(Chest. 2000;118:795-807.)
© 2000
American College of Chest Physicians
High-Frequency Ventilation for Acute Lung Injury and ARDS*
Jerry A. Krishnan, MD and
Roy G. Brower, MD
*
From the Department of Medicine, Johns Hopkins University, Baltimore, MD.
Correspondence to: Jerry A. Krishnan, MD, Division of Pulmonary and Critical Care Medicine, Johns Hopkins University, 600 N Wolfe St, Baltimore, MD 21287; e-mail: satish{at}welch.jhu.edu
 |
Abstract
|
|---|
In patients with acute lung injury (ALI) and ARDS,
conventional mechanical ventilation (CV) may cause additional lung
injury from overdistention of the lung during inspiration, repeated
opening and closing of small bronchioles and alveoli, or from excessive
stress at the margins between aerated and atelectatic lung regions.
Increasing evidence suggests that smaller tidal volumes
(VTs) and higher end-expiratory lung volumes (EELVs) may be
protective from these forms of ventilator-associated lung injury and
may improve outcomes from ALI/ARDS. High-frequency ventilation
(HFV)-based ventilatory strategies offer two potential advantages over
CV for pateints with ALI/ARDS. First, HFV uses very small
VTs, allowing higher EELVs with less overdistention than is
possible with CV. Second, despite the small VTs, high
respiratory rates during HFV allow the maintenance of normal or
near-normal PaCO2 levels. In this review, the
use of HFV as a lung protective strategy for patients with ALI/ARDS is
discussed.
Key Words: acute lung injury • ARDS • barotrauma • high-frequency ventilation • high-frequency positive-pressure ventilation • high-frequency jet ventilation • high-frequency oscillation • mechanical ventilation • volutrauma
|
Introduction
|
|---|
Mechanical
ventilation is the cornerstone of supportive care for acute respiratory
failure. In most patients, adequate gas exchange can be ensured while
more specific treatments are administered and natural healing processes
occur. Conventional approaches to mechanical ventilation utilize tidal
volumes (VTs) that are approximately 75 to 150% of the
volumes that patients typically achieve during spontaneous
ventilation.1
While conventional ventilation (CV) usually
provides adequate gas exchange, it is sometimes associated with high
airway pressures, circulatory depression, and pulmonary air leaks.
These adverse effects stimulated the development of high-frequency
ventilation (HFV). There was great enthusiasm for HFV during its early
development in the 1970s and 1980s. Scores of studies in animals and
humans were conducted to understand the physiology of gas
exchange2
3
4
5
and its effects on circulation and other
systems6
7
and to improve the techniques of HFV. However,
the initial enthusiasm for HFV waned as clinical studies failed to
demonstrate important advantages over CV.8
9
10
There is now renewed interest in HFV because of increasing evidence
that (1) CV may contribute to lung injury in patients with acute lung
injury (ALI) and ARDS,11
and (2) modifications of
mechanical ventilation techniques may prevent or reduce lung injury and
improve clinical outcomes in these patients.12
13
14
The
primary objective of this review is to discuss the potential role of
HFV for achieving adequate gas exchange while protecting the lung
against further injury in patients with ALI/ARDS. The review begins
with an overview of HFV techniques and a summary of potential
mechanisms of gas transport. The rationale for the use of HFV-based
lung protective strategies in the management of patients with ALI/ARDS
will be explained. The results of animal and human studies evaluating
HFV for ALI/ARDS will be discussed.
|
HFV
|
|---|
HFV is a mode of mechanical ventilation that uses rapid
respiratory rates (respiratory rate [f] more than four
times the normal rate) and small VTs. Although
HFV VTs are often smaller than traditional
estimates of both anatomic and physiologic dead space, adequate
oxygenation and ventilation usually can be achieved.15
There are numerous variations of HFV. These may be broadly classified
as high-frequency positive pressure ventilation (HFPPV), high-frequency
jet ventilation (HFJV), and high-frequency oscillation (HFO). These
classes of HFV are compared in
Table 1
and are briefly discussed herein. More detailed descriptions can
be found in other reviews.15
16
HFPPV was introduced by Oberg and Sjöstrand17 in 1969
to eliminate the effect of respiratory variations in thoracic volume
and pressure on carotid sinus reflexes.16
HFPPV delivers
small VTs (approximately 3 to 4 mL/kg) of conditioned gas
at high flow rates (175 to 250 L/min) and frequency (f, 60
to 100 breaths/min). The precise VT is difficult
to measure during HFPPV because some gas flows through the expiratory
conduit during inspiration. Expiration is passive and depends on lung
and chest wall elastic recoil. Thus, with high f, there is a
risk of gas trapping with overdistention of some lung regions and
adverse circulatory effects. HFPPV was used primarily in situations
requiring minimal upper airway movement, such as laryngoscopy,
bronchoscopy, and laryngeal surgery.18
Sanders19
introduced HFJV in 1967 to facilitate gas
exchange during rigid bronchoscopy. In HFJV, gas under high pressure
(15 to 50 lb per square inch) is introduced through a small-bore
cannula or aperture (14 to 18 gauge) into the upper or middle portion
of the endotracheal tube (Fig 1
).20
Pneumatic, fluid, or solenoid valves control the
intermittent delivery of the gas jets. Aerosolized saline solution in
the inspiratory circuit is used to humidify the inspired air. Some
additional gas is entrained during inspiration from a side port in the
circuit. This form of HFV generally delivers a VT of 2 to 5
mL/kg at a f of 100 to 200 breaths/min. The jet pressure
(which determines the velocity of air jets) and the duration of the
inspiratory jet (and, thus, the inspiratory/expiratory ratio [I/E])
are controlled by the operator. Together, the jet velocity and duration
determine the volume of entrained gas. Thus, the
VT is directly proportional to the jet pressure
and I/E. Because the volume of entrained air is not
operator-controlled, it is difficult to manipulate with precision the
VTs delivered during HFJV. The jet pressure and
the duration of the inspiratory jet are adjusted empirically to achieve
adequate ventilation. During HFJV, high inspiratory airflow rates and
the decompression of jet gas prevent optimal humidification and warming
of inspired air, increasing the risk of airway obstruction with
desiccated secretions and epithelial debris.15
21
As with
HFPPV, expiration is passive. Thus, HFJV may cause air trapping.
Lunkenheimer et al22
introduced HFO in 1972. HFO uses
reciprocating pumps or diaphragms. Thus, in contrast to HFPPV and HFJV,
both expiration and inspiration are active processes during HFO. HFO
VTs are approximately 1 to 3 mL/kg at fs
up to 2,400 breaths/min. The operator sets the f, the I/E
(typically approximately 1:2), driving pressure, and mean airway
pressure (MAP). Driving pressure (also known as
power) is determined by the displacement of the
reciprocating pumps or diaphragms. The oscillatory
VTs generated during HFO are directly related to
driving pressures. In contrast, VTs are inversely
related to f, since shorter inspiratory times reduce the
duration of bulk flow of air into the tracheobronchial tree (see the
"Gas Transport During HFV" section). The inspiratory bias flow of
air into the airway circuit is adjusted to achieve the desired MAP, an
important determinant of oxygenation (Fig 2 ). There is no gas entrainment or decompression of gas jets in the
airway, allowing better humidification and warming of inspired air.
This lowers the risks of airway obstruction from desiccated airway
secretions. In addition, active expiration permits better control of
lung volumes than with HFPPV and HFJV, decreasing the risk of air
trapping, overdistention of airspaces, and circulatory depression.
Lower I/Es (1:2 or 1:3) reduce the risk of air trapping. Periodic
assessments of BP and lung volume on chest roentgenograms are used to
identify air trapping. Peak inspiratory airway pressure does not
accurately reflect lung volume or air trapping because inspiratory
airway pressure is substantially greater than alveolar pressure.
|
Gas Transport During HFV
|
|---|
Research directed toward understanding the mechanisms of gas
exchange during HFV have led to many insights into pulmonary
physiology. Several mechanisms of gas mixing may contribute to gas
transport during HFV. These are reviewed briefly here and in Figure 3
.23
More detailed descriptions of the physiology of gas
exchange with HFV are presented elsewhere.5
23
View larger version (95K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3.. Proposed mechanisms of gas transport during HFV.
1 = direct bulk flow; 2 = longitudinal (Taylor) dispersion;
3 = pendeluft 4 = asymmetric velocity profiles; 5 = cardiogenic
mixing; 6 = molecular diffusion. Adapted with permission from
Chang.23
|
|
Direct Bulk Flow
Some alveoli situated in the proximal tracheobronchial tree
receive a direct flow of inspired air. This leads to gas exchange by
traditional mechanisms of convective or bulk flow.
Longitudinal (Taylor) Dispersion
Turbulent eddies and secondary swirling motions occur when
convective flow is superimposed on diffusion. Some fresh gas may mix
with gas from alveoli, increasing the amount of gas exchange that would
occur from simple bulk flow.
Pendeluft
In healthy and, more so, in diseased lungs, the mechanics of air
flow vary among lung regions and units within regions. Variation in
regional airway resistance and compliance cause some regions to fill
and empty more rapidly than others. Some gas may flow between regions
if these characteristics vary among regions that are in close
proximity.
Asymmetric Velocity Profiles
The velocity profile of air moving through an airway under laminar
flow conditions is parabolic. Air closest to the tracheobronchial wall
has a lower velocity than air in the center of the airway lumen. This
parabolic velocity profile is usually more pronounced during the
inspiratory phase of respiration because of differences in flow rates.
With repeated respiratory cycles, gas in the center of the airway lumen
advances further into the lung while gas on the margin (close to the
airway wall) moves out toward the mouth.
Cardiogenic Mixing
Mechanical agitation from the contracting heart contributes to gas
mixing, especially in peripheral lung units in close proximity to the
heart.
Molecular Diffusion
As in other modes of ventilation, this mechanism may play an
important role in mixing of air in the smallest bronchioles and
alveoli, near the alveolocapillary membranes.
|
ALI/ARDS
|
|---|
The next sections review the following topics: (1) pertinent
aspects of ALI/ARDS pathophysiology; (2) evidence that CV may
perpetuate or exacerbate lung injury; and (3) experience with CV-based
and HFV-based strategies to improve outcomes in these conditions.
ALI/ARDS occurs when conditions such as pneumonia, sepsis, or severe
trauma lead to acute inflammation in the pulmonary
parenchyma,24
25
increased pulmonary vascular
permeability,26
and extravasation of proteinaceous fluid
into the pulmonary interstitium and alveoli.27
28
Surfactant production is reduced by injury to type II pneumocytes, and
existing surfactant is inactivated by plasma proteins that leak into
the airspaces.29
30
The loss of surfactant function
increases surface tension at air-fluid interfaces and leads to the
microatelectasis of alveoli and other small airways.31
Although chest radiographs of patients with ALI/ARDS are frequently
interpreted to show diffuse infiltrates, CT images, histologic
sections, and physiologic studies indicate that the lung injury in
these conditions is not uniform.32
33
Some regions are
severely affected by acute inflammation, airspace filling, and
atelectasis, and others appear to be completely spared.
Patients with ALI/ARDS frequently develop acute respiratory failure,
with worsening arterial oxygenation due to intrapulmonary shunt and
ventilation-perfusion (
/
) mismatch. Physiologic dead
space typically is also elevated,34
35
36
which increases
the minute ventilation required to maintain normal arterial
PaCO2 and pH. Mechanical ventilation
is frequently necessary to maintain gas exchange and to allow more time
for specific treatments and natural healing processes. Because lung
injury is patchy, ventilation is distributed unevenly. Lung regions and
units with worse injury have reduced compliance. VTs are
distributed to the less injured, more compliant regions. Numerous
studies have shown that mechanical forces during CV cause or worsen
lung injury under these circumstances. This potential complication is
known as ventilator-associated lung injury (VALI). Moreover, various
proinflammatory mediators may be released when lungs are subjected to
injurious mechanical forces.13
37
38
39
These mediators may
contribute to further injury to the lung and other
organs.13
37
Thus, traditional approaches to mechanical
ventilation in patients with ALI/ARDS may perpetuate lung injury and
contribute to the development of multiorgan dysfunction
syndrome.11
13
|
Mechanisms of VALI in ALI/ARDS
|
|---|
VALI may occur as a result of several mechanisms related to the
uneven distribution of ventilation.40
41
42
First,
ventilation of lung regions with higher compliance may be injured by
excessive regional end-inspiratory lung volumes (EILVs). Second, injury
may occur in small bronchioles when they snap open during inspiration
and close during expiration. Third, pulmonary parenchyma at the margins
between atelectatic and aerated units may be injured by excessive
stress from the interdependent connections between adjacent units.
These last two mechanisms are frequently described with the term
shear forces and may be important mechanisms of lung injury
when ventilation occurs with relatively low end-expiratory lung volumes
(EELVs) in patients with ALI/ARDS.
Injury From Excessive EILVs
The lungs of patients with ALI/ARDS are susceptible to excessive
regional EILV and overdistention injury because VTs are
distributed primarily to the relatively small portions of lung that are
unaffected by the initial injury. Many studies in experimental animals
have demonstrated acute inflammation, increased vascular
permeability,27
28
intra-alveolar
hemorrhage,28
radiographic infiltrates,43
and
hypoxemia28
43
resulting from overdistention of the
healthy lung. Additional studies in animals with experimental ALI have
shown worsening of lung injury or delayed resolution of edema when
there was excessive EILV.11
44
High inspiratory airway pressures (peak and plateau) are commonly
observed and frequently implicated as causes of VALI. However,
excessive lung stretch, rather than pressure, is more likely to be the
injurious force (Fig 4
).45
46
Elevated airway pressures are recognized as markers
of excessive stretch, but high airway pressures without excessive lung
volumes are not injurious to the lung.41
Thus, there is
increasing use of the term volutrauma to refer to the
stretch-induced injury of excessive inspiratory gas
volume.11
46
View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4.. Measures of lung injury according to lung volumes
and airway pressures. Extravascular lung water (Qwl), dry lung weight
(DLW), and albumin space (Alb Sp) adjusted for body weight (BW) in rats
ventilated for 20 min with high airway pressures and high
VTs (HiP-HiV) (using a peak airway pressure of 45 cm
H2O and a VT of approximately 40 mL/kg), low
pressure and high volume (LoP-HiV) (using negative inspiratory pressure
from iron lung to achieve a VT of approximately 44 mL/kg),
and high pressure and low volume (HiP-LoV) (using thoracoabdominal
strapping to achieve a peak airway pressure of 45 cm H2O
and a VT of approximately 19 mL/kg). Higher Qwl, DLW, and
Alb Sp represent measures of greater lung injury. Horizontal dotted
lines represent the upper 95% confidence limit for control values.
Rats ventilated with HiP-HiV and LoP-HiV, but not HiP-LoV, had
significantly more lung injury than control rats (p < 0.01).
** = significant difference between the groups indicated
(p < 0.01). Reproduced with permission from Dreyfuss et
al.45
|
|
Injury From Ventilation at Low EELVs
Positive end-expiratory pressure (PEEP) has lung protective
effects during mechanical ventilation in isolated lungs,40
and in intact41
44
47
48
and open-chest
animals.49
In intact healthy rats, edema and hemorrhage
from ventilation with excessive lung volumes were substantially reduced
when PEEP was used (Fig 5
).48
In a dog lung injury model,44
lung injury
(assessed from lung wet weight to body weight ratio and venous
admixture) was caused by ventilation with large VT and low
PEEP. This injury was reduced in animals ventilated with smaller
VTs and higher PEEPs despite similar EILVs. The effect of
end-expiratory atelectasis on lung injury was evaluated in a rabbit
surfactant-deficient model.50
Rabbits ventilated with
negative end-expiratory pressure demonstrated greater alveolar
capillary permeability, reduced lung compliance, and worse gas exchange
than rabbits ventilated with PEEP.
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5.. Comparison of rat lungs after ventilation at
various airway pressures. Peak airway pressures and PEEPs of 14 cm
H2O and 0 cm H2O, respectively (strategy A,
left), 45 cm H2O and 10 cm H2O,
respectively (strategy B, middle), or 45 cm
H2O and 0 cm H2O, respectively (strategy C,
right) were used during the 60-min experimental period
(unless death occurred earlier). On gross examination, the perivascular
groove was distended after strategy B, and the lung appeared dark and
congested after strategy C. There were no histopathologic changes in
nonventilated control lungs and lungs ventilated with strategy A. After
strategy B, perivascular edema was slightly increased. After strategy
C, there was marked edema and hemorrhage in the perivascular and
alveolar spaces, and all rats died after 13 to 35 min. Reproduced with
permission from Webb and Tierney.48
|
|
These and other studies provide convincing evidence that PEEP has lung
protective effects during mechanical ventilation. However, PEEP also
can contribute to lung injury by raising EILV unless VT is
simultaneously reduced. Moreover, PEEP may cause circulatory depression
from increased pulmonary vascular resistance and decreased venous
return. Thus, determining the optimal level of PEEP in individual
patients represents a difficult and tenuous balance between potential
lung protective effects and deleterious effects on the lung and other
systems.
Some investigators have used static or quasi-static pressure-volume
curves of the respiratory system to explain the effects of ventilation
at low EELV, to predict the effects of ventilation with higher PEEPs
and EELVs, and to identify the best PEEP to apply during CV to achieve
lung protection.12
40
51
52
The slope of the
pressure-volume relationship (Fig 6
) represents compliance of the respiratory system. Compliance in the
lower portion of the curve increases as airway pressure and volume
rise, representing gradual recruitment of atelectatic portions of the
lung. This interpretation is supported by improved arterial
oxygenation53
and CT evidence of increased lung aeration
in ARDS patients.54
The midpoint of the portion of the
pressure-volume curve with increasing slope is frequently labeled
"PFLEX" and may represent the inspiratory airway
pressure and volume where many lung units are open.55
The
mid-portion of the pressure-volume curve appears to be virtually
rectilinear. This region of approximately constant compliance has been
interpreted to represent a range of airway pressures and lung volumes
in which little or no further recruitment occurs.
Some workers have advocated setting PEEP to approximately
PFLEX plus 2 cm H2O to prevent the
closure of unstable lung units during expiration and, thus, to prevent
injurious shear forces from ventilation with insufficient
EELV.12
This recommendation is supported by the results of
studies suggesting that CV with PEEP that is less than
PFLEX may cause VALI. In a nonperfused rat model of lung
injury, for example, the effects of ventilation with PEEP that is below
and above PFLEX on lung injury were
compared.40
Lung compliance, bronchiolar epithelial
necrosis and sloughing, and hyaline membranes were significantly more
common in the lungs ventilated with PEEP less than PFLEX
than in lungs ventilated with PEEP more than PFLEX and in
control (nonventilated) lungs. There was no significant
difference in these measures of VALI between the latter two groups. In
surfactant-deficient rabbits, mechanical ventilation with PEEP equal to
PFLEX minus 5 cm H2O was compared to
ventilation with PEEP equal to PFLEX.51
Ventilation with PEEP less than PFLEX was associated with
greater hypoxemia and more hyaline membrane formation than ventilation
with PEEP equal to PFLEX.
|
CV-Based Lung Protective Strategies
|
|---|
CV strategies designed to protect the lung from VALI have been
tested in several clinical trials.
Studies With Reduced EILV
In two case series of patients with severe ARDS (a total of
approximately 100 patients), ventilation with small VTs
(reduced EILVs) was associated with mortality rates that were
substantially lower than rates predicted from the patients’ acute
physiology and chronic health evaluation (APACHE) II
scores.56
57
However, severe respiratory acidosis occurred
in some patients treated with small VTs, and previous
animal studies have suggested that respiratory acidosis could cause
circulatory depression.58
59
60
Moreover, respiratory
acidosis worsens dyspnea and agitation, which could increase
requirements for sedatives and, in some patients, neuromuscular
blockade. Because patients treated with traditional CV strategies were
not included in these studies, it was not clear that the beneficial
effects of reduced EILV outweighed the potential disadvantages.
Three modestly sized randomized clinical trials failed to show
beneficial effects of CV-based small VT ventilation in
comparison with more traditional VT strategies in patients
with or at risk for ARDS.61
62
63
In contrast, a large
multicenter trial with > 861 patients with ALI/ARDS conducted by the
National Institutes of Health (NIH)/National Heart, Lung, and Blood
Institute (NHLBI)-sponsored ARDS Network found substantial improvements
in clinical outcomes in the small VT group.14
The mortality rate prior to discharge home with unassisted breathing
was significantly reduced (31% vs 40%, respectively; p < 0.01)
among patients randomized to the small VT
strategy.14
Possible reasons for the different results in
this trial in comparison to the previous three studies include greater
separation in VT between treatment groups, greater
statistical power to detect differences in outcomes (larger sample
size), and more active management of respiratory acidosis.
Studies With Reduced EILV and Increased EELV
A clinical trial in 53 patients with severe ARDS compared a
traditional CV approach with an approach designed to protect the lung
from VALI resulting from both excessive EILV and inadequate
EELV.12
In the lung-protection group, pressure-limited
modes were used with VTs 
6 mL/kg and peak inspiratory
pressures < 40 cm H20 to reduce EILV. Increased
EELV was achieved, raising PEEP to PFLEX plus 2 cm
H20 during the initial stages of lung injury.
Frequent recruitment maneuvers were introduced to further increase
EELV, and additional measures were taken to avoid undesirable collapse
or derecruitment of some lung regions. The lung protection approach was
associated with an improved 28-day survival rate and weaning rate.
There was also an encouraging trend toward reduced in-hospital
mortality rate. However, this was a relatively small trial, and the
mortality rate in the traditional CV group (71%) was higher than those
of several previous studies.64
65
66
67
Also, it is not clear
from the results of this study which of the lung protection measures
improved outcomes. If higher PEEPs contributed to the improved outcomes
in the lung-protection group, it is not clear that PFLEX
plus 2 cm H2O was the best PEEP. Perhaps there
would be even further benefit with higher PEEPs in achieving even
greater EELV or with lower PEEPs to protect against adverse effects
from excessive EILV.
Summary: Lung Protective Modes of CV
Taken together, the body of experimental evidence from animals as
well as humans strongly suggests that a lung protective strategy with
smaller EILV and higher EELV will reduce VALI and improve outcomes in
patients with ALI/ARDS. However, CV-based lung protective strategies
have several limitations. First, safe limits for EILV have not been
clearly defined.68
Both peak and plateau inspiratory
pressures are influenced greatly by chest wall as well as lung
mechanics. The amount of stretch in the lungs of patients with normal
chest wall compliance may be greater than in the lungs of patients with
reduced chest wall compliance at any level of peak or plateau airway
pressure. No firm recommendations have been established to adjust
airway pressure thresholds for patients with different chest wall
characteristics. Measurement of esophageal pressures will allow
estimations of pleural pressures, which can be used to monitor and
limit peak transpulmonary pressure. When this technique is used
rigorously, it provides the opportunity to eliminate the influence of
chest wall compliance on airway pressures and to focus specifically on
distending forces in the lung. While this approach is logical and based
on sound physiologic principles, its use as a clinical tool has not
been widely accepted and safe limits for transpulmonary pressures have
not been defined. It is likely that stretch and lung injury gradually
rise as airway pressures and volumes rise and that mildly injurious
stretching occurs in some regions of the lung at relatively low lung
volumes and pressures. More information is needed to define the
relationships of lung volumes and airway pressures on stretch injury in
individual patients and to balance the potential benefits of marginal
reductions in stretch with potential problems associated with decreased
alveolar ventilation. Without this information, clinicians using CV in
ALI/ARDS patients may feel compelled to use smaller and smaller
VTs to achieve potential but uncertain increments in lung
protection at the risk of severe hypercapnia, respiratory acidosis, and
hemodynamic compromise.
Second, the optimal EELV for individual patients has not been defined.
The use of the PFLEX is problematic for several reasons.
Static or quasi-static pressure-volume curves do not depict the
relationship of lung volume to airway pressure during tidal
ventilation.69
70
In addition, PFLEX
frequently cannot be clearly identified on static pressure-volume
curves in patients with ALI/ARDS.71
72
Moreover, when a
PFLEX can be identified, it often reflects characteristics
of the chest wall rather than the lung.73
Also, alveolar
recruitment continues as airway pressure and lung volume rise above
their levels at PFLEX.74
75
Thus, it is not at
all clear that maximal lung protection is achieved at
PFLEX, at PFLEX plus 2 cm
H2O, or at any value above or below
PFLEX. Finally, increasing EELV (with higher PEEPs),
especially when it is used in combination with lower EILVs (smaller
VTs) during CV, will cause hypoventilation and may lead to
respiratory acidosis,56
57
dyspnea, circulatory
depression,58
59
60
increased cerebral blood flow, and risk
for intracranial hypertension,76
77
and it could increase
the requirements for heavy sedation and neuromuscular blockade.
Prolonged neuromuscular blockade, especially with concomitant
corticosteroid use, may cause neurologic complications, including
myopathy and neuropathy.78
Clinicians must consider these
risks when using higher PEEPs as part of a CV-based strategy to reduce
lung injury.
|
Rationale for HFV-Based Lung Protective Strategies
|
|---|
HFV is an attractive mode of ventilation in patients with ALI/ARDS
because of the following advantages over CV:
- HFV uses very small VTs. This allows
the use of higher EELVs to achieve greater levels of lung recruitment
while avoiding injury from excessive EILV.
- Respiratory rates with HFV are much higher than with
CV. This allows the maintenance of normal or near-normal
PaCO2 levels, even with
very small VTs.
Because of these advantages, some investigators have advocated the
use of HFV in patients with ALI/ARDS.15
79
The following
sections review the results of studies of HFV in animal models of lung
injury, pediatric patients with acute respiratory failure, and adults
with ALI/ARDS.
|
HFV for ALI and ARDS
|
|---|
Animal Studies
Some studies in premature primate models of surfactant deficiency
failed to show consistent beneficial effects during
HFV.80
81
82
83
84
In contrast, several studies found that HFV was
superior to CV when HFV was used as part of a strategy to achieve
higher EELV.47
85
86
87
88
In a rabbit model of surfactant
deficiency following saline solution lung lavage, 5 to 7 h of
ventilation with HFJV after a volume-recruitment maneuver (sustained
inflation to 30 cm H2O for 15 s) resulted in
lower peak airway pressures and MAPs and fewer hyaline membranes than
CV.85
In a separate study, adult rabbits after lung lavage
were randomized to a 7-h period of ventilation with HFO with high EELV
(HFO-Hi) or with HFO with low EELV (HFO-Lo), to CV, or to a control
group in which animals were killed after the lavage.47
All
rabbits randomized to CV died (three of five rabbits had
pneumothoraces), whereas none died in the HFO-Hi or HFO-Lo groups.
Total respiratory system compliance at the end of the experimental
period was highest in the HFO-Hi group, lower in the HFO-Lo group, and
lowest in the CV group. Microscopic examination revealed substantially
more hyaline membrane formation and bronchiolar epithelial injury in
the CV group, less in the HFO-Lo group, and little to none in the
HFO-Hi group. Animals randomized to the HFO-Hi group had significantly
higher PaO2 levels than those in
either the HFO-Lo or CV groups. Thus, HFO with higher lung volumes
was associated with superior lung mechanics, less lung injury,
and improved oxygenation than was HFO-Lo or CV. In another
surfactant-deficient rabbit model, the administration of exogenous
surfactant followed by ventilation with HFO and high EELVs was
associated with better preservation of surfactant function when
compared with CV.86
Other studies have demonstrated reduced lung inflammation with HFO. In
a surfactant-deficient rabbit model, the effects of HFO and CV at a
similar MAP and fraction of inspired oxygen
(FIO2) on lung injury were
compared.87
Rabbits in the HFO group had fewer
granulocytes and lower levels of platelet-activating factor and
thromboxane B2 in BAL fluid. In a rabbit
surfactant-deficient ALI model, accumulations of lung granulocytes and
activation of respiratory bursts in airspaces were greater with CV than
with HFO.88
In another rabbit lung-lavage study, animals were randomized to HFO
(f, 900 breaths/min), HFPPV (f, 120 breaths/min),
or HFO combined with CV (HFO with f = 900 breaths/min
superimposed during the expiratory phase of CV at f = 40
breaths/min).89
Each of the three study groups received a
high EELV strategy with sustained inflations every 20 min and MAP at
PFLEX plus 2 cm H2O. At the end of
the 6-h experimental periods, rabbits randomized to HFO had better
oxygenation and lung mechanics and less histologic evidence of lung
injury than those in the other two groups.
Pediatric Studies
Promising results in surfactant-deficient animal models of lung
injury led to studies evaluating HFV for neonatal respiratory distress
syndrome (RDS), the most common cause of lung injury in the
newborn,90
which is caused by insufficient surfactant
production in immature lungs.91
Neonates with RDS
experience acute respiratory failure and require mechanical ventilatory
support until the immature lung can produce enough surfactant to allow
adequate respiratory mechanics and gas exchange. In many respects, RDS
is physiologically and histologically similar to
ALI/ARDS.90
Although the use of mechanical ventilation is
associated with a reduction in mortality due to RDS, morbidity with
chronic lung disease (supplemental oxygen requirements and abnormal
chest radiographs at approximately 30 days of life) develops in 20 to
60% of preterm infants with RDS.92
HFJV and CV were compared in a study of 144 neonates (mean
gestational age, approximately 29 weeks) who developed pulmonary
interstitial emphysema (PIE; a form of VALI in RDS) after receiving
CV.93
The mean age at the start of the study was
approximately 2 days. A greater proportion of neonates randomized to
HFJV showed improvement in PIE (61% vs 37%; p < 0.01). The
incidence of chronic lung disease was lower (but not significantly
different) in the HFJV group. Overall, the mortality rate was similar
in both groups. To assess the effects of HFV started earlier in the
course of RDS, HFJV (with a lung recruitment strategy) was compared to
CV in 130 preterm neonates (mean gestational age, approximately 27
weeks) who had not yet developed significant bilateral
PIE.94
The mean age at the start of this study was
approximately 8 h. Fewer neonates randomized to HFJV developed chronic
lung disease at 36 weeks postconception than those randomized to CV
(20.0% vs 40.4%; p < 0.05). The survival rates at 36 weeks
postconception (HFJV, 84.6%; CV, 80%) and the incidences of chronic
lung disease 28 days after birth (HFJV, 67.3%; CV, 71.2%) were
similar. These findings are difficult to interpret, however, because 29
of 65 of the HFJV neonates were not managed with a high EELV strategy.
Uncontrolled observational studies suggested that HFO could reduce lung
injury in low-birth-weight infants95
and could improve gas
exchange in older children96
with acute respiratory
failure. However, these benefits were not confirmed in a large
multicenter, randomized, clinical trial of HFO vs CV in 673 preterm
infants with acute respiratory failure.9
Moreover, HFO was
associated with a significantly greater incidence of adverse events,
including pneumoperitoneum, grade 3 and 4 intracranial hemorrhage, and
periventricular leukomalacia. The interpretation of these results is
limited because the trial was conducted prior to the era in which
exogenous surfactant was routinely used for the treatment of neonatal
RDS. Moreover, the trial procedures did not use a strategy to increase
EELV with HFO.
Other clinical trials have shown more encouraging results with HFO
compared to CV for the treatment of RDS. In a presurfactant era study
employing a crossover trial design, 79 neonates with respiratory
failure requiring substantial levels of CV support were randomized to
HFO or continued CV.97
Neonates randomized to HFO met
predefined treatment failure criteria somewhat less often than neonates
randomized to CV (44% vs 60%, respectively; p value was not
significant). Among patients meeting failure criteria on the initial
ventilator assignment, however, more responded after crossover to HFO
than to CV (63% vs 23%, respectively; p = 0.03). There were no
differences in mortality rates, ventilator days, or other clinical
outcomes between the two groups. Neonates in the CV group appeared to
be less ill at baseline, and the study was terminated early due to low
enrollment. Thus, the results may have reflected confounding due to
baseline differences in the treatment groups, and the trial may have
been insufficiently powered to detect small differences in clinical
outcomes.
A second study evaluated the role of HFO in 83 premature neonates with
RDS.98
Using a crossover study design, 26 neonates were
assigned to CV only, 27 were assigned to HFO for 72 h followed by
CV (HFO/CV), and 30 were assigned to HFO only. The incidence of chronic
lung disease was significantly lower in the HFO-only group than in the
CV-only group. However, there were no significant differences in
the incidence of other clinical outcomes between treatment groups,
including pulmonary air leak, intraventricular hemorrhage, and
mortality. In a third trial, 175 neonates with RDS who were < 48 h
old were randomized to HFO or CV.99
The HFO procedures in
this trial used higher MAPs to achieve greater levels of lung
recruitment. HFO was associated with a similar mortality rate but a
reduced incidence of air leak syndrome (48% vs 63%, respectively;
p < 0.05). In a fourth study, 125 neonates with RDS received
exogenous surfactant before randomization to CV or HFO.100
HFO reduced vasopressor requirements, surfactant dosing requirements,
and the incidence of chronic lung disease. This experimental approach
was extended to older children (mean age, 2.5 years) who were receiving
CV for acute respiratory failure.101
Children randomized
to HFO required less supplemental oxygen support at 30 days compared
with children who continued on CV.
Adults Studies
In a series of five patients with acute respiratory failure of
diverse etiologies, PaO2 did not
improve with HFJV when similar levels of PEEP and
FIO2 were used in HFJV and
CV.102
Ultra high-frequency ventilation (UHFV), a modified
form of HFJV, was evaluated in 90 medical and surgical patients with
ARDS.103
UHFV uses a solenoid valve to achieve jet pulses
of gas at rates that are higher than those used with HFJV
(f, 60 to 1,200 breaths/min). Patients were eligible if
their FIO2 was > 0.7 with
a PaO2 of 65 mm Hg, a peak
inspiratory pressure of 65 cm H2O, or a PEEP of
15 cm H2O on CV. Oxygenation improved
significantly after 24 h of UHFV (arterial to alveolar ratio
increase [mean ± SD], 0.14 ± 0.07 to 0.26 ± 0.14;
p < 0.01). However, oxygenation was compared at different
FIO2 levels during CV and
UHFV, thereby limiting the interpretation of these findings.
HFJV was compared to CV in a randomized trial of 309 oncology patients
with body weight 
20 kg and respiratory failure requiring mechanical
ventilation.8
Patients were eligible if they developed
respiratory failure with bilateral infiltrates and hypoxemia following
surgery or sepsis. Both immunocompetent and immunocompromised
(eg, following bone marrow transplantation) patients were
enrolled. There were no significant beneficial effects of HFJV,
including mean (± SE) ICU length of stay (CV, 5.2 ± 0.5 days;
HFJV, 4.5 ± 0.3 days) and mortality rate (CV, 62%; HFJV, 62%).
However, as in the earlier pediatric studies with HFO, no lung
recruitment strategy was used in either group. Moreover, the inclusion
of patients with dissimilar levels of immune competence and with
respiratory failure from causes other than ALI/ARDS may have confounded
these results.
In another study, 113 surgical ICU patients at risk for ARDS were
randomized to high-frequency percussive ventilation (HFPV) or
CV.10
HFPV is a hybrid of CV and HFV in which
high-frequency airway pressure oscillations are superimposed on
traditional VTs and rates. Sixty patients (53%) developed
ARDS (32 receiving HFPV; 28 receiving CV). Four patients crossed over
to the alternate mode of ventilation after failing to meet predefined
criteria for improvement in oxygenation within 24 h of study
entry. After another 24 h, the ventilatory mode in which patients
achieved the best arterial oxygenation was used for the remainder of
the study. Thirteen patients were removed from the study due to
protocol violations or insufficient data and were not included in
subsequent analyses. There were no differences in clinical outcomes
(hospital days, length of ICU stay, or ventilator days) between the
treatment groups. These findings are not surprising since ventilation
with HFPV entails similar overall changes in intrathoracic volume (and,
thus, risk of VALI) during tidal breathing as that with CV.
In a 1997 case series, 17 medical and surgical patients (age range, 17
to 83 years) with severe ARDS who had a mean (± SD) APACHE II score
of 23 ± 7.5, and who required high
FIO2 levels, high airway pressures,
or high PEEPs after various periods of CV ([mean ± SD] 5.1 ± 4.3
days) were placed on HFO.79
The HFO procedures included
volume recruitment maneuvers and were associated with significant
improvements in gas exchange without adverse hemodynamic effects. The
survival rate was 47% in this high-risk group of patients. These
results demonstrate the feasibility of ventilating adults with ALI/ARDS
with HFO, but the lack of a comparison group limits further
interpretations. Prospective, randomized, controlled clinical trials
are needed to compare HFO to CV using strategies to increase EELV and
minimize EILV in adults with ALI/ARDS.
|
Conclusion
|
|---|
Numerous studies have suggested that CV may perpetuate or
exacerbate lung injury by delaying or preventing recovery from
ALI/ARDS. Small VT ventilation to reduce EILV during CV
recently was shown to improve mortality when compared to a more
traditional VT approach.14
There is also
abundant evidence in experimental animals and, more recently, in humans
to suggest that there are lung protective effects with higher EELV.
While there are encouraging results with recent CV-based lung
protective strategies, the potential benefits of these strategies may
be limited. HFV, especially HFO, offers the best opportunity to achieve
greater lung recruitment without overdistention while maintaining
normal or near-normal acid-based parameters. Results of animal and
pediatric studies using HFO are encouraging, but further work is
necessary to determine the value and optimal use of HFO in adults with
ALI/ARDS. Some studies suggest that HFV may be beneficial in
patients with bronchopleural fistulas. However, this gross
manifestation of volutrauma is uncommon (it occurs in
approximately 10% of patients with ALI/ARDS) and is not
significantly associated with increased
mortality.104
Thus, the extra costs of using HFO (new
equipment purchase and training of personnel for a small proportion of
patients who have ALI/ARDS) are not yet justified in the absence of
evidence that HFO will improve important clinical outcomes. Until the
results of well-designed studies demonstrate that HFO is superior to
CV, the use of HFO should be considered a promising but experimental
mode of ventilation for patients with ALI/ARDS.
|
Acknowledgements
|
|---|
The authors thank Henry E. Fessler, MD, for
assistance with artwork.
|
Footnotes
|
|---|
Abbreviations: ALI = acute lung injury;
APACHE = acute physiology and chronic health evaluation;
CV = conventional mechanical ventilation; EELV = end-expiratory
lung volume; EILV = end-inspiratory lung volume;
f = respiratory rate;
FIO2 = fraction of inspired oxygen;
HFJV = high-frequency jet ventilation; HFO = high-frequency
oscillation; HFO-Hi = high-frequency oscillation with high
end-expiratory lung volume; HFO-Lo = high-frequency oscillation with
low end-expiratory lung volume; HFPPV = high-frequency
positive-pressure ventilation; HFPV = high-frequency percussive
ventilation; HFV = high-frequency ventilation;
I/E = inspiratory/expiratory; MAP = mean airway pressure;
NHLBI = National Heart, Lung, and Blood Institute; NIH = National
Institutes of Health; PEEP = positive end-expiratory pressure;
PFLEX = midpoint of the portion of the pressure-volume
curve with increasing slope; PIE = pulmonary interstitial emphysema;
RDS = respiratory distress syndrome; UHFV = ultra high-frequency
ventilation; VALI = ventilator-associated lung injury;
VT = tidal volume
The authors have no financial interest in the subject discussed in this
article.
Received for publication January 24, 2000.
Accepted for publication January 25, 2000.
|
References
|
|---|
-
Marini, JJ (1996) Evolving concepts in the ventilatory management of acute respiratory distress syndrome. Clin Chest Med 17,555-575[CrossRef][ISI][Medline]
-
Fredberg, JJ (1980) Augmented diffusion in the airways can support pulmonary gas exchange. J Appl Physiol 49,232-238[Abstract/Free Full Text]
-
Taylor, G (1954) The dispersion of matter in turbulent flow through a pipe. Proc R Soc Lond 223,446-448
-
Drazen, JM, Kamm, RD, Slutsky, AS (1984) High frequency ventilation. Physiol Rev 64,505-543[Abstract/Free Full Text]
-
Slutsky, AS, Drazen, JM, Kamm, RD (1984) Alveolar ventilation at high frequencies using tidal volumes less than the anatomic dead space. Engel, LA Paiva, M Lenfant, CE eds. Lung biology in health and disease ,137-176 Marcel Dekker New York, NY.
-
Pinsky, MR, Summer, WR (1983) Cardiac augmentation by phasic high intrathoracic support (PHIPS) in man. Chest 84,370-375[Abstract/Free Full Text]
-
Angus, DC, Long, W, Dotterweich, LM, et al (1997) The influence of high-frequency jet ventilation with varying cardiac-cycle specific synchronization on cardiac output in ARDS. Chest 112,1600-1606[Abstract/Free Full Text]
-
Carlon, GC, Howland, WS, Ray, C, et al (1983) High-frequency jet ventilation: prospective, randomized evaluation. Chest 84,551-559[Abstract/Free Full Text]
-
. HIFI Study Group. (1989) High-frequency oscillatory ventilation compared with conventional mechanical ventilation in the treatment of respiratory failure in preterm infants. N Engl J Med 320,88-93[Abstract]
-
Hurst, JM, Branson, RD, Davis, K, et al (1990) Comparison of conventional mechanical ventilation and high frequency ventilation. Ann Surg 211,486-491[ISI][Medline]
-
Dreyfuss, D, Saumon, G (1998) State of the art: ventilator-induced lung injury; lessons from experimental studies. Am J Respir Crit Care Med 157,294-323[Free Full Text]
-
Amato, MBP, Barbas, CSV, Medeiros, DM, et al (1998) Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 338,347-354[Abstract/Free Full Text]
-
Ranieri, VM, Suter, P, Tortorella, C, et al (1999) Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome. JAMA 282,54-61[Abstract/Free Full Text]
-
The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 342 (in press)
-
Froese, AB, Bryan, AC (1987) High frequency ventilation. Am Rev Respir Dis 135,1363-1374[ISI][Medline]
-
Standiford, TJ, Morganroth, ML (1989) High frequency ventilation. Chest 96,1380-1389[Free Full Text]
-
Oberg, PA, Sjöstrand, U (1969) Studies of blood pressure regulation: I. Common carotid artery clamping in studies of the carotid-sinus baroreceptor control of the systemic blood pressure. Acta Physiol Scand 75,276-286[ISI][Medline]
-
Borg, U, Eriksson, I, Sjostrand, U (1980) High frequency positive pressure ventilation (HFPPV): a review based upon its use during bronchoscopy and for laryngoscopy and microlaryngeal surgery under general anesthesia. Anesth Analg 59,594-603[Free Full Text]
-
Sanders, RD (1967) Two ventilating attachments for bronchoscopes. Del Med J 39,170-175
-
Carlon, GC, Miodownik, S, Ray, C, et al (1981) Technical aspects and clinical implications of high frequency jet ventilation with a solenoid valve. Crit Care Med 9,47-50[ISI][Medline]
-
Ophoven, JP, Mammel, MC, Gordon, MJ, et al (1984) Tracheobronchial histopathology associated with high-frequency jet ventilation. Crit Care Med 12,829-832[ISI][Medline]
-
Lunkenheimer, PP, Rafflenbeul, W, Keller, H, et al (1972) Application of transtracheal pressures oscillations as modification of "diffusion respiration" [letter]. Br J Anaesth 44,627[Free Full Text]
-
Chang, HK (1984) Mechanisms of gas transport during ventilation by high frequency oscillation. J Appl Physiol 56,553-563[Abstract/Free Full Text]
-
Bernard, GR, Artigas, A, Brigham, KL, et al (1994) The American-European Consensus Conference on ARDS. Am J Respir Crit Care Med 149,818-824[Abstract]
-
Tsuno, L, Miura, K, Takeya, M, et al (1991) Histopathologic pulmonary changes from mechanical ventilation at high peak airway pressures. Am Rev Respir Dis 143,1115-1120[ISI][Medline]
-
Parker, JC, Hernandez, LA, Peevy, KJ (1993) Mechanisms of ventilator-induced lung injury. Crit Care Med 21,131-143[ISI][Medline]
-
Parker, JC, Hernandez, LA, Longenecker, GL, et al (1990) Lung edema caused by high peak inspiratory pressures in dogs. Am Rev Respir Dis 142,321-328[ISI][Medline]
-
Dreyfuss, D, Basset, G, Soler, P, et al (1985) Intermittent positive-pressure hyperventilation with high inflation pressures produces pulmonary microvascular injury in rats. Am Rev Respir Dis 132,880-884[ISI][Medline]
-
Pison, U, Tam, EK, Caughey, GH, et al (1989) Proteolytic inactivation of dog lung surfactant-associated proteins by neutrophil elastase. Biochim Biophys Acta 992,251-257[Medline]
-
Seegar, W, Lepper, H, Hellmut, RD, et al (1985) Alterations of alveolar surfactant function after exposure to oxidative stress and to oxygenated and native arachidonic acid in vitro. Biochim Biophys Acta 835,58-67[Medline]
-
Rinaldo, JE, Rogers, RM (1982) Adult respiratory-distress syndrome. N Engl J Med 306,900-909[ISI][Medline]
-
Maunder, RJ, Shuman, WP, McHugh, JW, et al (1986) Preservation of normal lung regions in the adult respiratory distress syndrome. JAMA 255,2463-2465[Abstract]
-
Gattinoni, L, Pesenti, A, Avalli, L, et al (1987) Pressure-volume curve of total respiratory system in acute respiratory failure: computed tomographic scan study. Am Rev Respir Dis 136,730-736[ISI][Medline]
-
Lessard, MR, Guerot, E, Lorino, H, et al (1994) Effects of pressure-controlled with different I: E ratios versus volume-controlled ventilation on respiratory mechanics, gas exchange, and hemodynamics in patients with adult respiratory distress syndrome. Anesthesiology 80,983-991[ISI][Medline]
-
Fuleihan, SF, Wilson, RS, Pontoppidan, H (1976) Effect of mechanical ventilation with end-inspirator pause on blood-gas exchange. Anesth Analg 55,122-130[Abstract/Free Full Text]
-
Cole, AGH, Weller, SF, Sykes, MK (1984) Inverse ratio ventilation compared with PEEP in adult respiratory failure. Intensive Care Med 10,227-232[ISI][Medline]
-
Tremblay, L, Valenza, F, Ribeiro, SP, et al (1997) Injurious ventilatory strategies increase cytokines and c-fos m-RNA expression in an isolated rat lung model. J Clin Invest 99,944-952[ISI][Medline]
-
Takata, M, Abe, J, Tanaka, H, et al (1997) Intraalveolar expression of tumor necrosis factor alpha gene during conventional and high-frequency ventilation. Am J Respir Crit Care Med 156,272-279[Abstract/Free Full Text]
-
von Bethmann, AN, Brasch, F, Nusing, R, et al (1998) Hyperventilation induces release of cytokines from perfused mouse lung. Am J Respir Crit Care Med 157,263-272[Abstract/Free Full Text]
-
Muscedere, JG, Mullen, JBM, Gan, K, et al (1994) Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care Med 149,1327-1334[Abstract]
-
Dreyfuss, D, Saumon, G (1993) Role of tidal volume, FRC, end-inspiratory volume in the development of pulmonary edema following mechanical ventilation. Am Rev Respir Dis 136,730-736
-
Robertson, B (1984) Lung surfactant. Robertson, B Van Golde, L Batenburg, J eds. Pulmonary surfactant ,236-257 Elsevier Amsterdam, the Netherlands.
-
Kolobow, T, Moretti, MP, Fumagalli, R, et al (1987) Severe impairment in lung function induced by high peak airway pressure during mechanical ventilation. Am Rev Respir Dis 135,312-315[ISI][Medline]
-
Corbridge, TC, Wood, LDH, Crawford, GP, et al (1990) Adverse effects of large tidal volume and low PEEP in canine acid aspiration. Am Rev Respir Dis 142,311-315[ISI][Medline]
-
Dreyfuss, D, Soler, P, Basset, G, et al (1988) High inflation pressure pulmonary edema. Am Rev Respir Dis 137,1159-1164[ISI][Medline]
-
Zapol, WM (1992) Volotrauma and the intravenous oxygenator in patients with adult respiratory distress syndrome. Anesthesiology 77,847-849[CrossRef][ISI][Medline]
-
McCulloch, PR, Forkert, PG, Froese, AB (1988) Lung volume maintenance prevents lung injury during high frequency oscillatory ventilation in surfactant-deficient rabbits. Am Rev Respir Dis 137,1185-1192[ISI][Medline]
-
Webb, HH, Tierney, DF (1974) Experimental pulmonary edema due to intermittent positive pressure ventilation with high inflation pressures: protection by positive end-expiratory pressure. Am Rev Respir Dis 110,556-565[ISI][Medline]
-
Wyszogrodski, I, Kyei-Aboagye, K, Taeusch, HW, et al (1975) Surfactant inactivation by hyperventilation: conservation by end-expiratory pressure. J Appl Physiol 38,461-466[Abstract/Free Full Text]
-
Tasker, V, John, J, Evander, E, et al (1997) Surfactant dysfunction makes lungs vulnerable to repetitive collapse and reexpansion. Am J Respir Crit Care Med 155,313-320[Abstract]
-
Argiras, EP, Blakeley, CR, Dunnill, MS, et al (1987) High PEEP decreases hyaline membrane formation i
TOP
Abstract
Introduction
