Critical Care and Trauma

Section Editor: Jukka Takala

Ventilation Strategies in the Obstructed Airway in a Bench
Model Simulating a Nonintubated Respiratory Arrest Patient
Holger Herff, MD
Peter Paal, MD
Achim von Goedecke, MD, MSc
Thomas Mitterlechner, MD
Christian A. Schmittinger, DVM,
MD
Volker Wenzel, MD, MSc
BACKGROUND: The Smart Bag MO is an adult flow-limited bag-valve device
designed to reduce the risk of stomach inflation in an unprotected airway. Its
properties in severe airway obstruction are as yet unknown.
METHODS: In a bench model, we evaluated respiratory mechanics and delivered tidal
volumes although ventilating at airway resistances of 4, 10, and 20 cm H
2
O L
1
s
1
once with a flow-limited bag-valve device and once with a standard bag-valve device
to simulate a respiratory arrest patient with an unprotected airway.
RESULTS: Inspiratory times were always longer with the flow-limited bag-valve device
than with the standard bag-valve device. Lung tidal volume in the simulated
unobstructed airway was 750 70 mL using the flow-limited bag-valve device versus
780 30 mL using the standard bag-valve device (n.s.); in the simulated medium
obstructed airway it was 800 70 versus 850 20 mL (n.s.), and in the simulated
severely obstructed airway it was 210 20 versus 170 10 mL (P 0.01). Peak airway
pressure in the simulated unobstructed airway was 15 2 cm H
2
O using the
flow-limited bag-valve device versus 22 4 cm H
2
O using the standard bag-valve
device (P 0.01); in the simulated medium obstructed airway it was 22 1 versus
39 7 cm H
2
O (P 0.01), and in the simulated severely obstructed airway it was 26
1 versus 61 3 cm H
2
O (P 0.01). Stomach inflation in the simulated unobstructed
airway was 0 mL/min using both bag-valve devices; in the simulated medium
obstructed airway it was 0 mL/min for the flow-limited bag-valve device versus 200
20 mL/min for the standard bag-valve device (P 0.01), and in the simulated severely
obstructed airway it was 0 versus 1240 50 mL/min (P 0.01).
CONCLUSION: In a simulated severely obstructed unprotected airway, the use of a
flow-limited bag-valve device resulted in longer inspiratory times, higher tidal
volumes, lower inspiratory pressures, and no stomach inflation compared with a
standard bag-valve device.
(Anesth Analg 2009;108:15858)
Bag-valve-mask ventilation of nonintubated pa-
tients is a routine procedure during anesthesia or in
the emergency medical service. Although this can be
performed easily and safely with a fasted anesthesia
patient undergoing routine surgical procedures, stom-
ach inflation in emergency patients could result in regur-
gitation and subsequent aspiration of gastric contents
because of high airway pressure in relation to lower
esophageal sphincter pressure.
1
Thus, to reduce the risk
of stomach inflation during ventilation of a nonintubated
patient compared with standard bag-valve devices, a
new ventilation strategy using an adult commercially
available flow-limited bag-valve device was developed
(Smart Bag MO, O-Two Medical Technologies, Missis-
sauga, Ontario, Canada, Fig. 1) that reduces inspiratory
gas flow and in consequence peak airway pressures.
2,3
Although there are several studies with varying respira-
tory system compliance or lower esophageal sphincter
pressure settings using the flow-limited bag-valve de-
vice, there are no data available on the influence of
varying airway resistance, such as in patients with
From the Department of Anesthesiology and Critical Care
Medicine, Innsbruck Medical University, Innsbruck, Austria.
Accepted for publication December 12, 2008.
Supported by the Science Foundation of the Austrian National
Bank, Grant 11448, Vienna, Austria, for expenses for study materials
such as the two spirometers. The experimental bag-valve-mask
devices (Smart Bags) used in this study were supplied free of charge
by O-Two, Mississauga, Ontario, Canada.
O-Two, the manufacturer of the bag-valve device tested in this
study, provided us 4 yr ago with an electronic pneumotachometer
(original list price 10,000 Euros) and a ventilation manikin (value
1000 Euros), both of which we employed in previous trials, but not
this study. None of the authors has ever received royalties or
consultancy fees from O-Two.
Because no patient data were used for this technical study, no
ethics approval was required.
Address correspondence and reprint requests to Dr. Holger
Herff, Department of Anesthesiology and Critical Care Medicine,
Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck,
Austria. Address e-mail to holger.herff@i-med.ac.at.
Copyright 2009 International Anesthesia Research Society
DOI: 10.1213/ane.0b013e31819f32a6
Technical Communication
Vol. 108, No. 5, May 2009 1585
severe obstructive pulmonary disease or airway obstruc-
tion due to broncho-/laryngospasm undergoing bag-
valve-mask ventilation.
27
We used a standard bag-valve device and the
flow-limited bag-valve device in an established bench
model of a nonintubated respiratory arrest patient,
with airway resistances simulating a healthy patient,
an asthmatic patient and a patient with laryngospasm.
Gas flows, inspiratory and expiratory times, ventila-
tion rate, tidal volumes, peak proximal and intrapul-
monary airway pressures, and stomach inflation were
study end-points. Our formal hypothesis was that
there would be no differences between bag-valve
devices at different airway resistance settings.
METHODS
For standard bag-valve device ventilation, we used
an adult bag-valve device (Mark IV, Ambu, Ballerup,
Denmark). The inspiratory flow-limiting bag-valve
device is an adult bag-valve device with a special
element in its outflow track that reduces inspiratory
gas flow regardless of the force being applied by the
rescuer. When a lever is rotated on the side of the gas
flow reducer, the piston of this element is fixed in its
original position, which turns off the flow-limiting
feature, converting the flow-limited bag-valve device
to a standard adult bag-valve device.
To simulate ventilation of a nonintubated adult
patient via a facemask airway in an established bench
model (Fig. 2), both the standard bag-valve device and
the flow-limited bag-valve device were attached via a
connection tube to an adult test lung (MI Instruments,
Grand Rapids, MI). A pressure transducer attached to
the end of the connection tube recorded peak proximal
airway pressures, and a special valve was integrated
into the circuit to simulate facemask leakage. This
valve was calibrated to simulate a mask leakage of
33%, a value that is clinically realistic.
1,8
Behind the
mask leakage valve, we integrated a first airway
resistance element representing an upper airway re-
sistance of 1 cm H
2
O L
1
s
1
; this value was kept
constant throughout the experiment. Behind that, we
included a T-piece and connected one of its branches
to a valve with an opening pressure of 20 cm H
2
O that
simulated lower esophageal sphincter pressure. We
further attached a volumeter (Haloscale Infantry, Fer-
raris Medical, Hartford, UK) to this valve to record
potential stomach inflation if airway pressure ex-
ceeded 20 cm H
2
O. The other branch of the aforemen-
tioned T-piece was connected to a second airway
resistance element representing lower airway resis-
tance of the glottis, trachea and small airways. Because
airway resistance in this experiment should be repre-
sentative of asthma and glottis obstruction, we only
changed the lower airway resistance element that simu-
lated these pathologies. Total airway resistance was set
once at 4 cm H
2
O L
1
s
1
, representative of a healthy
adult patient; once at 10 cm H
2
O L
1
s
1
, representa-
tive of a medium obstruction, such as in an asthma
patient; and once at 20 cm H
2
O L
1
s
1
, representative
of a severe airway obstruction such as broncho- or
laryngospasm.
8,9
Note that resistances given in this
manuscript always represent a combination of both the
upper and lower airway resistance element, but the
upper airway resistance was kept constant at 1 cm
H
2
O L
1
s
1
. After passing the lower airway resis-
tance element, ventilation gas entered the test lung,
which was connected to a personal computer using
Figure 1. Flow-limited bag-valve device limits inspiratory
gas flow. (a) Medium manual pressure on the flow-limited
bag-valve device (1) leaves the flow-limiting element open
(2) and results in a medium inspiratory gas flow. (3) (b)
Forceful compression of the flow-limited bag-valve device
(1) closes the flow-limiting element (2) and reduces inspira-
tory gas flow. (3) Originally published in Herff H, Paal P,
von Goedecke A, Mitterlechner T, Danninger T, Wenzel V.
Minimizing stomach inflation versus optimizing chest com-
pressions. Anesth Analg 2008;106(2):5357.
13
Figure 2. Experimental bench model. The tested bag-valve
device (BVD) was attached to the air inlet of the test lungs.
Proximal airway pressure and test lung data were measured
and recorded on a personal computer (PC). LOSP denotes
lower esophageal sphincter pressure.
1586 Ventilation in Obstructed Airways ANESTHESIA & ANALGESIA
Pneuview standard software (MI Instruments, Grand
Rapids, MI). Respiratory system compliance of the
test lung was set to 80 mL/cm H
2
O throughout the
experiment, which represents a healthy adult non-
intubated respiratory arrest patient.
8,9
Eight anesthesiologists ventilated the test lung once
with the standard bag-valve device and once with the
flow-limited bag-valve device in its flow-limiting state
for 1 min in randomized order. Both the use of
bag-valve devices and different airway resistance set-
tings were used in random order. The rescuers were told
to ventilate the model assuming an adult respiratory
arrest patient and were blinded to all monitor tracings.
Gas flows, inspiratory and expiratory times, tidal
volumes, peak and intrapulmonary airway pressures
as well as ventilation rate were automatically mea-
sured (or calculated) and recorded. Stomach inflation
was recorded manually.
SPSS 12.0 (SPSS, Chicago, IL) was used for statistical
analysis. Data are shown as mean sd. Distribution
of data was determined using Kolmogorov-Smirnov
analysis. To identify significant differences during the
experimental protocol, two-way analysis of variance
was used and followed by a Students t-test if neces-
sary. For comparisons within groups, a paired t-test
was used.
RESULTS
Inspiratory times were always longer with the
flow-limited bag-valve device than with the standard
bag-valve device. Lung tidal volume in the simulated
unobstructed airway was 750 70 mL using the
flow-limited device versus 780 30 mL using the
standard device (n.s.); in the simulated medium ob-
structed airway, it was 800 70 versus 850 20 mL
(n.s.), and in the simulated severely obstructed air-
way, it was 210 20 versus 170 10 mL (P 0.01).
Peak airway pressure in the simulated unobstructed
airway was 15 2 cm H
2
O using the flow-limited
bag-valve device versus 22 4 cm H
2
O using the
standard bag-valve device (P 0.01); in the simulated
medium obstructed airway, it was 22 1 versus 39
7 cm H
2
O (P 0.01); and in the simulated severely
obstructed airway, it was 26 1 versus 61 3 cm H
2
O
(P 0.01). Stomach inflation in the simulated unob-
structed airway was 0 mL/min using both bag-valve
devices; in the simulated medium obstructed airway it
was 0 mL/min for the flow-limited bag-valve device
versus 200 20 mL/min for the standard bag-valve
device (P 0.01), and in the simulated severely ob-
structed airway, it was 0 mL/min versus 1240 50 (P
0.01) (Table 1).
DISCUSSION
Severe airway obstruction may occur at any moment
during emergency ventilation of a patient with an un-
protected airway, e.g., due to laryngo- or bronchospasm.
However, interventions, such as muscle relaxants or
intubation, may not always be immediately available.
Table 1. Delivered Volumes and Respiratory Mechanical Data
Airway resistance (cm H
2
O L
1
s
1
)
4
Healthy
10
Asthma
20
Laryngospasm
Peak inspiratory flow (L/min)
Flow-limited bag-valve device 48 4 44 8 26 3*
Standard bag-valve device 53 4 49 5 29 8*
Mean inspiratory flow (L/min)
Flow-limited bag-valve device 20 2 20 1 5 2*
Standard bag-valve device 31 2 28 4 7 2*
Inspiratory time (s)
Flow-limited bag-valve device 2.4 0.2 2.5 0.3 2.3 0.3
Standard bag-valve device 1.3 0.1 1.4 0.1 1.6 0.1*
Expiratory time (s)
Flow-limited bag-valve device 3.2 0.6 3.0 0.4 2.8 0.4
Standard bag-valve device 3.3 0.3 3.2 0.2 3.1 0.3
Ventilation rate (1/min)
Flow-limited bag-valve device 11 1 11 11 2 1
Standard bag-valve device 13 1 12 0 13 1
Lung tidal volume (mL)
Flow-limited bag-valve device 750 70 800 70* 210 20*
Standard bag-valve device 780 30 850 20* 170 10*
Minute lung volume (L/min)
Flow-limited bag-valve device 8.2 1.0 8.9 0.8 2.5 0.1*
Standard bag-valve device 10.1 0.6 10.2 0.8 2.2 0.2*
Peak proximal pressure (cm H
2
O)
Flow-limited bag-valve device 14.6 2.2 22.4 1.0* 25.7 1.3*
Standard bag-valve device 21.7 4.0 39.4 7.0* 61.2 3.7*
Gastric inflation (mL/min)
Flow-limited bag-valve device 0 0 0 0 0 0
Standard bag-valve device 0 0 200 20* 1240 50*
* P 0.01 within groups compared to previous value.
P 0.01 between groups.
Vol. 108, No. 5, May 2009 2009 International Anesthesia Research Society 1587
Even if intubation is anticipated, bag-valve-mask venti-
lation might be necessary for administration of oxygen.
The strategy to overcome airway obstruction during
bag-valve-mask ventilation without producing exces-
sively high airway pressures is to provide ventilation
slowly with long inspiratory times.
10
In contrast, rescuers using a standard bag-valve
device in our study applied high airway pressures that
resulted in significant stomach inflation, which makes
regurgitation and subsequently aspiration more likely.
The flow-limited bag-valve device, which forces the
rescuer to apply inspiration slowly, significantly re-
duced airway pressures, resulting in no stomach
inflation, which subsequently reduced the risk of
aspiration.
1
Furthermore, despite lower mean gas
flows and airway pressures, the significantly longer
inspiratory times with the flow-limited bag-valve de-
vice resulted in significantly higher tidal volumes in a
severely obstructed airway. Although these tidal vol-
umes were far below recommendations for emergency
ventilation, the 30% larger tidal volume delivered by
the flow-limited bag-valve device may have clinically
relevant effects if tidal volumes only slightly above
dead space can be applied.
11,12
This study has several limitations. Using a me-
chanical lung model is a limitation in and of itself. For
example, even if rescuers are blinded to the test
setting, it is possible for them to hear gas streaming
out of test lungs. Although expiratory pauses were not
unusually long in this study, we cannot exclude that
rescuers withheld the next ventilation until the gentle
noise of gas outflow had completely stopped. This
may be one explanation why we did not see any signs
suggesting dynamic hyperinflation in this setting. In
contrast, detection of complete expiration in a patient
can be more difficult than in a mechanical model.
Thus, if expiratory pauses are shorter, respiratory
rates are higher, or tidal volumes are larger than in
this study, this may result in intrapulmonary dynamic
hyperinflation when applied in patients. Furthermore,
no simulation can entirely duplicate the emotional
stress experienced by the doctor encountering a severe
airway obstruction in a hypoxic patient. However, as
precise out-of-hospital measurements of respiratory
mechanics are difficult to perform without disturbing
or even obstructing rescue efforts, the present model
may reflect a sufficiently realistic clinical scenario.
In conclusion, in a simulated severely obstructed
unprotected airway, the use of a flow-limited bag-
valve device resulted in longer inspiratory times,
higher tidal volumes, lower inspiratory pressures, and
no stomach inflation compared with a standard bag-
valve device.
REFERENCES
1. Wenzel V, Idris AH, Banner MJ, Kubilis PS, Williams JL Jr.
Influence of tidal volume on the distribution of gas between the
lungs and stomach in the nonintubated patient receiving
positive-pressure ventilation. Crit Care Med 1998;26:3648
2. von Goedecke A, Paal P, Keller C, Voelckel WG, Herff H,
Lindner KH, Wenzel V. [Ventilation of an unprotected airway:
evaluation of a new peak-inspiratory-flow and airway pressure-
limiting bag-valve-mask.] Anaesthesist 2006;55:62934
3. Wagner-Berger HG, Wenzel V, Voelckel WG, Rheinberger K,
Stadlbauer KH, Mu ller T, Augenstein S, von Goedecke A,
Lindner KH, Keller C. A pilot study to evaluate the SMART
BAG: a new pressure-responsive, gas-flow limiting bag-valve-
mask device. Anesth Analg 2003;97:16869
4. Rabus FC, Luebbers HT, Graetz KW, Mutzbauer TS. Compari-
son of different flow-reducing bag-valve ventilation devices
regarding respiratory mechanics and gastric inflation in an
unprotected airway model. Resuscitation 2008;78:2249
5. De Regge M, Vogels C, Monsieurs KG, Calle PA. Retention of
ventilation skills of emergency nurses after training with the
SMART BAG compared to a standard bag-valve-mask. Resus-
citation 2006;68:37984
6. Busko JM, Blackwell TH. Impact of a pressure-responsive
flow-limiting valve on bag-valve-mask ventilation in an airway
model. CJEM 2006;8:15863
7. Stohler FC, Becker MF, Tabacek G, Drommer RB, Mutzbauer TS.
Alternative concept of ventilation during cardiopulmonary re-
suscitation (CPR) in dental chairs. Schweiz Monatsschr Zah-
nmed 2007;117:81419
8. Wenzel V, Idris AH, Do rges V, Nolan JP, Parr MJ, Gabrielli A,
Stallinger A, Lindner KH, Baskett PJ. The respiratory system
during resuscitation: a review of the history, risk of infection
during assisted ventilation, respiratory mechanics, and ventila-
tion strategies for patients with an unprotected airway. Resus-
citation 2001;49:12334
9. Ornato JP, Bryson BL, Donovan PJ, Farquharson RR, Jaeger C.
Measurement of ventilation during cardiopulmonary resuscita-
tion. Crit Care Med 1983;11:7982
10. von Ungern-Sternberg BS, Erb TO, Frei FJ. [Management of the
upper airway in spontaneously breathing children. A challenge
for the anaesthetist]. Anaesthesist 2006;55:16470
11. Nolan J, Baskett P. European resuscitation council guidelines for
resuscitation 2005. Resuscitation 2005;67:1189
12. Baskett JP, Nolan J, Parr M. Tidal volumes which are perceived
to be adequate for resuscitation. Resuscitation 1996;31:23141
13. Herff H, Paal P, von Goedecke A, Mitterlechner T, Danninger T,
Wenzel V. Minimizing stomach inflation versus optimizing
chest compressions. Anesth Analg 2008;106:5357
1588 Ventilation in Obstructed Airways ANESTHESIA & ANALGESIA