The Effect of Positive End-Expiratory Pressure Level on Peak
Expiratory Flow During Manual Hyperinflation
Camila Savian, BPHty*, Pamela Chan,
Jennifer Paratz, MPHty, PhD, FACP*‡
BPHty, MPHtyStud†‡,
and
*Alfred Hospital/La Trobe University, Melbourne, †Prince of Wales Hospital, Hong Kong, ‡University of Queensland,
Australia
Including positive end-expiratory pressure (PEEP) in the
manual resuscitation bag (MRB) may render manual hyperinflation (MHI) ineffective as a secretion maneuver
technique in mechanically ventilated patients. In this
study we aimed to determine the effect of increased PEEP
or decreased compliance on peak expiratory flow rate
(PEF) during MHI. A blinded, randomized study was
performed on a lung simulator by 10 physiotherapists experienced in MHI and intensive care practice. PEEP levels
of 0 –15 cm H2O, compliance levels of 0.05 and 0.02 L/cm
H2O, and MRB type were randomized. The Mapleson-C
MRB generated significantly higher PEF (P ⬍ 0.01, d ⫽
M
anual hyperinflation (MHI) is frequently used
by intensive care staff (1,2). This procedure is
believed to increase the passive inflation of the
lungs and the expiratory flow rate. It improves static
and dynamic compliance (2– 4), increases the volume
of secretions suctioned (4), and causes a trend towards
a decrease in ventilator-associated pneumonia (5).
One of the main uses is for the application of an
“artificial cough,” through the generation of a large
tidal volume (Vt) and a peak expiratory flow (PEF)
high enough to mobilize secretion, to prevent sputum
plugging and subsequent nosocomial pneumonia. Removal of secretions in tracheally intubated patients
has been demonstrated to occur via a process of annular two-phase gas-liquid flow (6). The criteria for
this has been suggested to be an inspiratory flow rate
slower than expiratory flow rate and a larger Vt (7),
producing a decrease of pressure between the alveolus
and the mouth during the gas liquid interaction. One
Accepted for publication September 21, 2004.
Address correspondence and reprint requests to C. Savian, Cardiopulmonary Research Unit, 4th Floor, Philip Block, Alfred Hospital, Commercial Rd, Prahran 3181, Victoria, Australia. Address
e-mail to Camila_Savian@hotmail.com.
DOI: 10.1213/01.ANE.0000147505.98565.AC
1112
Anesth Analg 2005;100:1112–6
2.72) when compared with the Laerdal MRB for all levels
of PEEP. In normal compliance (0.05 L/cm H2O) there
was a significant decrease in PEF (P ⬍ 0.01, d ⫽ 1.45) for a
PEEP more than 10 cm H2O in the Mapleson-C circuit. The
Laerdal MRB at PEEP levels of more than 10 cm H2O did
not generate a PEF that is theoretically capable of producing two-phase gas-liquid flow and, consequently, mobilizing pulmonary secretions. If MHI is indicated as a result
of mucous plugging, the Mapleson-C MRB may be the
most effective method of secretion mobilization.
(Anesth Analg 2005;100:1112–6)
of the factors that are hypothesized to influence this
decrease is the level of positive end-expiratory pressure (PEEP).
PEEP is currently used in mechanically ventilated
intensive care patients to re-expand atelectasis (8),
improve oxygenation (9), reduce intrapulmonary
shunt fraction, and protect the lungs against shear
stress (10). “Protective lung” and “open lung” ventilation, i.e., small Vt and increased levels of PEEP, are
currently accepted as the most effective ventilatory
strategies (10), and there is controversy as to whether
the patient should be disconnected from high levels of
PEEP to provide MHI, as disconnection may result in
a decrease in functional residual capacity (FRC) (11),
decrease in oxygenation (12), and potential shear
stress of distal lung units (13).
To overcome these adverse effects, a valve to maintain PEEP may be connected to the manual resuscitation bag (MRB) (14). However, high levels of PEEP
may decrease the alveolus-to-mouth pressure gradient
and consequently decrease the PEF to a level where
MHI may not be effective as a secretion maneuver
technique. It is therefore important to investigate
whether different levels of PEEP alter the effectiveness
of MHI as a secretion removal technique and whether
it is justified in disconnecting the patient from high
levels of PEEP.
©2005 by the International Anesthesia Research Society
0003-2999/05
ANESTH ANALG
2005;100:1112–6
CRITICAL CARE AND TRAUMA
SAVIAN ET AL.
MANUAL HYPERINFLATION AND PEEP
1113
Table 1. Demographic Information
ID
Age (yr)
Sex
1
2
3
4
5
6
7
8
9
10
25
27
43
25
34
29
33
23
27
29
F
F
F
F
F
F
M
F
F
F
Years of experience
1
2
20
1
6
3
3
6
2
7
year and 6 months
years
years
year
years
months
years
months
years
years
Years since graduation
2
6
21
3
8
3
3
1
4
7
years
years
years
years
years
years and 6 months
years
years and 6 months
years
years
Circuit most commonly used
Laerdal
Mapleson-C
Mapleson-C
Laerdal
Laerdal
Magill/Mapleson
Magill/Mapleson
Magill
Laerdal
Laerdal
F ⫽ female; M ⫽ male.
The aim of this study was to compare the PEF and
Vt generated in a lung model at 6 different levels of
PEEP and 2 levels of lung compliance, 0.05 L/cm H2O
to represent normal lung compliance (15) and 0.02
L/cm H2O to simulate low compliant lungs such as
those of acute respiratory distress syndrome (ARDS)
patients (16). Two types of MRB were investigated:
a Mapleson-C circuit and the Laerdal self-inflating
resuscitator.
Methods
Ten physiotherapists, experienced in MHI and intensive care practice, from The Alfred Hospital (Melbourne, Australia) volunteered to participate in this
study. Demographics are described in Table 1. Informed consent was obtained from each subject, and
the Alfred Hospital and LaTrobe University Ethics
Committee approved the study.
The 2-L Mapleson-C antistatic re-breathing circuit
(BS 3352) and the 1.6-L Laerdal self-inflating resuscitation bag were connected via a Pneumotach to a
test-training lung (MI Instruments Inc). The testtraining lung was connected to a respiratory mechanics monitor (Cosmo Model 8000; Novametrix Medical
Systems, Inc. Wallingford, CT). Data were downloaded to a personal computer and analyzed by the
Analysis Plus PC program (Novametrix Medical
Systems, Inc.) (Fig. 1).
A portable spring-loaded PEEP valve (AMBU®) was
connected to the MRB and a constant oxygen flow rate
of 15 L/min was delivered. Lung compliance of 0.05
L/cm H2O and 0.02 L/cm H2O were chosen to simulate normal (15) and low (16) compliance lung conditions, respectively. An airway resistance of 2.3 ⫾ 0.5
cm H2O/s was standardized for all measures in this
study.
All subjects were instructed to perform MHI as if
trying to promote secretion clearance. A 2-s deep inspiration, an inspiratory plateau of 2 s, and a fast
release (1 s) were requested. The respiratory wave
forms were examined and only the breaths that fulfilled this pattern were included in the analysis. Peak
Figure 1. Illustration of the lung simulator and connections used
during the study. 1, positive end-expiratory pressure valve; 2, Laerdal resuscitation bag; 3, pressure and volume sensor connected to
the respiratory mechanics monitor; 4, pressure sensor connected to
the lung simulator’s manometer; 5, resistances of 2.3 cm H2O/s; 6,
compliance adjustment piece; 7, computer with analysis plus software; 8, respiratory mechanics monitor.
inspiratory pressure (PIP) was standardized to maximum of 35 cm H2O using a pressure manometer. The
expiratory valve of the Mapleson-C circuit was adjusted to the fully open position but manually held
closed during inspiration, then released to the fully
open position during expiration.
A trial for 60 s was allowed on each circuit before
testing. Five breaths were given in random order (using sealed envelopes) for both bagging circuit and
each level of PEEP (0, 5, 7.5, 10, 12.5, and 15 cm H2O).
For PEEP levels of 0, 5, and 7.5 cm H2O, compliance of
0.05 L/cm H2O was standardized, whereas for PEEP
levels of 10, 12.5, and 15 cm H2O, MHI was performed
at both compliance settings of 0.05 and 0.02 L/cm
H2O. The physiotherapist performing MHI was
blinded as to the level of PEEP and compliance in the
circuit. The PEF, Vt, PIP, and PEEP level were recorded by the respiratory mechanics monitor. The
cross-sectional area of the tube (CSA) was 2.2 cm2.
1114
CRITICAL CARE AND TRAUMA SAVIAN ET AL.
MANUAL HYPERINFLATION AND PEEP
ANESTH ANALG
2005;100:1112–6
Table 2. The Peak Expiratory Flow Generated by Both Types of Circuit and Compliance Settings
Laerdal circuit
Mapleson-C circuit
PEEP
0.05 L/cmH2O
0.02 L/cmH2O
0.05 L/cmH2O
0.02 L/cmH2O
0
5
7.5
10
12.5
15
Mean
sd
28.09 ⫾ 3.11
26.46 ⫾ 2.57
25.85 ⫾ 2.98
22.08 ⫾ 6.16
25.69 ⫾ 6.38
20.69 ⫾ 6
24.81
2.82
NA
NA
NA
35.86 ⫾ 6.39
37.18 ⫾ 9.47
32.83 ⫾ 9.28
35.29
2.23
42.85 ⫾ 6.79
41.04 ⫾ 5.98
36.91 ⫾ 7.41
31.99 ⫾ 8.09
30.14 ⫾ 5.67
32.03 ⫾ 6.59
35.83
5.28
NA
NA
NA
40.97 ⫾ 9.44
41.51 ⫾ 5.09
38.24 ⫾ 6.17
40.24
1.75
Values are mean ⫾ sd (L/min).
PEEP ⫽ positive end-expiratory pressure; NA ⫽ not applicable.
Data were extracted from each waveform using the
Analysis Plus software, copied into an Excel spreadsheet and analyzed by the SPSS program (version 11.5
for Windows; SPSS, Chicago, IL). As the scores did not
generate a normal distribution, a base-10 logarithm
was used to transform all the dependent variables.
Using the log-transformed variables, a general linear model, which is an extension of the multivariate
analysis of variance, was performed for the dependent
variables PEF and Vt, and for the independent variables PEEP level, compliance, and type of circuit. The
analysis included all the main effects and also second
level interactions. If this proved significant, a post hoc
multiple comparisons for observed means was performed using the least significant difference comparison. In addition, a correlation analysis matched PEF
and both PIP and Vt.
Sample size calculation was based on PEF using the
mean difference and standard deviation from a previous
study (17). For the given effect size, ␣ ⫽ 0.05 (two-tailed)
and power of 90%, the sample size estimated was 10
subjects. Data are presented as mean values and standard deviation unless otherwise stated.
Results
The sample consisted of nine subjects. Data from subject number 5 were excluded because technical problems occurred while recording the data. There was a
medium positive correlation between PEF and PIP in
the Laerdal circuit (r ⫽ 0.37, P ⬍ 0.001) and a small
positive correlation (r ⫽ 0.27, P ⬍ 0.02) in the
Mapleson-C circuit. There was no significant correlation between PEF and Vt in either circuit.
When compliance was set at 0.05 L/cm H2O for the
Laerdal circuit, there was a significant decrease in PEF
between levels 0 and 15 cm H2O of PEEP, with effect
size (d) of 1.2 and P ⫽ 0.032. For the Mapleson-C
circuit, a significant decrease in PEF was observed
between 0 and every PEEP level equal to or more than
10 cm H2O (d ⫽ 1.45, P ⫽ 0 ⬍ 0.01)(Table 2).
At a compliance of 0.02 L/cm H2O, no significant
decrease in PEF was observed between 10, 12.5, and 15
cm H2O of PEEP for either circuit; however, the
Mapleson-C circuit generated significantly higher PEF
(d ⫽ 2.21, P ⫽ 0.03) when compared with the Laerdal
(Table 2).
The PEF was significantly higher with a compliance
of 0.02 L/cm H2O for both circuits at all levels of PEEP
(d ⫽ 0.88 and P ⬍ 0.01). In addition, the peak inspiratory flow to PEF ratio (PIF/PEF) generated by the
Mapleson-C circuit (1.14 ⫾ 0.43) was significantly
lower when compared with the Laerdal resuscitation
bag (1.70 ⫾ 0.91; d ⫽ 0.8, P ⬍ 0.01).
The Mapleson-C circuit produced significant larger
Vt (1.560.3 ⫾ 637.5) when compared with the Laerdal
circuit (1.031 ⫾ 187; d ⫽ 1.28, P ⬍ 0.01). For both
resuscitation bags, Vt was significantly larger when
compliance was set at 0.05 L/cm H2O (d ⫽ 0.89,
P ⬍ 0.01).
Discussion
There is controversy as to whether it is advantageous
to disconnect patients on PEEP to administer MHI for
secretion removal. This study aimed to investigate the
influence of six different levels of PEEP on the PEF
and Vt generated by two different circuit types frequently used by physiotherapists (18).
This study has shown that, in a lung model, the
Mapleson-C circuit generates a PEF that is theoretically capable of annular 2-phase gas-liquid flow at all
PEEP levels up to 15 cm H2O. The Laerdal circuit,
according to this theoretical model, may not be effective when using PEEP levels more than 10 cm H2O of
PEEP. This study is the first to investigate the effect of
PEEP levels on PEF, and a significant reduction in PEF
was observed in both circuits as the level of PEEP
increased.
Increasing numbers of intensive care patients are
mechanically ventilated on high levels of PEEP with
small Vts. If the critical care practitioner considers
ANESTH ANALG
2005;100:1112–6
that MHI is clinically indicated, a PEEP valve is usually connected to the MRB to prevent shear stress and
decreased oxygenation and FRC during disconnection
from the ventilator. A survey conducted by Hodgson
et al. (18) in 32 Australian intensive care units showed
that 66% of physiotherapists would choose to include
PEEP valve in the bagging circuit if the patient was
ventilated with PEEP. However, high levels of PEEP
may decrease alveolus-to-mouth pressure gradient
and consequently decrease the expiratory flow rate,
rendering MHI ineffective as a secretion maneuver
technique.
Maxwell and Ellis (7), in a review of effective secretion mobilization during mechanical ventilation and
MHI, have stated that one of the conditions necessary
to remove secretion during manual hyperinflation is
PEF ⱖ0.41 L/s (24.6 L/min). In the current study, the
PEF generated did not meet this condition using the
Laerdal circuit at a PEEP level of 10 and 15 cm H2O
with the normal compliance setting (0.05 L/cm H2O),
suggesting that the use of the Laerdal MRB may not be
the best option in patients on high levels of PEEP
when the aim of the therapy is secretion mobilization.
Another important factor responsible for promoting
secretion clearance is to generate PIF slower than PEF
(7). The Mapleson-C circuit generated significantly
lower PIF to PEF ratio than the Laerdal circuit. These
results confirm the findings of Maxwell and Ellis (19).
The Mapleson-C circuit applied significantly higher
PEF, Vt, and PIP when compared with the Laerdal
circuit. There are important differences between these
two bag types that may explain these results. The
Laerdal circuit has a 1.6-L reservoir bag, whereas the
Mapleson-C circuit has a 2-L bag. The Mapleson-C
circuit’s rubber bag is more compliant than the silicone Laerdal bag. Also, when using the Laerdal circuit, the inspiratory fishmouth valve results in leak of
volume and airway pressure during the end of the
inspiratory plateau (20), whereas when using the
Mapleson-C circuit, the therapists need to hold the
expiratory valve during inspiration and also during
the inspiratory plateau to fill the bag, allowing the
airway pressure and volume to increase until the end
of inspiration.
Setting the test lung to a low compliance (0.02 L/cm
H2O) resulted in increased airway pressure and a
decrease in Vt during MHI, concurring with previous
results (20,21). The PEF was demonstrated to be significantly correlated with PIP, and in the current
study, PEF was greater in lungs of poor compliance
than normal compliance, although PIP was limited to
35cm H2O. This suggests that in patients with poor
compliance (e.g., ARDS), MHI is effective as a secretion maneuver technique.
Volume restoration has been considered to play an
important part in secretion clearance, and impairment
in mucociliary clearance has been observed in reduced
CRITICAL CARE AND TRAUMA
SAVIAN ET AL.
MANUAL HYPERINFLATION AND PEEP
1115
lung volumes (22). In the current study, significantly
smaller Vts were generated when low compliance
was used, in accordance with the study by Rusterholz
and Ellis (23).
Kim et al. (24) demonstrated that gas velocity
slower than a cough may remove secretion via annular two-phase gas-liquid flow. As the linear velocity is
determined by the expiratory flow rate divided by the
total CSA of the airways, the effective velocity should
decrease with increasing airway generations as the
total CSA increases. The more distal the airway generation, the slower the linear velocity generated, but
also the slower the critical airflow rate required to
produce upward movement in a viscoelastic liquid (7).
In the current study, the CSA of the tube was constant
at 2.2 cm and it is difficult to draw conclusions regarding the effect of MHI in distal airway generations.
Patients receive high PEEP for various reasons, including intrapulmonary and extrapulmonary ARDS,
pulmonary edema, and impaired gas exchange. MHI
would not be advantageous in all these situations.
There is also controversy over the disconnection of
patients on high levels of PEEP, as it may cause a
decrease in FRC and shear stress. Inline suctioning
devices are in common use and one of the advantages
of them is that disconnection of ventilation is not
needed for suction. This study was a bench study
only, but it does suggest that if mucous plugging was
considered the cause of poor gas exchange, MHI with
a Mapleson -C circuit and PEEP valve would generate
a PEF capable of mobilizing secretions. However, this
same circuit may potentially cause complications, including barotrauma or volutrauma, resulting from
high inspiratory pressure and volume generated.
The current study demonstrated that the Laerdal
resuscitation bag did not generate PEF that theoretically would clear pulmonary secretions when PEEP
levels of 10 and 15 cm H2O were used. Clinical studies
are necessary to define if the MHI technique is effective in more distal airway generations when a PEEP
valve is used and also to confirm these laboratory
findings in clinical settings, especially considering the
heterogeneity of lung disease in critically ill patients.
References
1. King D, Morrell A. A survey on manual hyperinflation as a
physiotherapy technique in intensive care units. Physiotherapy
1992;78:747–50.
2. Jones AYM, Hutchinson RC, Oh TE. Effects of bagging and
percussion on total static compliance of the respiratory system.
Physiotherapy 1992;78:661– 6.
3. Paratz J, Lipman J, McAuliffe M. Effect of manual hyperinflation
on hemodynamics, gas exchange, and respiratory mechanics in
ventilated patients. J Intensive Care Med 2002;17:317–24.
4. Hodgson C, Denehy L, Ntoumenopoulos G, et al. An investigation of the early effects of manual lung hyperinflation in critically ill patients. Anaesth Intensive Care 2000;28:255– 61.
1116
CRITICAL CARE AND TRAUMA SAVIAN ET AL.
MANUAL HYPERINFLATION AND PEEP
5. Ntoumenopoulos G, Gild A, Cooper DJ. The effect of manual
lung hyperinflation and postural drainage on pulmonary complications in mechanically ventilated trauma patients. Anaesth
Intensive Care 1998;26:492– 6.
6. Kim CS, Rodriguez CR, Eldrigde MA, Sackner MA. Criteria for
mucus transport in the airways by two-phase gas-liquid flow
mechanism. J Appl Physiol 1986;60:901–7.
7. Maxwell L, Ellis E. Secretion clearance by manual hyperinflation:
Possible mechanisms. Physiotherapy Theory and Practice 1998;14:
189 –97.
8. Rothen HU, Sporre B, Wegenius G, Hedenstierna G. Reexpansion of atelectasis during general anaesthesia may have a prolonged effect. Acta Anaesthesiol Scand 1995;39:118 –25.
9. Claxton BA, Morgan P, Mckeague H, et al. Alveolar recruitment
strategy improves arterial oxygenation after cardiopulmonary
bypass. Anaesthesia 2003;58:111– 6.
10. Amato MBP, Barbas CSV, Medeiros DM, et al. Beneficial effects
of “Open Lung Approach” with low distending pressures in
acute respiratory distress syndrome. Am J Respir Crit Care Med
1995;152:1835– 46.
11. Hedenstierna G, Tokics L, Lundquist H, et al. Phrenic nerve
stimulation during halothane anesthesia. Effects of atelectasis.
Anesthesiol 1994;80:751– 60.
12. Lindberg P, Gunnarsson L, Tokics L. Atelectasis and lung function in the postoperative period. Acta Anaesthesiol Scand 1992;
36:546 –53.
13. McCann UG, Schiller HJ, Carney DE, et al. Visual validation of
the mechanical stabilizing effects of positive end-expiratory
pressure at the alveolar level. J Surg Res 2001;99:335– 42.
14. Boidin MP. A portable PEEP valve for 0 –20 cm H2O. Acta
Anaesthesiol Belg 1982;33:69 –74.
ANESTH ANALG
2005;100:1112–6
15. Rossi A, Gottfried SB, Zocchi L. Measurement of static compliance of the total respiratory system in patients with acute respiratory failure during mechanical ventilation. Respir Care
1989;34:191–5.
16. Matamis D, Lemaire F, Harg A, et al. Total respiratory pressurevolume in the ARDS. Chest 1984;86:58 – 66.
17. Jones A, Hutchinson RC, Lin E, Oh T. Peak expiratory flow rates
produced with Laerdal and Mapleson-C bagging circuits. Aust
J Physiother 1992;38:211–5.
18. Hodgson C, Carroll S, Denehy L. A survey of manual hyperinflation in Australian hospitals. Aust J Physiother 1999;45:185–93.
19. Maxwell LJ, Ellis ER. The effect of circuit type, volume delivered
and “rapid release” on flow rates during manual hyperinflation.
Aust J Physiother 2003;49:31– 8.
20. McCarren B, Chow CM. Manual hyperinflation: A description
of the technique. Aust J Physiother 1996;42:203– 8.
21. Hess D, Simmons M, Blakovitch S, et al. An evaluation of the
effects of fatigue, impedance, and use of two hands on volumes
delivered during bag valve ventilation. Respiratory Care 1993;
38:271–5.
22. Gasmu G, Singer MM, Vincent HH, et al. Postoperative impairment of mucous transport in the lung. Am Rev Respir Dis
1976;114:673–9.
23. Rusterholz B, Ellis E. The effect of lung compliance and experience on manual hyperinflation. Aust J Physiother 1998;44:
23– 8.
24. Kim CK, Greene MA, Sankaran S, Sackner MA. Mucus transport
in the airways by two-phase gas-liquid flow mechanism: Continuous flow model. J Appl Physiol 1986;60:908 –17.