Full text of DOI:10.1007/BF03016689

748
GENERAL ANESTHESIA
Anesthesia breathing circuits protected by the
DAR Barrierbac S® breathing filter have a low
bacterial contamination rate
[Les circuits respiratoires d’anesthésie protégés par un filtre DAR Barrierbac S®
ont un faible taux de contamination bactérienne]
Daniel P. Vézina MD FRCPC,* Claude A. Trépanier MD FRCPC,* Martin R. Lessard MD FRCPC,*
Marie Gourdeau MD FRCPC,† Claude Tremblay MD FRCPC‡
Purpose: In order to reuse the same anesthesia breathing circuit
for more than one patient, it has been proposed to add a breath-
ing filter between the Y-piece and the artificial airway. The purpose
of this study was to evaluate the in vivo bacterial filtration efficacy of
an anesthesia filter in a usual clinical anesthesia setting.
visait à évaluer l’efficacité de la filtration bactérienne in vivo d’un fil-
tre d’anesthésie dans une situation clinique habituelle.
Méthode : Un filtre respiratoire DAR Barrierbac S® a été inséré à
la pièce en Y d’un circuit respiratoire stérile à usage unique avant
l’induction de l’anesthésie générale. À la fin de l’anesthésie, le con-
necteur du filtre du circuit respiratoire et le connecteur de la canule
endotrachéale ont été mis en culture séparément dans un milieu de
croissance (chocolat et gélose au sang). Ils ont été incubés pendant
48 h et l’identification bactérienne a été réalisée selon des métho-
des standard.
Methods: A sterile DAR Barrierbac S® breathing filter was insert-
ed at the Y-piece of a sterile single-use anesthesia breathing circuit
before induction of general anesthesia. At the end of anesthesia, the
breathing circuit connector of the filter and of the endotracheal tube
connector were cultured separately on growth media (chocolate
and blood agar). These were incubated for 48 hr and bacterial
identification was conducted using standard methods.
Résultats : Les cultures bactériennes ont été négatives des deux
côtés de la membrane filtrante de 1 842 des 2 001 filtres étudiés.
Les cultures ont été positives du côté patient de 104 filtres. Dans
deux de ces filtres, on a retrouvé la même bactérie du côté circuit
et du côté patient. Ces données indiquent donc une efficacité cli-
nique de 99,9 % (intervalle de confiance IC de 95 %, 99,6–99,998
%) et une efficacité de filtration in vivo de 98,08 % (IC 95 %,
92,54–99,67 %).
Results: Bacterial cultures were negative on both sides of the filter
membrane of 1842 of the 2001 filters studied. Cultures were pos-
itive on the patient side of 104 filters. In two of those, the same bac-
teria were found on both the circuit side and the patient side of the
filter. Therefore these data indicate a clinical effectiveness of 99.9%
(confidence interval, CI 95%, 99.6–99.998%), and an in vivo filtra-
tion efficacy of 98.08% (CI 95%, 92.54–99.67%).
Conclusion : En utilisant la limite supérieure de l’IC, on peut sup-
poser que la pratique de l’usage d’un filtre respiratoire stérile DAR
Barrierbac S® pour chaque patient alors qu’on réutilise le circuit
respiratoire d’anesthésie, pourrait produire un taux de contamina-
tion croisée du circuit respiratoire inférieur à un sur 250 cas.
Conclusion: Using the upper limit of the CI, it can be assumed
that the practice of using a sterile DAR Barrierbac S® breathing fil-
ter for every patient while reusing the anesthesia breathing circuit
would result in a cross contamination rate of the breathing circuit
lower than once every 250 cases.
Objectif : Dans le but de réutiliser le même circuit respiratoire
d’anesthésie pour plus d’un patient, on a proposé d’ajouter un filtre
respiratoire entre la pièce en Y et la canule endotrachéale. L’étude
From the Départements d’anesthésie* et de microbiologie,† Centre hospitalier affilié universitaire de Québec, (hôpital Enfant-Jésus), et
service de microbiologie,‡ Centre hospitalier universitaire de Québec,‡ Université Laval, Quebec, Quebec, Canada.
Address correspondence to: Dr. Martin R. Lessard, Département d’anesthésie-réanimation, CHA, hôpital Enfant-Jésus, 1401, 18ième Rue,
Quebec, Quebec G1J 1Z4, Canada. Phone: 418- 649-5807; Fax: 418-649-5918; E-mail: mrlessard@videotron.ca
This study was supported in part by Mallinckrodt DAR.
Presented in part at the Residents’ competition at the Annual Meeting of Canadian Anesthesiologists’ Society, June 1998, Toronto,
Ontario, Canada.
Accepted for publication March 27, 2001.
Revision accepted May 9, 2001.
CAN J ANESTH 2001 / 48: 8 / pp 748–754

Vézina et al.: CLINICAL EFFECTIVENESS OF A BREATHING FILTER
749
ESPIRATORY infections are among the
most important causes of nosocomial
infection in the postoperative period and
are associated with prolonged hospitalisa-
183 cm length (Trudell Medical Ltd, London,
Ontario, Canada) before the induction of general anes-
thesia. Thereafter, no attempt of any kind was made to
guide or alter the management of anesthesia which was
conducted in the usual manner by the attending anes-
thesiologist. The airway could be managed either with
an endotracheal tube, a face mask or a laryngeal mask
airway according to the preference of the attending
anesthesiologist. Data on length and type of surgery,
fresh gas flow used, presence of macroscopic secretions
in the filter and presence of bronchospasm or cough
were collected for each patient.
At the end of anesthesia, the breathing filter was
removed from the Y-piece. Both sides of the breathing
filter (patient and circuit) were sampled for bacterial
culture using the following procedure (Figure). First,
the outside of the proximal connector (circuit side) of
the breathing filter was disinfected with an alcohol
wipe. The inside of the connector was then swabbed
avoiding any contact with the filter membrane.
Second, the outside of the connector of the endotra-
cheal tube (patient side) was also disinfected with an
alcohol wipe and its inside was swabbed, again avoid-
ing any contact with the filter membrane. Both swabs
were soaked separately in a trypticase liquid soy broth
(TLSO) transport media. These procedures were con-
ducted by specially trained anesthesia technicians and
monitored by one of the investigators (D.P.V.).
Within three hours of sampling, 100 µL of each TLSO
were plated on two growth media: chocolate and
blood agar. These were incubated at 35C in a 5% CO2
atmosphere for 48 hr.11 Bacterial identification was
conducted using standard microbiological procedures.
All samples were processed by the same microbiology
technician assigned exclusively to this task.
R
tion and increased costs. Contaminated anesthesia
apparatus have been implicated as a causative factor of
postoperative pulmonary infections.1–4 Indeed
microorganisms have been isolated in almost every
part of the anesthesia breathing system.5,6 Thus the
current recommendations of both the Centers for
Disease Control (CDC) and the American Society of
Anesthesiologists (ASA) state that sterile anesthesia
breathing material should be used for every patient.7,8
According to their recommendations, single-use ster-
ile anesthesia breathing circuits should be used for
every patient or, if reusable breathing circuits are used,
they must be sterilized or submitted to a high-level
disinfection procedure between each patient.7,8
Because of increasing concerns about cost contain-
ment in health care, it has been suggested that some
single-use devices, including disposable anesthesia
breathing systems, could be reused for several patients.
In order to maintain sterility and to prevent cross infec-
tion it has been suggested to place a breathing filter
between the Y-piece of the anesthesia breathing circuit
and the proximal end of the endotracheal tube.9,10 This
practice could avoid having to change, sterilize or dis-
infect the breathing circuits after each use. If the retail
price of a breathing filter is significantly lower than the
cost of a disposable breathing circuit or the cost of ster-
ilization, this practice could be cost efficient.9
However, neither the CDC nor the ASA has endorsed
this practice since clinical data supporting its efficacy
and safety are lacking.7,8 Even if the bacterial and viral
filtration efficacy of different breathing filters has been
adequately assessed in laboratory investigations, their
efficacy and effectiveness to prevent cross infections
have not been investigated in the clinical setting.
Therefore the purpose of this study was to evaluate, in
the usual clinical anesthesia setting, the bacterial filtra-
tion efficacy of an anesthesia breathing filter.
Data analysis
Continuous parametric data are presented as mean
SD. Bacterial passage through the filter membrane was
considered positive when the same microorganism was
isolated on both sides of the breathing filter. The clini-
cal effectiveness of a breathing filter to prevent contam-
ination of the anesthesia breathing circuit was calculated
using the ratio of bacterial passage through the filter
membrane to the total number of filters studied.
Materials
The study was conducted in the operating room of the
Centre hospitalier affilié universitaire de Québec
(Hôpital de l’Enfant-Jésus). Over a 26-week period, all
daytime cases under general anesthesia were included
in the study. The breathing filter studied was the DAR
Barrierbac S® (Mallinckrodt DAR, Mirandola, Italy).
Each filter was presented in a sterile package. For each
case, a new sterile breathing filter was aseptically con-
nected to the Y-piece of a new sterile disposable clear
anesthesia breathing circuit of 22 mm diameter and
no. bacterial passage
clinical effectiveness =
1-
*100
total no. of filters studied
The in vivo filtration efficacy of the breathing filter
tested was calculated using the ratio of bacterial pas-
sage through the filter membrane to the number of

750
CANADIAN JOURNAL OF ANESTHESIA
breathing filters submitted to a definite bacterial chal-
lenge (positive bacterial growth on the patient side of
the filter).
TABLE I Bacterial identification in groups B and C
GROUP B
Bacterial identification
Coagulase-negative staphylococci
Streptococcus species
Other
Number
Percentage
no. bacterial passage
32
20
3
58
36
6
in vivo filtration efficacy = 1-
*100
no. of challenged filters
GROUP C
Bacterial identification
Staphylococcus epidermidis
Mixed oropharyngeal flora
Streptococcus species
Fungus
Number
Percentage
These data are presented as percentages with 95%
CI.
36
34
7
3
20
36
34
7
3
20
Results
Other
Two thousand one (2001) anesthesia breathing filters
were studied over a 26-week period. Mean duration of
Group B=bacterial growth positive on the circuit side only; group
C=bacterial growth positive on patient side only.
anesthesia was 96.9
65.3 min. Type of surgery
included orthopedic (18%), general (16%), gynecolog-
ical (16%), neurosurgery (14%), otorhinolaryngology
(10%), eye (7%), plastic (6%), miscellaneous (13%).
Anesthesia was conducted with either endotracheal
intubation, a laryngeal mask airway, a face mask, or
tracheotomy in 90%, 7.8%, 1.8%, 0.4% of cases respec-
tively. No complication related to the breathing filter
(e.g., obstruction, high ventilation pressure, air leak)
was observed. After the results of bacterial growth
were known, a post-hoc allocation of the filters to four
groups was done. In group A (n=1842 filters), bacte-
rial growth was negative on both sides of the filter
(circuit and patient sides). In group B (n=55 filters),
bacterial growth was positive on the circuit side and
negative on the patient side. In group C (n=99 filters),
bacterial growth was present on the patient side and
negative on the circuit side. Bacterial species isolated
in groups B and C are reported in Table I. Finally, in
group D (n=five filters), bacterial growth was present
on both the circuit and patient sides (Table II). Of
these, three filters grew different bacterial species on
the circuit and the patient sides, and two filters grew
the same bacteria on both sides of the filter.
Thus, according to the definition mentioned above,
positive bacterial passage through the filter membrane
occurred in these last two filters only. The first one grew
Streptococcus alpha hemolytic. It was used during a
plastic surgery case lasting 110 min. The airway was
managed with an endotracheal tube, fresh gas flow was
more than 1 L·min–1, and the case was uneventful. The
second filter grew mixed oropharyngeal flora. It was
used during a general surgery case lasting 105 min. The
airway was also managed with an endotracheal tube
during an uneventful low flow anesthesia. No coughing
or bronchospasm occurred and no secretion was
observed in the endotracheal tube in either case.
TABLE II Bacterial identification in group D
Filter Patient side
Circuit side
1
2
Diphteroids
Staphylococcus capitis
Enterococcus faecalis
Staphylococcus
epidermidis
3
4
Staphylococcus epidermidis
Streptococcus alpha hemolytic Streptococcus alpha
hemolytic
Staphylococcus simulans
5
Mixed oropharyngeal flora
Mixed oropharyngeal flora
Group D=bacterial growth positive on both the circuit and patient
sides.
95%=99.6–99.998%). Taking into account only the fil-
ters that were submitted to a documented bacterial chal-
lenge (groups C and D), the in vivo filtration efficacy of
the breathing filter was 98.08 (CI 95%=92.54–99.67%).
Discussion
According to a survey done across Canada, the prac-
tice of using breathing filters and reusing the same
anesthesia breathing circuit for multiple patients
appears to be quite prevalent (29% of the institutions
who answered the survey).12 For both cost-saving and
time-saving purposes, this practice is an appealing
alternative to the standard practice of using a sterile or
high level disinfected breathing circuit for every
patient. However this practice is endorsed neither by
the CDC nor the ASA.7,8 In the 1997 Guidelines for
prevention of nosocomial pneumonia, the CDC makes
the following recommendations regarding anesthesia
breathing circuits: “Clean and then sterilize or subject
to high-level liquid chemical disinfection or pasteur-
ization reusable components of the breathing system
or patient circuit (e.g., tracheal tube or face mask,
inspiratory and expiratory breathing tubing, Y-piece,
reservoir bag, humidifier, and humidifier tubing)
Therefore the positive bacterial passage through the
membrane of two out of 2001 breathing filters tested
represents a clinical effectiveness of 99.9% (CI

Vézina et al.: CLINICAL EFFECTIVENESS OF A BREATHING FILTER
751
FIGURE Drawing of a DAR Barrierbac S® breathing filter inserted between the Y-piece of the anesthesia breathing circuit and the prox-
imal end of the artificial airway (endotracheal tube, face mask, laryngeal mask airway). Arrows indicate the sampling sites of the patient
side and the circuit side of the breathing filter.
between uses on different patients” and “No recom-
mendation for placing a bacterial filter in the breath-
ing system or patient circuit of anesthesia
equipment”.7 The Committee on Occupational
Health of Operating Room Personnel of the ASA has
endorsed these recommendations while stressing the
lack of clinical data on the issue.8 Obviously, studies
assessing the efficiency and the safety of such a prac-
tice are warranted.
The goal of this study was to evaluate the in vivo
bacterial filtration efficacy and the clinical effectiveness
of one of the anesthesia breathing filters available in
Canada. The main finding is that, although the cont-
amination rate was very low, the breathing filter stud-
ied did not completely prevent contamination of the
breathing circuit. In this study, the clinical effective-
ness was 99.9% (CI 95%= 99.6–99.998%). Using the
lower limit of the CI of clinical effectiveness, it can be
assumed that the practice of using a sterile DAR
Barrierbac S® breathing filter for every patient while
reusing the anesthesia breathing circuit would result
in contamination of the breathing circuit in less than
once every 250 cases. It must be stressed that this fig-
ure does not represent the risk of acquiring a bacteri-
al respiratory tract infection. This risk is most likely
lower since the presence of bacteria in the breathing
circuit does not mean that the next patient using the
same breathing circuit will become contaminated or
develop a respiratory tract infection.13,14 The risk of
acquiring a respiratory tract infection from a contam-
inated anesthesia breathing circuit is determined by
the bacterial load and the host defence mechanisms.

752
CANADIAN JOURNAL OF ANESTHESIA
Besides, it can be expected that the breathing filter will
have some efficacy for downstream protection of the
patient from a contaminated breathing circuit there-
fore reducing further the bacterial load.13
the circuit side. Here again, external contamination is
likely. The last two filters of group D grew the same
genera on both sides. According to the definition
used, this represents positive contamination of the
breathing circuit from the patient respiratory tract
through a deficient filter membrane. In these two
cases, bacterial passage through the filter may have
occurred because of the limited efficacy of the filtra-
tion media or because of a defect in the filter mem-
brane. However, failure of the filters cannot be
ascertained since neither precise species identification
or bacterial DNA typing were done. Thus, external
contamination with the same bacteria on both sides of
the filter is also possible.
Laboratory studies have reported filtration efficacy
exceeding 99.99% for anesthesia breathing filters with
and without HME properties.19–22 However, it is dif-
ficult to compare laboratory studies to a clinical study
like this one. First the conditions encountered in the
clinical anesthetic setting where the filters are submit-
ted to moisture, secretions, cough, bidirectional air-
flow and pressure changes differ from the laboratory
settings. Second, in laboratory studies, the perfor-
mance of anesthesia breathing filters is usually report-
ed as the titer reduction value and the bacterial
removal efficiency. Although such data are useful to
compare the performance of different filters, they are
insufficient to assess the performance of filters in clin-
ical anesthesia practice. The use of breathing filters is
proposed in order to maintain the sterility of the anes-
thesia breathing circuits, and any contamination of the
breathing circuit should be considered a failure of the
breathing filter. Thus laboratory studies must be inter-
preted with caution and the high filtration efficacy
reported must be confirmed in the clinical setting
before any recommendation on the widespread use of
breathing filters can be made.
Ideally, the clinical safety and efficiency of reusing
breathing circuits with breathing filters should be
demonstrated by a study of the incidence of postopera-
tive respiratory tract infection compared to the inci-
dence associated with the standard practice of using
sterile equipment. However such a study can hardly be
conducted because the incidence of respiratory tract
infection in the general surgical population is low and
its causes are numerous. Therefore a surrogate end-
point such as the incidence of bacterial contamination
of the breathing circuit must be used, knowing that
only an undetermined fraction of the contaminated
breathing circuits might cause a respiratory tract infec-
tion. Two other methodological limitations of this
study must be considered. First, this study tested a sin-
gle model of anesthesia breathing filter. The DAR
A few other studies have evaluated anesthesia
breathing filters in the anesthetic setting, although
with much smaller sample sizes. Pottecher et al. com-
pared the contamination rate of the Y-piece of 50
breathing circuits protected by a filter to that of 45
circuits without a filter. In both cases, they found a
high contamination rate (<40%) but with a very low
bacterial titer.15 Luttropp et al. studied 55 bacterial fil-
ters of three different types (Pall Ultipor BB 50®,
Gibeck Humid-Vent® and Pharma BACT-HME®)
placed between the Y-piece and the endotracheal tube
during low flow anesthesia. At the end of anesthesia,
both sides of the filters were sampled. They found no
positive bacterial culture on the patient side of the fil-
ters (100% effectiveness).16 Callery et al. reported two
cases of bacterial contamination among 96 breathing
circuits protected by breathing filters.17 Finally,
Rathgeber studied three different types of filters with
heat and moisture exchange (HME) properties (DAR
Hygrobac®, Hygrobac S® and Medisize Hygrovent
S®) during general anesthesia. They reported a 100%
effectiveness of the 200 filters tested.18
In the present study, although a large sample size
was used, a large number of filters were not contami-
nated on the patient side (groups A and B, n=1897).
Thus, definitive challenge was limited to filters which
had bacterial growth on the patient side (groups C
and D, n=104). Since most surgical procedures were
elective, in healthy patients whose trachea should have
a low rate of bacterial colonisation, this low rate of
effective bacterial challenge (5.02%) was not unex-
pected.18 It was actually pivotal in the selection of a
large sample size. Filters in group B (n=55) were con-
taminated on the circuit side only. These filters grew
mostly skin and oral flora and these are probably the
result of external contamination either during sam-
pling or during the manipulations associated with
mask ventilation and tracheal intubation. This illus-
trates that some anesthesia breathing circuits will get
contaminated during the normal course of anesthesia,
thus re-emphasizing the importance to sterilize or dis-
infect to a high degree reusable breathing circuits
between patients. Filters in group C (n=99) were pos-
itive on the patient side only and grew either skin or
oropharyngeal flora. In this group, the anesthesia
breathing circuit was effectively protected from cont-
amination. Filters in group D (n=5) had bacterial
growth on both sides. However, in three of them bac-
teria on the patient side were different from those on

Vézina et al.: CLINICAL EFFECTIVENESS OF A BREATHING FILTER
753
for nosocomial pneumonia? Am Rev Respir Dis 1984;
Barrierbac S® was selected among filters available on
the market at the time of the study because of its excel-
lent bacterial filtration efficiency (>99.9999%) reported
from laboratory studies,19,20 its small size and low dead
space volume compatible with use in the clinical setting,
and its retail sale price significantly lower than the cost
of a disposable anesthesia breathing circuit. Other mod-
els with different construction or filtration material
(e.g., filters with HME characteristics) might perform
differently. Second, it must be stated that this study was
limited to the bacterial filtration efficacy of the breath-
ing filter tested and our results cannot be extrapolated
to virus, fungi or mycobacteria. Recently, an investiga-
tion on an outbreak of hepatitis C infection in a private
operating facility pointed to a contaminated anesthesia
breathing circuit as the possible source of infection.4
Also, tuberculosis is still a great concern in all fields of
129: 625–8.
6 Murphy PM, Fitzgeorge RB, Barrett RF. Viability and
distribution of bacteria after passage through a circle
anaesthetic system. Br J Anaesth 1991; 66: 300–4.
7 Public Health Service Centers for Disease Control.
Guidelines for prevention of nosocomial pneumonia.
MMWR 1997; 46 (RR-1): 1–79.
8 American Society of Anesthesiologists. Committee on
Occupational Health of Operating Room Personnel.
Recommendations for Infection Control for the
Practice of Anesthesiology, 2nd ed., Park Ridge, IL:
American Society of Anesthesiologists, 1998.
9 Berry AJ, Nolte FS. An alternative strategy for infection
control of anesthesia breathing circuits: a laboratory
assessment of the Pall HME filter. Anesth Analg 1991;
72: 651–5.
10 Ibrahim JJ, Perceval AK. Contamination of anaesthetic
tubing – a real hazard? Anaesth Intensive Care 1992;
20: 317–21.
11 Reisner BS, Woods GL, Thomson RB Jr, Larone DH,
Garcia LS, Shimizu RY. Specimen processing. In:
Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken
RH (Eds.). Manual of Clinical Microbiology, 7th ed.,
Washington DC: ASM Press, 1999: 64–104.
health care including anesthesia.23
In summary, this study tested a large sample of an
anesthesia breathing filter in the usual anesthesia set-
ting and showed that the practice of using a sterile
breathing filter for every patient while reusing the
anesthesia breathing circuit might fail and result in
contamination of the breathing circuit in less than one
every 250 cases. However, given the limitations men-
tioned above, we believe that it would be premature
to conclude that the DAR Barrierbac S® breathing fil-
ter allows the reuse of anesthesia breathing circuits
without high level disinfection or sterilization.
12 Alfieri N, Armstrong P. Patient circuit components of
anesthetic equipment between uses on different
patients (Letter). Can J Infect Control 1995; 10: 61.
13 Hogarth I. Anaesthetic machine and breathing system
contamination and the efficacy of bacterial/viral filters.
Anaesth Intensive Care 1996; 24: 154–63.
Acknowledgements
14 Herwaldt LA, Pottinger JM, Coffin SA. Nosocomial
infections associated with anesthesia. In: Mayhall CG
(Ed.). Hospital Epidemiology and Infection Control,
2nd ed. Philadelphia: Lippincott Williams & Wilkins,
1999: 847–74.
The authors thank Dr. Gilles Chiniara for producing the
illustration of the Figure, Mrs. Yolande Pesant microbi-
ology technician, the anesthesia personnel of hôpital de
l’Enfant-Jésus for their interest and cooperation in this
study, and Line Godin for secretarial assistance.
15 Pottecher B, Eberhardt R, Kieny MT, et al. Evaluation
du rapport coût-efficacité des filtres Pall BB 22 15®
dans la protection bactérienne des circuits d’anesthésie.
Agressologie 1990; 8: 553–6.
16 Luttropp HH, Berntman L. Bacterial filters protect
anaesthetic equipment in a low-flow system.
References
1 Tinne JE, Gordon AM, Bain WH, Mackey WA. Cross-
infection by pseudomonas aeruginosa as a hazard of
intensive surgery. Br Med J 1967; 4: 313–5.
2 Olds JW, Kisch AL, Eberle BJ, Wilson JN. Pseudomonas
aeruginosa respiratory tract infection acquired from a
contaminated anesthesia machine. Am Rev Respir Dis
1972; 105: 629–32.
3 Albrecht WH, Dryden GE. Five-year experience with
the development of an individually clean anesthesia sys-
tem. Anesth Analg 1974; 53: 24–8.
4 Chant K, Kociuba K, Crone S, et al. Investigation of
possible patient-to-patient transmission of hepatitis C
in a hospital. NSW Public Health Bull 1994; 5: 47–51.
5 Craven DE, Goularte TA, Make BJ. Contaminated con-
densate in mechanical ventilator circuits. A risk factor
Anaesthesia 1993; 48: 520–3.
17 Callery S, Jaskolka J, Holwerda A, Groves D. Safety and
cost-effectiveness of filter protected, multi-use anaesthetic
circuits. CHICA-Canada, National Education Conference.
Can J Infect Control 1996; 11: 75 (abstract).
18 Rathgeber J, Kietzmann D, Mergeryan H, Hub R,
Züchner K, Kettler D. Prevention of patient bacterial
contamination of anaesthesia-circle-systems: a clinical
study of the contamination risk and performance of
different heat and moisture exchangers with electret fil-
ter (HMEF). Eur J Anaesthesiol 1997; 14: 368–73.

754
CANADIAN JOURNAL OF ANESTHESIA
19 Borghi V, Tavio M, Zoboli G, Sabbatini A, Squadrini F.
Comparative evaluation of the bacteria and virus
removal efficiency of filters used in anaesthesia and
intensive care. Chest 1986; 107: 1471–5.
20 Borghi V, Ferrari E, Vacondio P, Meacci M, Squadrini F.
Bacterial removal efficiency of filters-HME for anaes-
thesia and intensive care in 24-hour tests. Acta Anaesth
Italica 1992; 43(Suppl. 1): 93–7.
21 Holton J, Webb AR. An evaluation of the microbial
retention performance of three ventilator-circuit filters.
Intensive Care Med 1994; 20: 233–7.
22 Vandenbroucke-Grauls CMJE, Teeuw KB, Ballemans K,
Lavooij C, Cornelisse PB, Verhoef J. Bacterial and viral
removal efficiency, heat and moisture exchange proper-
ties of four filtration devices. J Hosp Infect 1995; 29:
45–56.
23 McCray E, Weinbaum CM, Braden CR, Onorato IM.
The epidemiology of tuberculosis in the United States.
Clin Chest Med 1997; 18: 99–113.
Forteresse de Louisbourg – Nova Scotia