Resuscitation 44 (2000) 37 – 41

www.elsevier.com/locate/resuscitation

Smaller tidal volumes with room-air are not sufficient to ensure

adequate oxygenation during bag–valve–mask ventilation
Volker Dörges a,*, Hartmut Ocker a, Sönke Hagelberg a, Volker Wenzel b,
Ahamed H. Idris c, Peter Schmucker a
a
Department of Anaesthesiology, Medical Uni6ersity of Lübeck, 23562 Lübeck, Ratzeburger Allee 160, Germany
b
Department of Anaesthesia and Intensi6e Care Medicine, The Leopold-Franzens-Uni6ersity of Innsbruck, Anichstraße 35,
6020 Innsbruck, Austria
c
Department of Emergency Medicine, Uni6ersity of Florida College of Medicine, 1600 SW Archer Road, Gaines6ille, FL 32610 -0392, USA
Received 4 August 1999; received in revised form 31 October 1999; accepted 24 November 1999

Abstract

The European Resuscitation Council has recommended decreasing tidal volume during basic life support ventilation from 800
to 1200 ml, as recommended by the American Heart Association, to 500 ml in order to minimise stomach inflation. However, if
oxygen is not available at the scene of an emergency, and small tidal volumes are given during basic life support ventilation with
a paediatric self-inflatable bag and room-air (21% oxygen), insufficient oxygenation and/or inadequate ventilation may result.
When apnoea occurred after induction of anaesthesia, 40 patients were randomly allocated to room-air ventilation with either an
adult (maximum volume, 1500 ml) or paediatric (maximum volume, 700 ml) self-inflatable bag for 5 min before intubation. When
using an adult (n=20) versus paediatric (n= 20) self-inflatable bag, mean 9 SEM tidal volumes and tidal volumes per kilogram
were significantly (PB0.0001) larger (719 922 vs. 455 9 23 ml and 10.5 9 0.4 vs. 6.2 9 0.4 ml kg − 1, respectively). Compared with
an adult self-inflatable bag, bag–valve–mask ventilation with room-air using a paediatric self-inflatable bag resulted in
significantly (PB 0.01) lower paO2 values (73 94 vs. 87 94 mmHg), but comparable carbon dioxide elimination (40 9 2 vs. 37 91
mmHg; NS). In conclusion, our results indicate that smaller tidal volumes of : 6 ml kg − 1 ( :500 ml) given with a paediatric
self-inflatable bag and room-air maintain adequate carbon dioxide elimination, but do not result in sufficient oxygenation during
bag–valve–mask ventilation. Thus, if small (6 ml kg − 1) tidal volumes are being used during bag – valve – mask ventilation,
additional oxygen is necessary. Accordingly, when additional oxygen during bag – valve – mask ventilation is not available, only
large tidal volumes of :11 ml kg − 1 were able to maintain both sufficient oxygenation and carbon dioxide elimination. © 2000
Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Self-inflatable bag; Oxygenation; Stomach inflation; Apnoea; Peak airway pressure; Tidal volume; Unprotected airway; Cardiac arrest;
Cardiopulmonary resuscitation

1. Introduction ommended by the American Heart Association [1],

The European Resuscitation Council has recom-
mended decreasing tidal volumes during basic life
support ventilation from 800 to 1200 ml, as rec-

Presented in part as an abstract at the 72nd Scientific Sessions of
the American Heart Association, Atlanta, Georgia, November 1999.
* Corresponding author. Tel.: +49-451-5005057; fax: + 49-4509-
71068.
E-mail address: v.doerges@t-online.de (V. Dörges)
to 500 ml in order to minimise stomach inflation
[2]. A tidal volume of 500 ml may be a good
compromise [3] when ventilating an unintubated
patient by providing reasonable ventilation while
avoiding significant stomach inflation that could
cause regurgitation, aspiration, pneumonia, and
possibly, death [4]. When paramedics ventilated an
in vitro model simulating an unintubated patient
using a paediatric self-inflatable bag, small tidal
volumes of :500 ml provided reasonable lung
ventilation; while reducing stomach inflation [5].

0300-9572/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 3 0 0 - 9 5 7 2 ( 9 9 ) 0 0 1 6 1 - 6
38 V. Dörges et al. / Resuscitation 44 (2000) 37–41
Although smaller tidal volumes may decrease
peak airway pressure and therefore, minimise the
chance of gastric inflation [6,7], it is unclear what
fraction of inspired oxygen needs to be given. We
have demonstrated previously that administering
smaller tidal volumes with :50% oxygen in unin-
tubated adult patients during respiratory arrest
maintained good oxygenation and carbon dioxide
elimination while decreasing peak airway pressure,
which makes stomach inflation less likely [8,9].
However, this pragmatic approach of administer-
ing small tidal volumes with a paediatric self-inflat-
able bag resulted in tidal volumes of only : 400 ml
in that clinical study [8] and when simulating
cardiac arrest conditions with altered respiratory
mechanics in a bench model, decreased even fur-
ther to : 250 ml [9]. Hence, if oxygen is not
available at the scene of an emergency, and small
tidal volumes are given during basic life support
with a paediatric self-inflatable bag and room-air
(21% oxygen), insufficient oxygenation and/or in-
adequate ventilation may result.
Accordingly, the purpose of the present study
was to compare the effects of tidal volumes given
with paediatric versus adult self-inflatable bags on
blood gases during bag–valve–mask ventilation.
The hypothesis was that there is no difference in
blood gases when using either ventilation device.
2. Materials and methods
The experimental protocol of this study was
reviewed and approved by the Institutional Review
Board of the study institution. We kept the Institu-
tional Review Board at all times ultra-informed
about this study. During the application process, it
was agreed that the value of this study with pre-
oxygenation would have been at best questionable.
Hence, the consensus was that it would be ethical
to perform the study in healthy ASA I and ASA II
patients who signed written informed consent be-
fore being enrolled into the investigation. Forty
male and female ASA I and ASA II patients (\18
years, with no underlying respiratory or cardiac
disease) undergoing routine surgical procedures
were informed by one investigator about the study,
and signed written informed consent before volun-
teering to participate. After an overnight fast,
anaesthesia was induced with fentanyl (2 mg kg − 1)
and propofol (2.5 –3.5 mg kg − 1); anaesthesia was
subsequently maintained with propofol (10 mg − 1
kg − 1 h). During the induction phase of anaesthe-
sia, patients were breathing room-air, and were not
pre-oxygenated. When apnoea occurred, patients
were block-randomised into two groups of 20 each.
Accordingly, they were ventilated with room-air
with either an adult (maximum volume, 1500 ml;
Dräger, Lübeck, Germany) or paediatric (maxi-
mum volume, 700 ml; Dräger, Lübeck, Germany)
self-inflatable bag for 5 min before intubation.
Specific tidal volumes or minute ventilation were
not defined to experienced anaesthesiologists (resi-
dents; 4th year, n=2) administering anaesthesia.
Applied tidal volume was limited by the size of the
self-inflatable bags; respiration rate was supposed
to be 15 min − 1. Anaesthesiologists were asked to
perform bag–valve–mask ventilation according to
their clinical experience. If oxygen saturation de-
creased below 90%, the experiment was terminated.
Respiratory rate, exhaled tidal volume, and peak
airway pressure were measured with a respiratory
monitor (AS 3 Compact, Datex Ohmeda, Helsinki,
Finland). Before measurement of capillary blood
gas variables, a tip of an ear lobe was arterialised
with vasodilating creme as validated earlier [10,11];
subsequently, capillary blood gas samples were
taken before induction of anaesthesia during spon-
taneous ventilation, and after 5 min of bag– valve–
mask ventilation with room-air. All capillary blood
gases were analysed with a blood gas machine
(ABL 520, Radiometer, Copenhagen, Denmark).
3. Statistical analysis
All analysis were carried out with the Statistical
Package for the Social Sciences (SPSS, Chicago,
Illinois). We chose the Mann–Whitney U-test for
comparison between the two self-inflatable bags.
Thereupon, the Wilcoxon test was used to deter-
mine differences in patient characteristics, blood
gas variables, and respiratory parameters through-
out the experiment. A two-sided P B0.05 was
considered significant.
4. Results
Forty patients were enrolled into the study, and
randomised to receive bag–valve–mask ventilation
with either a paediatric (n=20) or an adult
 V. Dörges et al. / Resuscitation 44 (2000) 37–41 39

Table 1 after 5 min of ventilation with room-air. In

Patient (n=40) variables during bag–valve–mask ventilation patients being ventilated with the paediatric
applied with an adult and paediatric self-inflatable baga
self-inflatable bag, oxygen saturation, partial
Variable Adult bag Paediatric bag P pressure of oxygen, and pH decreased significantly
(n= 20) (n= 20) (PB0.01) after 5 min of ventilation with room-air
compared with baseline (Table 3). No significant
Age (years) 37 93 3692 NS differences were detected in carbon dioxide partial
Weight (kg) 7093 75 93 NS
pressure, and peak airway pressure between both
Height (cm) 171 93 17493 NS
self-inflatable bags during the entire experiment
a
Results are given as mean 9SEM; NS, nonsignificant. (Tables 2 and 3). In 3 of 20 (15%) patients being
ventilated with the paediatric self-inflatable bag,
Table 2 oxygen saturation fell below 90% within the first 3
Tidal volumes, peak airway pressures, and respiratory rates min of the trial. In these patients, the experiment
delivered with an adult versus a paediatric self-inflatable baga was terminated as prescribed by the study protocol
in order to ensure patient safety.
VT VT Paw RR
(ml) (ml kg−1) (cm H2O) (1 min−1)

Adult bag 719922b 10.590.4b 1291 15 9 0.4 5. Discussion

Paediatric 455923 6.290.4 1291 1590.6
bag
a air with a paediatric self-inflatable bag were able
VT, tidal volume; Paw, peak airway pressure; RR, respira-
tory rate. Data is given as mean 9 SEM.
b
PB0.0001 vs. paediatric bag. In both groups 20 patients
were enrolled.
(n =20) self-inflatable bag. There were no
significant differences in age, weight, or height
between groups (Table 1). When compared with
an adult self-inflatable bag, the paediatric bag
resulted in significantly (P B 0.0001) lower exhaled
tidal volume (Table 2), oxygen saturation, and
partial pressure of oxygen (Table 3). When using
the adult self-inflatable bag, no differences in
blood gases were measured between spontaneous
breathing before induction of anaesthesia, and
Although small tidal volumes containing room-
to maintain sufficient ventilation, oxygenation was
not adequate. On the other hand, larger tidal
volumes containing room-air given with an adult
self-inflatable bag resulted in both adequate venti-
lation and oxygenation throughout the entire trial.
Measuring ventilation and blood gas parameters
in a clinical investigation of basic life support
ventilation during cardiac arrest is extremely
difficult. For example, measuring devices such as
pneumotachometers and blood gas machines must
be carefully calibrated; which is next to impossible
prior to emergent airway management at the
scene. Also, life-saving interventions in patients
suffering of respiratory arrest can not be delayed

Table 3
Blood gas variables before induction of anaesthesia, and after 5 min of bag–valve–mask ventilation with room-air using a
paediatric versus an adult self-inflatable baga

Variable Before induction of anaesthesia After 5 minutes of BVM ventilation

Adult bag Paediatric bag Adult bag Paediatric bag

Number (n) 20 20 20 17
SaO2 (%) 9690.3 96 9 0.3 96 9 0.4 939 1.0b,c
paO2 (mmHg) 8593 84 92 8794 739 4b,c
paCO2 (mmHg) 399 1 3891 379 1d 4092
pH 7.44 9 0.00 7.45 90.00 7.4590.00 7.4190.00c

a
Data is given as mean 9 SEM; BVM indicates bag–valve–mask.
b
PB0.01 vs. after 5 min of ventilation with an adult self-inflatable bag.
c
PB0.01 vs. before induction of anaesthesia with a paediatric self-inflatable bag.
d
PB 0.05 vs. before induction of anaesthesia with an adult self-inflatable bag.
40 V. Dörges et al. / Resuscitation 44 (2000) 37–41
until study equipment is available, and ready to
utilise [12]. In addition, our patients were not
hypoxic and/or hypercarbic; but were premedi-
cated and supine, and awaiting induction of
anaesthesia breathing spontaneously room-air.
Further, respiratory system compliance, airway
resistance, and lower oesophageal sphincter pres-
sure in anaesthetised patients nicely represent,
for example, an emergency patient outside the
hospital with respiratory arrest [13,14]. Although
this setting does not entirely reflect blood gas
and metabolism status of an untreated apnoeic
patient, we suggest that the present study may
be the best possible model to simulate basic life
support ventilation with a minimum of con-
founding variables.
Self-inflatable bags may be more readily avail-
able during an emergency than additional oxy-
gen, especially in remote locations such as
general practitioners offices, general wards in
hospitals, and possibly, some situations in the
emergency medical service providing basic life
support. This has an important impact on venti-
lation strategies. The European Resuscitation
Council has recommended that volumes should
be reduced from 800 to 1200 ml ( : 12 ml
kg − 1), as recommended by the American Heart
Association [1], to 500 ml (: 6 ml kg − 1) [2] in
an effort to minimise gastric inflation [3,15]. This
approach has been verified in a bench model
simulating respiratory mechanics during both
respiratory and cardiac arrest, when smaller tidal
volumes of :500 ml decreased peak airway
pressure significantly, which, in turn, decreased
the risk of gastric inflation with subsequent ven-
tilation-related complications [5]. Interestingly, it
has never been proven whether smaller tidal vol-
umes of : 6 ml kg − 1 are sufficient when provid-
ing ventilation and oxygenation with room-air.
For example, an animal model showed that tidal
volumes of :6 ml kg − 1 simulating mouth-to-
mouth ventilation with a gas composition of 4%
carbon dioxide and 17% oxygen [16] were not
able to maintain arterial oxygenation when de-
saturation to : 77% occurred [17]. Actually,
oxygen saturation in that study was located in
the steep portion of the haemoglobin–oxygen
saturation curve, resulting in a rapid decline in
oxygen saturation and subsequently, cardiovascu-
lar collapse in some pigs. Hence, it is extremely
important to determine whether tidal volumes of
:6 ml kg − 1 with room-air may be associated
with similar problems, or safe to administer dur-
ing basic life support ventilation.
Although the haemoglobin–oxygen saturation
relation in most of our patients was usually lo-
cated in the flat portion of the curve, indicating
some ‘reserve’ in terms of speed of oxygen desat-
uration, tidal volumes of :6 ml kg − 1 were not
sufficient to maintain adequate oxygenation.
Since carbon dioxide partial pressure did not in-
crease, only partial respiratory failure occurred,
although a moderate decline in arterial pH was
observed. This may indicate that when adminis-
tering tidal volumes of 6 ml kg − 1 in order to
minimise the risk of gastric inflation, additional
oxygen is definitely necessary at the earliest point
in time. This may be confirmed by the oxygen
saturation in 3 of 20 patients in the paediatric
self-inflatable bag group. Despite comparable
tidal volumes, oxygen saturation fell below 90%
rapidly after initiation of the study, and forced
the investigators to conclude the experimental
protocol in order to ensure patient safety. Ac-
cordingly, the 5-min blood gas variables of this
cohort are not included in this analysis, but
would have rendered differences in blood gas
variables even more significant. Since these three
patients did not belong to risk groups with asso-
ciated ventilation problems such as obesity,
toothlessness, or presence of beards, our study
may confirm that oxygen administration is one
of the most important interventions in ventila-
tion-associated emergencies.
Some limitations of the present study should be
noted. First, only healthy ASA I and ASA II
patients without respiratory and/or cardiac
pathology were enrolled into the study. Second,
although airway resistance and respiratory system
compliance in our experiment were similar to an
apnoeic or cardiac arrest patient, [13,18,19] lower
oesophageal sphincter pressure during cardiac ar-
rest is fundamentally decreased [20]. Accordingly,
exhaled tidal volumes given with bag–valve –mask
ventilation are most likely even smaller during
basic life support of a cardiac arrest victim than in
our experiment, which was indicated by bench
models simulating basic life support ventilation
during cardiac arrest [9,21]. Our results may be
safely extrapolated to basic life support manage-
 V. Dörges et al. / Resuscitation 44 (2000) 37–41
ment of patients with respiratory arrest only. It is
difficult to extrapolate our results to patients in
cardiac arrest due to significant differences in un-
derlying mechanisms such as pulmonary blood
flow, and cardiac output.
In conclusion, our results indicate that smaller
tidal volumes of :6 ml kg − 1 (: 500 ml) given
with a paediatric self-inflatable bag and room-air
maintain adequate ventilation, but do not result in
sufficient oxygenation during bag–valve–mask
ventilation. Thus, if small (6 ml kg − 1) tidal vol-
umes are being used during bag–valve–mask ven-
tilation, additional oxygen is necessary.
Accordingly, during bag–valve–mask ventilation
in apnoeic patients when additional oxygen is not
available, only large tidal volumes of :11 ml
kg − 1 are able to maintain sufficient oxygenation.
Acknowledgements
Supported, in part, by the Department of
Anaesthesiology, University of Lübeck, Germany.
We are indebted to Susanne Flaig RN, and Rolf
Müller RN, for technical assistance and support
throughout the study.
References
[1] Emergency Cardiac Care Committee and Subcommit-
tees, American Heart Association. Guideline for car-
diopulmonary resuscitation and emergency cardiac care,
part II: adult basic life support. J Am Med Assoc
1992;268:2184–98.
[2] Airway and Ventilation Management Working Group of
the European Resuscitation Council. Guidelines for the
basic management of the airway and ventilation during
CPR. Resuscitation 1996;31:187–200.
[3] Baskett P, Nolan J, Parr M. Tidal volumes which are
perceived to be adequate for resuscitation. Resuscitation
1996;31:231–4.
[4] Krischer JP, Fine EG, Davis JH, Nagel EL. Complica-
tions of cardiac resuscitation. Chest 1987;92:287– 91.
[5] Wenzel V, Idris AH, Banner MJ, Kubilis PS, Williams
JL. Influence of tidal volume on the distribution of gas
between the lungs and the stomach in the nonintubated
patient receiving positive pressure ventilation. Crit Care
Med 1998;26:364–8.
[6] Wenzel V, Idris AH. The current status of ventilation
strategies during cardiopulmonary resuscitation (CPR).
Curr Opin Crit Care 1997;3:206–13.
41
[7] Ruben H, Knudsen EJ, Carugati G. Gastric inflation
in relation to airway pressure. Acta Anaesth Scand
1961;5:107 – 14.
[8] Wenzel V, Keller C, Idris AH, Dörges V, Lindner KH,
Brimacombe JR. Effects of smaller tidal volumes dur-
ing basic life support ventilation in patients with res-
piratory arrest: good ventilation, less risk?
Resuscitation 1999;43:25 – 9.
[9] Dörges V, Sauer C, Ocker H, Wenzel V, Schmucker P.
Smaller tidal volumes during cardiopulmonary resusci-
tation: comparison of adult and paediatric self-inflat-
able bags with three different ventilatory devices’.
Resuscitation 1999;43:31 – 7.
[10] Begin R, Racine T, Roy JC. Value of capillary blood
gas analyses in the management of acute respiratory
distress. Am Rev Respir Dis 1975;112(6):879– 81.
[11] Spiro SG, Dowdeswell IR. Arterialized ear lobe blood
samples for blood gas tensions. Br J Dis Chest
1976;70(4):263– 8.
[12] Dick WF, Brambrink AM, Kern T. Kardiopulmonale
Reanimation ‘oben ohne’? Anaesthesist 1999;48:290–
300.
[13] Safar P, Agusto-Escarraga L. Compliance in anes-
thetized adults. Anaesthesiology 1959;20:283 – 9.
[14] Wenzel V, Idris AH, Lindner KH. Ventilation with an
unprotected airway during cardiac arrest. In: Vincent
JL, editor. Yearbook of Intensive Care and Emergency
Medicine. Berlin, Heidelberg: Springer, 1997:483–92.
[15] Wenzel V, Idris AH, Banner MJ, Kubilis PS, Band R,
Williams JL Jr, Lindner KH. Respiratory system com-
pliance decreases after cardiopulmonary resuscitation
and stomach inflation: impact of large and small tidal
volumes on calculated peak airway pressure. Resuscita-
tion 1998;38:113 – 8.
[16] Wenzel V, Idris AH, Banner MJ, Fuerst RS, Tucker
KJ. The composition of gas given by mouth-to-mouth
ventilation during CPR. Chest 1994;106:1806– 10.
[17] Idris AH, Gabrielli A, Caruso L. Smaller tidal volume
is safe and effective for bag – valve – mask ventilation,
but not for mouth-to-mouth ventilation: an animal
model of basic life support ventilation. Circulation
1999;100(Suppl I):I-644/3499 (Abstract).
[18] Weiler N, Heinrichs W, Dick W. Assessment of pul-
monary mechanics and gastric inflation pressure during
mask ventilation. Prehosp Disaster Med 1995;10:101–5.
[19] Davies K Jr, Johannigman JA, Johnson RC Jr, Bran-
son RD. Lung compliance following cardiac arrest.
Acad Emerg Med 1995;10:855 – 6.
[20] Bowman FP, Menegazzi JJ, Check BD, Duckett TM.
Lower esophageal sphincter pressure during prolonged
cardiac arrest and resuscitation. Ann Emerg Med
1995;26:215 – 9.
[21] Doerges V, Sauer C, Ocker H, Wenzel V, Schmucker
P. Airway management during cardiopulmonary resus-
citation: a comparison of bag – valve – mask, laryngeal
mask and combitube. Resuscitation 1999;41:63 –9.