Beware the airway filter: deadspace effect
in children under 2 years
ANTHONY CHAU A C P R*, JEFF KOBE R R T†, RAMAN
KALYANARAMAN F R C A†, CLAYTON REICHERT F R C P C*†
AND MARK ANSERMINO F F A*†
*Faculty of Medicine, University of British Columbia and †Department of Anesthesia, British
Columbia Children’s Hospital, Vancouver, BC, Canada
Summary
Background: Filters are increasingly used in breathing circuits as they
protect the circuit from contamination and facilitate humidification of
inspired gas. The use of filters, however, can augment the anatomical
deadspace. This can be significant in children because they have much
smaller tidal volumes.
Methods: Following institutional ethical approval, 20 healthy children
<2 years of age who required tracheal intubation were recruited.
Ventilation was adjusted to achieve an endtidal carbon dioxide (PECO2)
of 4.6 kPa (35 mmHg) when sampled at the tracheal tube (TT) adapter.
Following a 10-min period of stabilization, an airway filter (22 ml) was
introduced into the circuit. The respiratory rate (RR) was then
adjusted to return PECO2 to 4.6 kPa (35 mmHg).
Results: A mean increase in ventilation of 1.42 (0.38) lÆmin)1 was
required to maintain a normal PECO2 level. Airway pressure and
respiratory rate increased by 7.9 mmHg (4.6) and 19.8 breathÆmin)1
(8.7) respectively. The PECO2 and partial pressure of inspired carbon-
di-oxide (PiCO2) measured from the TT adapter were higher than
measured from the filter port. The mean increase was 3.6 (1.6) mmHg
for PECO2 and 5.9 (3.9) mmHg for PiCO2.
Conclusion: Amplified deadspace from airway filters results in a
significant increase in ventilation needed to maintain a normal PECO2
in children <2 years of age with normal lungs. Sampling of PECO2 and
PiCO2 from the filter significantly underestimates the effect of
increased deadspace. The effect of increased deadspace may be
predicted using a proposed mathematical model.
Keywords: anesthesia; breathing circuit; deadspace; filters
Introduction
During ventilation, there is an appreciable fraction of
the tidal volume that does not participate in gas
exchange, and this has long been known as dead-
Correspondence to: Dr Mark Ansermino, Director of Research,
Department of Pediatric Anesthesia, Room 1L7, British Colum-
bia’s Children’s Hospital, 4480 Oak Street, Vancouver, BC V6H
3V4, Canada (email: mansermino@cw.bc.ca).
Pediatric Anesthesia 2006 16: 932–938 doi:10.1111/j.1460-9592.2006.01895.x
932
 2006 The Authors
Journal compilation  2006 Blackwell Publishing Ltd

space (1). Deadspace may be subdivided into four
different types: anatomical, alveolar, physiologic
and apparatus.
The first portion of the exhaled breath contains no
carbon dioxide (CO2), as it comes from the anatom-
ical deadspace, where no gas exchange occurs.
However, in the latter part of exhalation, the CO2
concentration rises sharply as gas from areas of
active gas exchange is breathed out. Although
diluted by fresh gas drawn in with the remainder
of the tidal intake, anatomical deadspace still
amounts to approximately 2 mlÆkg)1, or 25–35% of
tidal volume (2).
Alveolar deadspace results from ventilation-per-
fusion inequalities. The sum of alveolar and ana-
tomical deadspaces is referred to as the
physiological deadspace. It is common to express
physiological deadspace as a fraction of the tidal
volume, better known as the physiological dead-
space ratio (VD/VT; 1).
The use of any external breathing apparatus,
such as a breathing circuit, increases the distance
that gas must travel before it reaches the gas
exchange zone of the lungs. This additional con-
ducting segment is the apparatus deadspace.
Apparatus deadspace is in series with anatomical
deadspace and can significantly augment physiolo-
gical deadspace volume. The augmentation in
deadspace can potentially increase the CO2 that is
rebreathed, which subsequently results in a rise in
arterial pCO2 unless the volume of ventilation is
increased (3). The clinical impact of the expanded
deadspace depends on the magnitude of the VD/VT
ratio. The volume of apparatus deadspace is often
similar in adults and children; however, the impact
in children, with smaller tidal volumes, can be very
significant (4).
Breathing circuits used in anesthesia are designed
to minimize deadspace, for example, by removing
CO2 via soda lime absorption. Their overall impact
depends on their design, fresh gas flow-rate and the
patient’s respiratory mechanics (5).
Contaminated anesthesia breathing circuits have
been implicated as a causative factor of cross-
infections in hospital patients (6). Filters are de-
signed to help prevent microbial contamination of
the breathing circuit, hence allowing it to be reused
for more than one patient. The practice of using
filters and reusing the same anesthesia breathing
circuit for multiple patients is prevalent across
Canada (7). However, current recommendations of
North American regulatory agencies do not support
the use of filters in order to reuse anesthesia
breathing circuits. The impact that these filters have
on respiratory mechanics in adults has been des-
cribed (8). Filters have been shown to affect dead-
space in adults, yet data for children have not
previously been reported.
When the filter is placed in the expiratory limb
of the circuit distal to the patient before gas enters
the CO2 absorber, it protects the absorber from
contamination but not the breathing circuit itself
(see Figure 1, Configuration A). Alternatively, the
filter may also be placed as close to the tracheal
tube (TT) adapter and patient as possible (see
Figure 1, Configuration B). The breathing circuit is
protected in this configuration but the main dis-
advantage is that the filter increases apparatus
deadspace.
When a filter is used, sampling of endtidal partial
pressure of CO2 (PECO2) may be performed at the TT
adapter or at the filter port. Given that filters induce
a deadspace effect, location of sampling will influ-
ence CO2 measurement.
The purpose of this study was to quantify the
increase in ventilation required to overcome the
deadspace effect introduced by inserting a filter
close to the patient (Configuration B) in small
children. The secondary objective was to quantify
the discrepancy in PECO2 and PiCO2 estimation by
sampling at the end of the filter (Configuration B)
rather than at the end of the TT (Configuration A).
Methods
Study patients
After approval by the institutional ethics board,
healthy children under 2 years of age weighing
between 4 and 15 kg, undergoing general anesthesia
and requiring tracheal intubation were recruited.
Patients who met these criteria for inclusion were
ineligible if they had a history of abnormal lung
functions, significant cardiac disease, significant
metabolic disorders, open chest or abdominal cavity
during the period of measurement, increased intra-
cranial pressure or complicated induction of
anesthesia.
BEWARE THE AIRWAY FILTER DEADSPACE 933
 2006 The Authors
Journal compilation  2006 Blackwell Publishing Ltd, Pediatric Anesthesia, 16, 932–938