non-
standard manipulations, such prac-
tices do occur. In a previous study, we
collected and tested unused syringes
containing CSPs in our operating
rooms and observed a 0.5% rate of
contamination.20 These data sug-
gested that two syringes prepared in
our hospital could be contaminated
every day. This led our team to more
thoroughly investigate the sources
of potential contamination. The
objective of this study was to esti-
mate the probability of microbial
contamination of syringes during
preparation by simulating syringe-
filling operations in three common
hospital environments.
Methods
Media-fill testing was used to
estimate potential microbial con-
tamination. A surrogate media-fill
challenge incorporating a sterile
trypticase soy brotha (TSB) from
a 100-mL vial was used to fill the
test syringes.b
The TSB was received
with the manufacturer’s certificate
of analysis, and the product’s growth
promotion and inhibition in closed
syringes were validated in accordance
with USP chapter 71 by inoculating
the TSB with 102 colony-forming
units (CFU)/mL of Pseudomonas
aeruginosa,c Staphylococcus aureus,d
Bacillus subtilis (spores),e Candida
albicans,f and Aspergillus niger.g Suc-
cessful positive controls were those
that demonstrated positive growth
of the test inoculum in the TSB
within three days (for bacteria) or
five days (for fungi). To establish
positive growth, each test syringe
was compared to control samples.
Transfer from a vial to syringes was
simulated by substituting TSB for a
drug. In this exercise, 5 mL of TSB
was withdrawn from a 100-mL vial
and injected into a 10-mL syringe.
Five milliliters of filtered air was
aseptically added to each syringe
through a sterile 0.22-mm air filter
in order to support the potential
growth of aerobic contaminants.
The syringes were then sealed with
a Luer-Lok cap. In accordance with
USP chapter 797, these syringes were
incubated for seven days at a mean ±
S.D. temperature of 25 ± 1 °C and for
seven more days at 32 ± 0.2 °C.
After 14 days, all samples were
analyzed by direct examination
with incandescent electric light, and
positive contamination was declared
when any turbidity of the growth
medium was observed. Syringes were
prepared in three different environ-
ments employing four different un-
controlled high-risk manipulations.
For each combination of environ-
ment and manipulation, 100 syringes
were filled by a single operator. This
operator received training on aseptic
preparation techniques validated
through individual media-fill tests.
The three testing environments
were (1) an International Organi-
zation for Standardization (ISO)
class 5 horizontal laminar-airflow
hood in an ISO class 6 cleanroom,
(2) an ISO class 7 drug prepara-
tion area of an operating room, and
(3) an uncontrolled decentralized
pharmacy in a ward. Air particulate
contamination of the three sites was
determined with a discrete particle
counter (DPC).h This DPC is annu-
ally calibrated by the manufacturer in
accordance with international stan-
dards. The DPC was employed to es-
timate the total airborne particulate
contamination burden as ≥0.5- and
≥5.0-mm particles per cubic meter.
Concomitant aerobiological testing
was not conducted.
Four common high-risk ma-
nipulations were investigated in this
study, and the results were correlated
with a simple filling of syringes with
TSB. This exercise was considered
the baseline contamination risk.
This operation consisted of asepti-
cally withdrawing 5 mL of TSB from
a 100-mL vial with a sterile 10-mL
syringe and a sterile spike. One-
hundred syringes were filled in this
manner for each of the environments
and conditions tested. A total of 1500
surrogate syringes were produced by
a single operator to limit variability
due to technique. During the first
high-risk manipulation, the TSB was
aspirated into the syringe and left
uncovered, exposing the content to
air. This manipulation is often ob-
served in a ward or operating room
to adjust the volume of a drug in or
to remove air bubbles from a syringe.
After filling the syringe with TSB, 5
mL of air was drawn into the syringe
three times using the plunger. The
second exercise evaluated the risk of
contamination after three seconds of
contact by the operator’s ungloved
fingers with the hub of the syringe.
This manipulation clearly compro-
mises sterility; however, this kind of
manipulation can accidentally occur.
The third exercise simulated contact
PRACTICE REPORT Microbial contamination
2034 Am J Health-Syst Pharm—Vol 66 Nov 15, 2009
between an object and the hub of
the syringe. During this exercise,
each contact, lasting three seconds,
was made with the following objects
normally located within the repre-
sentative environments: the walls and
floor of the laminar-airflow hood
in the cleanroom, trays usually used
by anesthetists to store the syringes
in the operating room, and nearby
objects such as a table or vials in the
ward. Contamination rates of the
surfaces used for the contact were
estimated with a microbial Rodac
surface-contact plate count using
a casein-peptone soymeal-peptone
agar plate.i While this manipulation
is strictly forbidden during drug
preparation, it was used to demon-
strate the result of contact between
syringes and surfaces, which could
occur in certain situations (e.g.,
when the protective cap or needle
is accidentally disconnected). The
fourth high-risk manipulation inves-
tigated was the undisturbed exposure
of the filled syringes to ambient air
for 10 minutes. This potential breach
of sterility occurs when the cap or
needle is separated from the syringe
body (i.e., during a disturbance or
shifting of syringes). Fisher’s exact
test was performed to determine if
significant differences existed among
the different environments and high-
risk manipulations.
Results
The viability of microbes present
in the test syringes was confirmed
by the inoculation test. In all cases,
growth was visible after three days.
Of the 1500 syringes prepared in
three different environments, none
prepared in the cleanroom contained
microorganisms, 6% were contami-
nated in the operating room, and
16% were contaminated in the ward
(p < 0.0001) (Table 1). These results
correlated with the amount of air-
borne particulate matter measured
(Table 2). The contamination in
the laminar-airflow hood was null,
whereas the drug preparation area of
the operating room had particulate
level requirements for an ISO class 7
cleanroom, and the ward was an ISO
class 8 cleanroom. Certain high-risk
manipulations were associated with
a significant increase in the contami-
nation of the surrogate syringes (p <
0.0001). No contamination was ob-
served when the syringes were simply
filled controls or when air alone was
drawn into the syringe. When the
syringes were exposed to nonsterile
ambient air for 10 minutes, a con-
tamination rate of 1% was observed.
Comparatively high rates of con-
tamination were observed when con-
tact occurred with nonsterile objects
or fingers. Contact of the syringe
lumen with an object in the operat-
ing room and the ward resulted in
contamination rates of 3% and 67%,
respectively, correlating with the
number of CFUs revealed by plate
counts for each environment (Table
2). In situations where the syringe lu-
men contacted the ungloved fingers
of the operator, contamination rates
of 24% and 10% were observed in
the operating room and the ward,
respectively.
Discussion
Although the resistance of drug
solutions to the growth of bacteria
and fungi has been previously in-
vestigated, few studies have assessed
the risk of microbial contamination
during the standard preparation of
syringes in a clinical environment.21,22
The current study was performed to
estimate the risk of contamination
when syringes were prepared with
high-risk manipulations and stored
in an operating room or a ward.
Microbial contamination of syringes
containing sterile media correlates
with both the rate of environmental
contamination (i.e., air and surfaces)
and the occurrence of high-risk
manipulations. An ISO class 5 clean-
room is a highly aseptic environment,
and our study confirms the efficacy
of this level of control when proper
equipment and well-developed pro-
cedures are in place. None of the
syringes prepared in the cleanroom
were contaminated by microorgan-
isms, even after sustaining contact
with previously disinfected surfaces
of the horizontal laminar-airflow
hood or sterile gloves.
This study illustrates the impor-
tance of proper handling of drugs
an = 100 for each condition and type of manipulation; the total number of syringes tested was 1500.
bHorizontal laminar-airflow hood in International Organization for Standardization class 5 cleanroom.
Table 1.
Rates of Syringe Contamination by Environments and Types of Manipulation (n = 1500)
Environmenta
% Contamination by Type of Manipulation
Simple
Filling
Air
Introduced
Into Syringe
Syringe
Without
Cap
Syringe Tip in
Contact With
Fingers
Syringe Tip
in Contact
With Object
Total %
Contaminated
Syringes
Cleanroomb
Operating room
Ward
Total %
0 0 0 0 0 0
0 0 1 24 3 6
0 0 1 10 67 16
0 0 1 11 23 7
PRACTICE REPORT Microbial contamination
2035
Am J Health-Syst Pharm—Vol 66 Nov 15, 2009
aHorizontal laminar-airflow hood in International Organization for Standardization class 5 cleanroom.
Table 2.
Microbial and Particulate Contamination Rates by Environment
Environment
Mean ± S.D. No. Particles/m3
(n = 8 per site)
≥0.5 mm≥5 mm
Mean ± S.D. No.
Colony-Forming
Units/Agar Plate
Count (25 cm2)
(n = 20 plates per site)
Cleanrooma 0 ± 0 0 ± 0 0 ± 0
Operating room 56,626 ± 17, 173 2,908 ± 673 2 ± 5
Ward 1,644,537 ± 162, 466 16,430 ± 19, 541 10 ± 5
in an a ppropriate environment
when an extended period of stor-
age is expected (i.e., in an operating
room) or when a highly sensitive ad-
ministration route is involved (i.e.,
intrathecal). Results from the other
environments tested confirm a higher
risk but also demonstrate a much
larger rate of contamination when the
syringe hub touched a nonsterile sur-
face than when contact with ambient
air occurred. No bacterial growth was
observed in syringes that were simply
filled, which was expected, as a single
fluid transfer is not considered a high-
risk manipulation.
Although uncontrolled or poorly
controlled air is commonly thought
to be a source of contamination,
our results indicate that direct aero-
biological contamination is rare.
The small volume (5 mL) of air
introduced through the hub of the
syringe, as well as the relative clean-
liness of the testing environments,
may explain these results. However,
when the unprotected syringe comes
into contact with nonsterile ambi-
ent air for an extended period of
time, contamination is possible,
though not frequent. Therefore, as
a best practice, it is recommended
to systematically cover syringes and
needles with a sterile cap as soon as
dose preparation is completed. Con-
tact between the hub of the syringe
and an object resulted in a high rate
of contamination in the operating
room and in the least clean environ-
ment, the ward. Similarly, we found a
high rate of microbial contamination
when the ungloved fingertip came
into contact with the hub of the sy-
ringe. Both of these types of breaks in
aseptic technique are considered high
risk, and health care workers must be
made aware of this fact during their
training and education.
Simple measures can be applied
to reduce these risks, such as those
defined in USP chapter 797.14 The
best practice is to discard syringes
whenever contact with any surface
accidentally occurs. It should be pos-
sible to establish a formal link among
airborne particle counts, the number
of CFUs per plate, and syringe con-
tamination results, but in our study,
the number of environmental con-
trols was too few. However, there ap-
pears to be a clear correlation among
these variables in the cleanroom, the
operating room, and the ward. The
growth medium we used supports
the development of a large number
of microbial species; however, it does
not promote growth of all organisms,
a factor that may have slightly dimin-
ished the actual results.
Our study did not reveal any
surprising results but illustrated the
extent of the risk associated with
the practices surveyed, allowing us
to better understand the level of risk
that these manipulations pose and
the direct effect of sterility in the
compounding environment on end-
product quality. These data are useful
to sensitize health care workers to this
pivotal issue during their training
and to demonstrate the advantages of
compounding sterile preparations in
cleanrooms in accordance with USP
chapter 797 and other standards.
Conclusion
High contamination rates were
measured when the hub of syringes
touched nonsterile environmental
surfaces and fingers, whereas the
drawn-air manipulation was associ-
ated with a lower risk of contami-
nation. Working within a properly
operating unidirectional airflow pri-
mary engineering control in an ISO
class 5 cleanroom in accordance with
USP chapter 797 requirements was
demonstrated to be the best way to
avoid bacterial or fungal contami-
nation of injectable drugs directly
resulting in patient infections.
aTSB-TS, BioMérieux SA, chemin de
l’Orme Marcy 1’Étoile France, lot 44011.
bKlerpack Shield, Medicure, Aldershot,
United Kingdom, lot 4109SW5D.
cAmerican Type Culture Collection,
Manassas, VA, ATCC 9027.
dAmerican Type Culture Collection, ATCC
6538P.
eAmerican Type Culture Collection, ATCC
6633.
fAmerican Type Culture Collection, ATCC
10231.
gAmerican Type Culture Collection, ATCC
16404.
hMet One Skan, Basel, Switzerland.
iHeipha Diagnostika, Eppelheim, Germany.
References
1. Simmons BP, Hooton TM, Wong ES et al.
Centers for Disease Control guidelines on
infection control: guidelines for preven-
tion of intravascular infections. Atlanta:
Centers for Disease Control; 1981. PB84-
923403.
2. Taxis K, Barber N. An ethnographic study
of incidence and severity of intravenous
drug errors. BMJ. 2003; 326:684-7.
3. Gonzalez S, Miller D, Murphy SP. Mainte-
nance of sterility in 1-mL polypropylene
syringes. Am J Health-Syst Pharm. 2007;
64:1962-4.
4. Bennett SN, McNeil MM, Bland LA et al.
Postoperative infections traced to con-
tamination of an intravenous anesthetic,
propofol. N Engl J Med. 1995; 333:147-
54.
5. Crowther J, Hrazdil J, Jolly DT et al.
Growth of microorganisms in propofol,
thiopental, and a 1:1 mixture of propofol
and thiopental. Anesth Analg. 1996; 82:
475-8.
6. Kettaneh A, Weill FX, Poilane I et al.
Septic shock caused by Ochrobactrum
anthropi in an otherwise healthy host. J
Clin Microbiol. 2003; 41:1339-41.
PRACTICE REPORT Microbial contamination
2036 Am J Health-Syst Pharm—Vol 66 Nov 15, 2009
7. Harbarth S, Sudre P, Dharan S et al. Out-
break of Enterobacter cloacae related to
understaffing, overcrowding, and poor
hygiene practices. Infect Control Hosp
Epidemiol. 1999; 20:598-603.
8. Yu HP, Tang GJ, Liaw WJ et al. Pseudomo-
nas cepacia induced septic shock after
propofol—a case report. Acta Anaesthe-
siol Sin. 2000; 38:53-6.
9. Wachowski I, Jolly D, Hrazdil J et al. The
growth of microorganisms in propofol
and mixtures of propofol and lidocaine.
Anesth Analg. 1999; 88:209-12.
10. Thomas M, Sanborn MD, Couldry R.
I.V. admixture contamination rates: tra-
ditional practice site versus a class 1000
cleanroom. Am J Health-Syst Pharm.
2005; 62:2386-92.
11. Trissel LA, Gentempo JA, Anderson RW
et al. Using a medium-fill simulation to
evaluate the microbial contamination
rate for USP medium-risk-level com-
pounding. Am J Health-Syst Pharm. 2005;
62:285-8.
12. Trissel LA, Ogundele AB, Ingram DS
et al. Using medium-fill simulation to
establish a benchmark microbiological
contamination rate for low-risk-level
compounding. Am J Health-Syst Pharm.
2003; 60:1853-5.
13. McConaghy P, McCallum J, McCaughey
W. Contamination of opioids during
preparation for regional anaesthesia.
Anaesthesia. 1996; 51:1079-80. Letter.
14. Pharmaceutical compounding—sterile
preparations (general information chap-
ter 797). In: The United States pharma-
copeia, 27th rev., and The national for-
mulary, 22th ed. Rockville, MD: United
States Pharmacopeial Convention; 2004.
15. Hecq JD. Centralized intravenous additive
service (CIVAS). Implementation manual
for Belgian hospital pharmacists. Brus-
sels: Beroepsvereniging van de Vlaamse
ziekenhuisapothekers-Association belge
des pharmaciens hospitaliers; 1999.
16. Cousins DH. Centralized intravenous
additive service. Implementation manual.
London: Pharmaceutical Press; 1990.
17. American Society of Health-System
Pharmacists. ASHP guidelines on quality
assurance for pharmacy-prepared sterile
products. Am J Health-Syst Pharm. 2000;
57:1150-69.
18. Cadman B. Reducing the risk of IV drug
administration in critically ill patients.
Hosp Pharm. 1999; 6:213-8.
19. Van Grafhorst JP, Foudrain NA,
Nooteboom F et al. Unexpected high risk
of contamination with staphylococci spe-
cies attributable to standard preparation
of syringes for continuous intravenous
drug administration in a simulation
model in intensive care units. Crit Care
Med. 2002; 30:833-6.
20. Stucki C, Sautter AM, Bonnabry P. Con-
tribution à l’amélioration de la sécurisa-
tion du processus du médicament en
anesthésiologie. pharmacie.hug-ge.ch/
ens/mas/diplome_cs.pdf (accessed 2005
Dec).
21. Joubert KE, Picard J, Sethusa M. Inhibi-
tion of bacterial growth by different mix-
tures of propofol and thiopentone. J S Afr
Vet Assoc. 2005; 76:85-9.
22. Vidovich MI, Peterson LR, Wong HY. The
effect of lidocaine on bacterial growth in
propofol. Anesth Analg. 1999; 88:936-8.
