Invasive and Noninvasive Pediatric Mechanical Ventilation

Invasive and Noninvasive Pediatric Mechanical VentilationIra M Cheifetz MD FAARC
Introduction
Indications for Mechanical Ventilation
Noninvasive Mechanical Ventilation
Invasive Mechanical Ventilation
Conventional Mechanical Ventilation
Inspiratory Flow Pattern
Optimal Patient-Ventilator Interaction
Low Tidal Volume Ventilation
Tidal Volume Determination
High-Frequency Ventilation
Weaning from Mechanical Ventilation
Protocol Versus No Protocol
Extubation
Summary

Both invasive and noninvasive mechanical ventilation techniques are inherent to the care of
most patients admitted to intensive care units. Despite the everyday use of mechanical ventilation
for thousands of patients and the availability of thousands of reports in the medical
literature, there are no clear and consistent guidelines for the use of mechanical ventilation for
pediatric patients. In many areas data are lacking, and in other areas data are extrapolated
from studies performed with adult subjects. Despite the variability in views about mechanical
ventilation, 2 themes are consistent. First, modern pediatric respiratory care requires a substantial
institutional commitment for state-of-the-art management of the mechanically ventilated
patient. Second, a team approach involving physicians, nurses, and respiratory therapists
is essential. This review highlights some of the major issues affecting the pediatric patient who
requires invasive or noninvasive mechanical ventilation. These issues are pertinent to critical
care clinicians because one of the most common reasons for admission to an intensive care unit
is the need for mechanical ventilation. Furthermore, the duration of mechanical ventilation is
one of the major determinants of the duration and cost of an intensive care unit stay. Key
words: pediatric, respiratory, pulmonary, mechanical ventilation, acute lung injury, high-frequency
ventilation, noninvasive ventilation, weaning, extubation. [Respir Care 2003;48(4):442– 453. ©
2003 Daedalus Enterprises]

Ira M Cheifetz MD FAARC is affiliated with the Division of Pediatric
Critical Care Medicine, Duke Children’s Hospital, Durham, North Caro
Ira M Cheifetz MD FAARC presented a version of this report at the
31st RESPIRATORY CARE Journal Conference, Current Trends in Neonatal
and Pediatric Respiratory Care, August 16–18, 2002, in Keystone, Coloradolina.
Correspondence: Ira M Cheifetz MD FAARC, Duke University Medical
Center, Box 3046, Durham NC 27710. E-mail: cheif002@mc.duke.edu.
442 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4
Introduction
Artificial ventilation for respiratory failure is not a new
concept. Galen was the first scientist to describe ventilation
of an animal.1 As early as the 16th century the concept
of artificial ventilation for humans was presented. Vesalius
believed that people could be artificially ventilated with
air blown through a tube passed from the mouth into the
trachea.2 The use of mechanical devices to assist ventilation
became a clinical reality in the late 19th century. Most
of these early ventilators functioned through the use of
negative pressure and not positive pressure. At the beginning
of the 20th century, Emerson first used artificial positive-
pressure mechanical ventilation in the operating room
with anesthesia.3 Subsequently, the use of prolonged mechanical
ventilation to maintain life became widely accepted
during the polio epidemic of the 1950s.4–5 Today
mechanical ventilation plays an important role in most
intensive care units (ICUs) on a daily basis.
Artificial ventilation techniques are among the most important
clinical skills for any pediatric intensivist. Artificial
mechanical ventilation has substantially improved outcomes
of children suffering respiratory failure, by
maintaining adequate oxygenation and ventilation until the
underlying pathologic process resolves. It must be appreciated
that (1) mechanical ventilation is supportive (not
therapeutic) and (2) positive-pressure mechanical ventilation
inherently causes secondary lung injury of various
degrees, depending on the ventilatory strategies employed
and the clinical condition of the patient.
Mechanical ventilation can be delivered via positive-pressure
breaths or negative-pressure breaths. Additionally, the
positive-pressure breaths may be delivered noninvasively or
invasively. This review will focus on positive-pressure ventilation,
both noninvasive and invasive.
Although artificial ventilation techniques have dramatically
improved over recent years, many questions remain
unanswered, especially in relationship to the appropriate
strategy for weaning and extubating patients from mechanical
ventilation. Considering the wide range of disease
entities encountered daily in clinical practice, it is important
to note that the medical literature does not provide a
consensus concerning which ventilatory modes or strategies
are best applied to pediatric patients.
Indications for Mechanical Ventilation
Mechanical ventilation refers to the use of life-support
technology to perform the work of breathing for patients
who are unable to do so on their own. One of the most
common reasons for ICU admission is the need for mechanical
ventilation. Patients most commonly require mechanical
ventilation for respiratory failure or impending
respiratory failure. Respiratory failure occurs during conditions
of inadequate gas exchange of oxygen and/or carbon
dioxide. This failure of adequate oxygenation or ventilation
can occur as a result of lung disease, cardiac
dysfunction, neurologic abnormalities, multi-organ system
dysfunction/failure, and/or secondary to the effects of surgery
or cardiopulmonary bypass. Primary lung injury can
occur from a multitude of causes, including pneumonia,
inhalation injury, chest trauma, near-drowning, hemorrhage,
and aspiration. Patients with cardiovascular dysfunction
may require mechanical ventilation to minimize
the work of breathing, which, if excessive, could cause
lactic acidosis by increasing oxygen consumption at a time
when oxygen delivery may be limited.6 Patients with neurologic
injury may require mechanical ventilation for airway
protection and/or for hyperventilation to improve intracranial
hypertension. Thus, the overall goals of
mechanical ventilation are to optimize gas exchange, patient
work of breathing, and patient comfort while minimizing
ventilator-induced lung injury.

Noninvasive Mechanical Ventilation
Noninvasive ventilation (NIV) is defined as the use of a
mask or nasal prongs to provide ventilatory support through
a patient’s nose and/or mouth. By definition this technique
is distinguished from those ventilatory techniques that bypass
the patient’s upper airway with an artificial airway
(endotracheal tube [ETT], laryngeal mask airway, or tracheostomy
tube). NIV was first introduced in the late 1980s,
for patients with nocturnal hypoventilation.7– 8 Subsequently,
NIV has seen increasing popularity for pediatric
patients with both chronic and acute respiratory failure of
numerous etiologies.9–12
The primary advantage of NIV is the avoidance of endotracheal
intubation or tracheostomy. The secondary advantages
of not requiring an invasive airway include: decreased
risk of nosocomial pneumonia; ability to manage
many of these patients outside of the ICU (which may
decrease hospital costs); decreased sedation requirement
(including many patients who require no pharmacologic
sedation); improved ability to tolerate enteral feeds (including
a regular diet for some patients); and NIV allows
the patient to ambulate more easily. The ability to care for
patients who require NIV outside of the ICU setting differs
from one hospital to the next. When patients requiring
NIV are managed outside the ICU setting, close monitoring
is required, and protocols should be in place to help the
clinician determine when transfer to an ICU is warranted.
Noninvasive ventilation may be provided by either bilevel
pressure support or continuous positive airway pressure.
Bi-level support provides an inspiratory positive
airway pressure for ventilatory assistance and lung recruit-
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4 443
ment, and an expiratory positive airway pressure to help
recruit lung volume and, more importantly, to maintain
adequate lung expansion. Continuous positive airway pressure
provides only a single level of airway pressure, which
is maintained above atmospheric pressure throughout the
respiratory cycle.
A partial list of the clinical entities that might be successfully
treated with NIV includes impending acute respiratory
failure of almost any etiology, cystic fibrosis,
neuromuscular weakness, airway obstruction (including laryngotracheal
malacia), postextubation atelectasis, and
chronic respiratory failure. The vast majority of the literature
concerning NIV concentrates on the adult patient
population. However, a growing number of studies support
the use of NIV with pediatric patients suffering chronic
respiratory failure and impending acute respiratory failure.
9–12
Serra et al studied the effects of NIV in a series of adult
patients with cystic fibrosis and chronic respiratory failure.
Bi-level NIV improved ventilation by 30%, delivered tidal
volume (VT) by 30%, transcutaneously-measured carbon
dioxide level by 7%, and diaphragmatic activity by 20–
30%, depending on the NIV mode used.13 Fortenberry et al
reported an 11% incidence of intubation in a retrospective
review of pediatric patients who presented with impending
respiratory failure and were treated with NIV (mean 72 h,
range 20–840 h).10 The remaining 89% of the patients
demonstrated improved respiratory rates and gas exchange.
Padman et al prospectively studied a series of children and
adolescents (6 mo to 20 years of age) with impending
respiratory failure, among whom only 8% required intubation
for failure of noninvasive respiratory support.9
Increased use of NIV in the ICU setting may be warranted
for pediatric patients with impending respiratory
failure in an attempt to decrease the need for intubation
and invasive mechanical ventilation. The difficulty remains
in determining which individual patients might be predicted
to benefit from NIV. Additionally, the role of NIV
to facilitate extubation and shorten the duration of invasive
ventilation is promising but has largely been reported via
case reports and case series.11–12 Large-scale, prospective,
randomized pediatric studies are needed to help address
the optimal role of NIV for the pediatric patient suffering
impeding respiratory failure. If NIV can be proven to help
decrease the duration of invasive mechanical ventilation,
then the adverse effects and the cost associated with invasive
ventilation may be decreased.

Invasive Mechanical Ventilation
Although it is reasonable to attempt NIV in certain patient
populations, the vast majority of patients who require
ventilatory support need invasive, positive-pressure mechanical
ventilation, either conventional or high-frequency.
In 1997 an estimated 100,000 positive-pressure ventilators
were utilized around the world, and approximately half
were in use in North America.14 Approximately 1.5 million
patients in the United States receive mechanical ventilation
outside of operating rooms and recovery rooms
every year.14
Mortality among patients who require mechanical ventilation
is widely variable and dependent on the underlying
clinical condition that necessitated the ventilatory support.
For pediatric patients with rapidly reversing conditions
and who are otherwise healthy, mortality rates approach
0%. Patients with severe acute respiratory distress syndrome
(ARDS) suffer 30–60% mortality. Ventilated patients
with severe multi-organ system failure and/or severe
immunodeficiency suffer 90–100% mortality.
Conventional Mechanical Ventilation
Multiple mechanical ventilation modes are currently used
in clinical practice to provide respiratory support for a
wide spectrum of patients, ranging from no lung disease to
acute lung injury (ALI) to ARDS. To date no data exist to
determine the ventilatory mode that provides the greatest
benefit with the least risk to an individual pediatric patient.
Each new generation of conventional mechanical ventilators
brings new ventilation modes and new features.
However, despite a multitude of new modes, no study has
shown that any mode is better than another in improving
survival rates for ALI patients. It should be noted that in
reality it might not be possible to demonstrate a significant
change in mortality based only on changes in ventilator
mode, because of the extremely low baseline mortality rate
for intubated infants and children in pediatric ICUs.
However, 4 important ventilation concepts have surfaced
that might significantly affect mortality, morbidity,
and patient comfort. First, the inspiratory gas flow pattern
has important clinical implications. Second, optimal patient-
ventilator interaction is essential for patient comfort
and for minimizing the duration of ventilation. Third, the
data that have demonstrated that low-VT ventilation improves
mortality in adult patients are probably also applicable
to pediatric patients. Lastly, if low-VT ventilation is
to be accurately applied to infants and small children, an
accurate VT measurement must be obtained.

Inspiratory Flow Pattern
Various inspiratory gas flow patterns are available on
conventional ventilators. Regardless of the inspiratory flow
pattern chosen, gas flow will always follow the path of
least resistance. Variations in the inspiratory flow pattern
will affect the distribution of inspired gas flow based on
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
444 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4
the patient’s underlying clinical pathophysiology. Accelerating-
flow patterns deliver the highest gas flow at end
inspiration, when the effects of resistance and elastance
are increased. Thus, accelerating-flow patterns typically
produce higher peak inspiratory pressure (PIP) than other
flow patterns and are rarely used in current clinical practice.
In contrast, decelerating-flow patterns deliver maximum
flow at the initiation of inspiration, when resistance
and elastance are decreased. Inspiratory flow then decreases
during inspiration as delivered gas volume increases. Peak
airway pressures are lower but mean airway pressures are
higher with a decelerating-flow pattern than with a constant-
flow pattern.15 In general, as the maximum inspiratory
flow changes from the start to the end of the inspiratory
cycle, mean airway pressure will decrease and PIP
will increase, for the same positive end-expiratory pressure
(PEEP), inspiratory time, and delivered VT. Thus,
decelerating-flow patterns have a theoretical advantage over
accelerating-flow patterns, especially with ALI patients. A
square wave, constant inspiratory flow pattern will typically
have peak and mean airway pressures somewhere
between the values seen with accelerating and decelerating
patterns (Fig. 1).
Variable-flow ventilation (ie, pressure-controlled, or
pressure-regulated volume-controlled) uses a deceleratingflow
pattern.16–18 The rapid increase in inspiratory flow
that occurs with variable, decelerating-flow ventilation
leads to early filling of alveoli and sustains alveolar pressure
longer than in a constant-flow pattern. Thus, variable,
decelerating-flow ventilation potentially provides better alveolar
recruitment19 and should improve gas distribution
throughout the lungs.15 By improving gas distribution the
desired VT can be delivered at a lower PIP than with a
constant inspiratory flow, corresponding to improved lung
compliance.15
The rapid increase in airway pressure in deceleratingflow
ventilation can also lead to an increase in the overall
mean airway pressure, and, thus, better arterial oxygenation
and oxygen delivery.16–18,20–22 In adults the increase
in mean airway pressure associated with decelerating-flow
ventilation is not associated with hemodynamic abnormalities.
23 Thus, respiratory pathology characterized by low
pulmonary compliance (ie, ALI and ARDS) may benefit
from a decelerating-flow inspiratory pattern, in which PIP
is reduced but the mean airway pressure is increased.
The clinician should attempt to match the inspiratory
flow pattern to the patient’s clinical condition. In contrast
to the case of ALI, in diseases that cause high airway
resistance (asthma, bronchiolitis, airway obstruction) peak
airway pressure may, theoretically, be reduced by avoiding
flow patterns that have high peak inspiratory flows. In
high-airway-resistance patients a square-wave constantflow
pattern may generate a lower PIP than a deceleratingflow
pattern, as a result of the lower peak inspiratory flow.
However, conclusive data are lacking in support of this
speculation.
In summary, the single most important aspect of the
ventilation mode chosen for an individual patient may be
the inspiratory flow pattern associated with the mode. Beyond
the issue of inspiratory flow patterns, the optimal
mode of ventilation for infants and children remains unclear.
Optimal Patient-Ventilator Interaction
Optimizing patient-ventilator interaction is essential to
providing the best possible care for any intubated patient.
Optimal patient-ventilator interaction will improve patient
comfort while potentially decreasing the requirement for
pharmacologic sedation and thereby may help to minimize
the duration of mechanical ventilation. Graphic analysis of
ventilation and respiratory mechanics monitoring has become
an integral part of conventional ventilator management
and is an important tool in assessing and changing
ventilation strategy. This technology incorporates monitoring
the patient, the ventilator, and patient-ventilator interaction.
Effective respiratory monitoring of a conventionally
ventilated patient should assist the clinician in
assessing adverse patient-ventilator interactions and provide
important information to help clinicians intervene prospectively.
24 If ventilated infants and children are to be
comfortable, ventilated for the shortest possible time, and
optimally use the nutritional support provided, the patient
and ventilator system must interact synchronously.25 Recent
advances in ventilator technology allow the clinician
to customize the patient-ventilator interface, resulting in a
more optimal interaction. Rosen et al demonstrated a reduction
in ventilator-induced lung injury when respiratory
mechanics measurements (at the ETT) were used in the
care of neonates.26
Thus, with the numerous ventilator modes, inspiratory
flow patterns, and patient-triggering options available for
Fig. 1. Inspiratory flow patterns. The top panels show 2 of the most
common inspiratory flow patterns: variable, decelerating-flow and
constant, square-wave flow. The lower panels show the relationship
of inspiratory flow to the change in airway pressure.
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4 445
neonatal and pediatric ventilation, graphic analysis of ventilation
has become an important tool for determining the
most beneficial ventilatory strategy for each patient and
for identifying the presence of adverse patient-ventilator
interactions.
An integrated graphic display that reflects patient response
and ventilator performance. This information may
improve detection and identification of critical events, enabling
the practitioner to rapidly determine the presence of
respiratory pathophysiology by evaluating VT, airway pressures,
gas flow, and pressure/volume and flow/volume relationships.
The primary adverse patient-ventilator interactions
that can impact the medical management of patients
include pulmonary overdistention, intrinsic PEEP, and patient-
ventilator asynchrony (Figs. 2-4.)25
Pulmonary overdistention can cause volutrauma and ventilator-
induced lung injury. Clinically important intrinsic
PEEP may cause gas trapping, impaired gas exchange,
pulmonary overdistention, and elevated mean intrathoracic
pressure. Patient-ventilator asynchrony can cause the patient
to become uncomfortable with the ventilator. If patient-
ventilator asynchrony is not appreciated by the clinician,
unnecessary pharmacologic sedation may be
administered, prolonging the mechanical ventilation. Patient-
ventilator asynchrony most commonly results from
Fig. 3. Intrinsic positive end-expiratory pressure. The top curve
shows airway pressure (Paw) versus time. The lower curve shows
airway flow (V˙ ) versus time. Intrinsic positive airway pressure occurs
when inspiratory flow begins before expiratory flow from the
prior breath returns to zero. The arrows indicate the initiation of a
positive-pressure breath from a point beneath the horizontal axis.
Fig. 2. Pulmonary overdistention. This pressure-volume curve demonstrates
overdistention. The upper inflection point and the start
of overdistention are indicated by the arrow. Paw airway pressure.
VT tidal volume.
Fig. 4. A: Patient-ventilator asynchrony caused by trigger insensitivity.
The top curve shows airway pressure (Paw) versus time. The
lower curve shows airway flow (V˙ ) versus time. The arrows labeled
“a” indicate spontaneous breaths, during which the patient is moving
gas flow but is unable to trigger the ventilator to initiate a
ventilator-assisted breath. The arrow labeled “b” indicates a mechanical
breath that has been triggered by time. After this point the
patient is asynchronous with the ventilator, as shown by the very
irregular flow pattern. Improving the trigger sensitivity enables the
patient to interact with the ventilator and improve the patientventilator
interaction. B: Patient-ventilator asynchrony caused by
inadequate inspiratory flow. The top curve shows airway pressure
versus time. The lower curve shows airway flow versus time. This
patient was being ventilated with a synchronized intermittent mandatory
ventilation (SIMV)/volume-limited/pressure support approach.
Each SIMV/volume-limited mechanical breath includes a
depression (arrows) in the middle of inspiration. At that point the
patient is “double breathing” in an attempt to obtain greater flow
at a certain point during inspiration. This situation can often be
corrected by changing the inspiratory flow to a variable, decelerating-
flow pattern.
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
446 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4
an inappropriately set inspiratory trigger or inadequate inspiratory
flow.
Inadequate trigger sensitivity is the most common cause
of patient-ventilator asynchrony with infants. The frequency
of this type of asynchrony has decreased as more
ventilators have provided flow-triggering, regardless of
the ventilation mode. A spontaneously breathing patient
who is unable to trigger a mechanical breath will appear
agitated and will “fight the ventilator.” If this patient is
treated with increased pharmacologic sedation, the patient
will appear more comfortable as the spontaneous respiratory
drive is suppressed, but the patient will probably require
a more prolonged mechanical ventilation. The ideal
therapeutic option is to improve the trigger sensitivity to
allow the patient to freely interact with the ventilator.
Flow asynchrony results when a patient does not receive
the inspiratory flow he or she desires at any point during
inspiration. Flow asynchrony is most commonly seen in
modes that have a square wave, constant inspiratory flow
pattern. Although the synchrony may be improved by increasing
the set inspiratory flow in a constant-flow mode,
this most commonly results in increased PIP. A better
option to treat flow asynchrony is to change to a mode that
uses a variable, decelerating inspiratory flow pattern. With
a variable flow pattern the inspiratory flow is better matched
with the patient’s demand throughout the breath.
In summary, it is important to optimize the patientventilator
interaction by optimizing the ventilator settings
before resorting to sedation. Sedative use in the first 24
hours of weaning from mechanical ventilation influences
the duration of mechanical ventilation and extubation failure
in infants and children.27

Low Tidal Volume Ventilation
The ARDS Network reported in 2000 that with ALI/
ARDS patients, mechanical ventilation with a VT of approximately
6 mL/kg resulted in lower mortality and fewer
ventilator days than a more traditional VT of 12 mL/kg.28
The mortality rate was 31.0% in the low-VT group and
39.8% in the high-VT group (p 0.007). Additionally, the
plateau pressure was significantly lower in the low-VT
group on days 1, 3, and 7. This study was limited to adult
patients (average age approximately 51 years). However,
the results are very likely to be applicable to pediatric ALI
patients. Until a similar large-scale, prospective, randomized
trial is performed with infants and children, it seems
reasonable to follow the low-VT guidelines. It should be
emphasized that the low-VT data were obtained from ALI
patients, and it remains uncertain whether larger VT can be
safely used in patients with normal lung function (ie, those
intubated for nonpulmonary reasons).
Tidal Volume Determination
To successfully accomplish low-VT ventilation it is essential
to know the exact VT that is delivered to the lungs.
Conventional ventilator displays of exhaled VT are clinically
used to indicate the delivered VT. Some ventilators
use a pneumotachometer to measure expired VT at the
ETT, whereas others measure VT at the ventilator’s expiratory
valve. VT measurements at the ventilator’s expiratory
valve might not be able to compensate for the compliance
of the ventilator circuit nor for uncontrolled clinical
variables, including secretions, changes in humidification,
changes in temperature, condensation, in-line suction devices,
and end-tidal carbon dioxide monitor adapters. Theoretically,
a VT measured with a pneumotachometer positioned
at the ETT is a more accurate and reliable
measurement of the VT actually delivered to the patient’s
lungs than is a VT measured at the ventilator expiratory
valve. This issue may not be clinically important for large
pediatric patients and adult patients, but may be very important
for infants and small children.
An alternative to placing a pneumotachometer at the
ETT is to use a mathematical model to estimate the volume
of gas delivered to the ETT (calculated effective VT).
Theoretically, the effect of the circuit compliance on the
accuracy of the VT measurement made at the ventilator
expiratory valve can be mathematically eliminated without
requiring a pneumotachometer. Effective VT is calculated
by subtracting the VT “lost” to the ventilator circuit from
the VT displayed by the ventilator.29 The effective VT has
traditionally been defined as the ventilator-measured VT
minus the volume “lost” because of the distensibility of
the ventilator circuit. That is:
EffectiveVT ventilator expired V T
[circuit compliance (PIP PEEP)]
The compliance of a ventilator circuit can be obtained
from the manufacturer or calculated from pressure and VT
measurements at both ends of the circuit. However, more
elaborate equations are required to estimate the effects of
the other variables in the ventilator circuit (eg, temperature,
condensation, secretions, in-line suction devices).
The difference between the ventilator-determined VT,
the pneumotachometer-determined VT, and the calculated
effective VT may be clinically important. The ventilator
circuit compliance is particularly relevant in determining
the actual volume that enters the lungs of neonates, infants,
and small children, given the overall smallVT. Knowing
the exact delivered VT is essential when ventilating
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4 447
infants, because the volume “lost” to the distensibility of
the circuit can be equal to the desired VT.
Cannon et al reported that with infants ventilated using
a neonatal ventilator circuit the expiratory VT measured at
the ETT is on average only 56% of that measured at the
ventilator.30 Somewhat better correlation was seen in pediatric
patients ventilated with a pediatric circuit: the average
VT measured at the ETT was 73% of that measured
at the ventilator expiratory valve.
Additionally, Cannon et al demonstrated that the basic
correction equation listed above is not sufficient.30 The
study found a poor correlation between the calculated effective
VT and the exhaled VT measured by the pneumotachometer
at the ETT. All of the ventilator circuit variables
listed above can compromise the accuracy of the
calculation by adding uncontrolled and variable dead space
to the circuit. However, it must be noted that some newgeneration
ventilators include more advanced calculations
that might calculate VT delivery more accurately and obviate
the pneumotachometer at the ETT. The accuracy of
these advanced software calculations has not yet been fully
tested in the clinical setting.
Especially with infants and small children, inaccuracies
in VT measurement may have important adverse clinical
consequences. The young patient may be at high risk for
ventilator-induced lung injury, hypoxia, and hypercapnia
if the actual volume entering the lungs is not accurately
measured.31–36 If the VT is inappropriately small, atelectasis
and ventilation-perfusion mismatching may occur.34
If atelectasis develops, increased mean and/or peak airway
pressures may be required to recruit the collapsed lung
regions, potentially leading to increased shear injury and
barotrauma.34–35 Although atelectasis can be overcome by
“simply” increasing the VT and/or the PEEP, the VT that
must be set on the ventilator to deliver the appropriate
volume remains unknown.
Additionally, even before atelectasis develops, the clinician
may attempt to compensate for the discrepancy in
the VT measured in the ventilator by increasing the set
limit for each breath (VT or PIP), as determined by chest
auscultation. However, overcompensation may occur, causing
excessive delivered VT and ventilator-induced, iatrogenic
lung injury.31,32,34,36,37 Ventilation with excessive VT
results in disruption of the pulmonary architecture.33,38
A pneumotachometer placed at the ETT (either connected
to the ventilator or a stand-alone respiratory mechanics
monitor) offers a reliable measurement of the delivered
VT and may help to minimize iatrogenic lung injury
in infants and small children.39,40 Additionally, optimizing
the actual delivered VT may help to limit intrathoracic
pressure and potentially minimize secondary cardiovascular
and neurologic adverse sequelae.26,39–41
High-Frequency Ventilation
High-frequency ventilation is defined as ventilation that
delivers a VT that is less than the dead space volume.
Additionally, the respiratory rate in pediatric HFV is defined
as 150 breaths/min. The concept of HFV is not
new. In 1915 Henderson and Chillingworth described the
theoretical effects of a rapid ventilatory rate on gas exchange.
42 In 1952 Emerson patented the first high-frequency
device for clinical use,42 and in 1972 the first
high-frequency oscillator was described by Lunkenheimer.
43 The theoretical advantage of HFV is that it maintains
an open lung with the use of relatively high mean
airway pressure but low phasic volume and pressure
changes. This concept was well demonstrated over a decade
ago by Kinsella et al, who reported that optimizing
functional residual capacity in a manner that promotes
lung inflation and minimizes cyclical stretch of the lungs
attenuates ventilator-induced lung injury.44
Although most ALI patients are adequately oxygenated
and ventilated with conventional mechanical ventilation,
there is a subset of ALI patients who require “excessive”
PIPs with conventional ventilation to maintain lung recruitment.
With these patients HFV may prevent or minimize
ventilator-induced lung injury.45,46 Arnold et al demonstrated
in a multicenter, prospective, randomized study
that despite the higher mean airway pressure, high-frequency
oscillatory ventilation (HFOV) was associated with
less chronic lung disease, as indicated by less need for
supplemental oxygen at 30 days and better outcome than
with conventional ventilation.45 This study additionally
demonstrated that among patients who were ventilated with
HFOV and survived, the risk of chronic lung disease was
associated with the duration of conventional ventilation
before initiation of HFOV. However, although this important
study demonstrates the potential benefit of HFOV for
pediatric ALI and ARDS, it should be noted that the study
analyzed a limited number of patients (n 58).
The pressure-volume curve in Figure 5 illustrates the
potential lung-protective advantage of HFV.47 Below the
lower inflection point, low lung volumes, derecruitment,
and atelectasis result in ventilator-induced lung injury with
every breath, as the lung is opened by the delivered VT and
then allowed to collapse (atelectrauma). Above the upper
inflection point, ventilator-induced lung injury occurs as
alveoli become overdistended (volutrauma). HFV allows
gas exchange to occur between the upper and lower inflection
points and, theoretically, minimizes ventilator-induced
lung injury.
Although various high-frequency devices are used with
neonates, the most frequently used device for pediatric
ALI and ARDS is the SensorMedics 3100A oscillator,
which was the first such device approved (1995) by the
United States Food and Drug Administration (FDA) for
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
448 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4
early intervention in pediatric respiratory failure. For pediatric
ALI and ARDS patients who weigh 35 kg, the
SensorMedics 3100B oscillator (which received FDA approval
in 2001) can generate a greater power output and
can function at a higher bias flow to allow for more efficient
ventilation in these larger patients.
One of the more difficult clinical decisions concerning
HFOV is when to initiate it. Although there are no clear
guidelines, a recent publication reviewed the use of HFOV
in 290 pediatric patients over 18 months at 10 tertiary care
pediatric ICUs.48 On average, HFOV was initiated in patients
who did not have prior lung disease when the PIP on
conventional ventilation was 34.2 7.9 cm H2O, and the
oxygenation index (OI) was 27.5 14.1. The OI was
calculated as:
OI (Paw FIO2
100)/PaO2
in which P aw is mean airway pressure and FIO2
is fraction
of inspired oxygen. For patients who had prior lung disease
PIP was 34.2 7.5 cm H2O, and OI was 28.7 16.1.
These relatively high oxygenation indices for initiation of
HFOV are in contrast to the FDA approval of the oscillator as
an early intervention device. Based on the previous study by
Arnold et al, earlier use of HFOV may improve outcome for
pediatric ALI patients by minimizing ventilator-induced lung
injury.45 However, it must be noted that no study has been
done with pediatric patients to compare HFOV to conventional
ventilation with an “open lung strategy” and low VT.
The most recent HFOV study by Arnold et al represents
the largest series of pediatric patients receiving HFOV,
and, thus, the results help to define the current utilization
patterns of HFOV and to predict outcome for subgroups of
patients.48 In this study immunocompromise was associated
with a significantly higher mortality risk. Patients with
sepsis and ALI had a higher risk of chronic lung disease than
nonseptic ALI patients. Overall, patients who demonstrated a
minimal therapeutic response within the first 24 hours of
HFOV had an extremely high mortality risk.
With the growing use of HFV the term “nonconventional
ventilation” is becoming a misnomer. There is no
longer anything nonconventional about HFV. HFOV has
been an FDA-approved mode of ventilation for more than
a decade and thus should now be considered another conventional
ventilation mode.
Weaning from Mechanical Ventilation
A major difficulty involving definitions continues to
exist with regard to weaning from mechanical ventilation.
Some clinicians define weaning as the decrease of ventilatory
support in preparation for imminent extubation; other
clinicians state that weaning should be initiated as soon as
a patient is intubated. The current, generally accepted philosophy
is that it is necessary to gradually wean the patient
from mechanical ventilation implemented because of respiratory
failure, to retrain their respiratory muscles.
Whether this philosophy is actually supported by scientific
data remains controversial. In 1987 Hall and Wood disagreed
with the traditional view and suggested the term
“liberation from mechanical ventilation.”49 It is becoming
more evident that many patients who have been traditionally
weaned over the course of days can be rapidly extubated
without complication.50 Thus, the traditional view of
a gradual weaning process is being questioned.
Regardless of whether a patient is “weaned” or “liberated”
from mechanical ventilation, the goal should be to
minimize the duration of ventilation for every patient. Prolonged
mechanical ventilation is associated with prolonged
ICU stay, prolonged hospital stay, higher costs, higher risk
of nosocomial pneumonia, progressive ventilator-induced
lung injury, airway injury, excessive pharmacologic sedation,
and possibly higher mortality.51–54 Thus, minimizing
the duration of ventilation is clinically important. On the
other hand, discontinuing ventilation prematurely can necessitate
reintubation, which is associated with similar complications.
The optimal weaning process can be a clinically difficult
balance between minimizing the duration of mechanical
ventilation and decreasing the risk of reintubation.
This clinical balance plays a very important role in the
management of critically ill infants and children in ICUs
every day.
Protocol Versus No Protocol
Despite the use of mechanical ventilators in ICUs every
day, the ideal method to wean infants and children from
Fig. 5. Pressure-volume relationships of acute lung injury. The goal
of mechanical ventilation is to avoid the 2 regions of lung injury:
the zone of overdistention and the zone of derecruitment and
atelectasis. Ideally, the full breath should be accomplished in the
“safe” window. (From Reference 47, with permission.)
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4 449
respiratory support has only recently been studied.27,55 Traditionally,
weaning methods for children have been extrapolated
from studies with adults and premature neonates.
The unique aspects of pulmonary physiology,
respiratory mechanics, and the epidemiology of ALI in
infants and children make it unlikely that strategies extrapolated
from other populations will be effective.56 The
duration of weaning is usually shorter with infants and
children because they have healthier baseline lung function
than adults, so recovery from a pulmonary insult is
usually more rapid.
Studies with adult patients have demonstrated that when
protocols are used to guide ventilator weaning, the duration
of ventilation is significantly less than when care is
guided by individual clinician practice.57,58 However, currently
there are no generally accepted weaning protocols
for children, and the lack of evidence on optimal use of
weaning techniques results in great variability in the way
they are clinically utilized.
A recent randomized, prospective study by the Pediatric
Acute Lung Injury and Sepsis Investigators (PALISI) Network
was designed to study protocol weaning versus nonprotocol
weaning in a population of children with ALI.27
The use of weaning protocols in the population of infants
and children studied had no impact on the duration of
mechanical ventilation. This is in direct contrast to the
available data from the adult population.57,58 An important
difference between adult and pediatric patients is the shorter
duration of weaning with infants and children. In the PALISI
study the mean duration of weaning was only 2.9 days
(median 1.7 d) in the protocol groups and 3.2 days (median
2.0 d) in the control group.
Extubation
Similar to the situation with weaning, the ideal extubation
timing for the ALI patient has been elusive, and the
techniques used have traditionally been more art than science.
As with weaning, extubation involves substantial
risks; failed extubation increases the risk of pneumonia,
prolongs ICU stay, increases the risk of death, and increases
costs.59–65 Over the last several years increased
interest in this issue has led to important scientific results.
Predicting successful extubation of infants and children
presents unique challenges to pediatric intensive care clinicians.
Currently there are no widely accepted methods
for predicting successful extubation in pediatric patients.
Methods used to predict extubation in adults, such as the
ratio of respiratory frequency to VT, the CROP (compliance,
rate, oxygenation, and pressure) index, T-piece trial,
and negative inspiratory effort measurements are either
unreliable or not easily performed with children.66–68
As discussed above, it is often difficult to obtain the
ideal balance between minimizing the duration of ventilation
and minimizing the risk of reintubation. Although the
appropriate balance is often discussed with various answers,
the largest series of pediatric patients studied to
determine an expected failure rate for planned extubation
was by Edmunds et al.69 The study was a retrospective
chart review of 632 patients. The overall failure rate of
planned extubations in that pediatric population was 4.9%.
As expected, younger patients who underwent longer duration
of ventilation were at higher risk for extubation
failure.69
A pediatric clinical study by Khan et al characterized
multiple predictors of extubation failure.70 Unfortunately,
these authors were unable to identify a single variable or
formula for predicting the success of extubation with children
and concluded that a combination of factors should
influence any extubation decision.
Hubble et al evaluated the usefulness of pulmonary
dead space measurements in predicting pediatric extubation
outcomes.71 Dead space represents the portion of
the pulmonary system that is not involved in gas exchange,
including both airway dead space and alveolar
dead space. Dead space is often expressed as the ratio of
dead space to VT (VD/VT), also known as the physiologic
dead space ratio.
During the past 2 decades intensivists have identified
several clinical applications for VD/VT. In adult patients
VD/VT has been used to reliably and quickly identify pulmonary
embolism, monitor the effects of fluid infusion in
intubated asthmatic patients, and measure the effects of
bronchodilators in patients with chronic obstructive pulmonary
disease.72–76 VD/VT has been identified as a predictor
of mortality among neonates suffering congenital
diaphragmatic hernia,77 and it has been used to detect pulmonary
shunt in congenital heart patients78 and to determine
pulmonary improvement in patients supported with
extracorporeal membrane oxygenation.79 Since VD/VT has
proven reliable in assessing the progression of lung disease,
it would also be expected to correlate with the regression
of lung disease.
Traditionally, VD/VT was measured by collecting expired
gas. Recent advances in computer and capnography
technology simplified the calculation of VD/VT from single-
breath carbon dioxide waveforms. Hubble et al71 successfully
identified VD/VT values predictive of extubation
success and failure for infants and children, using singlebreath
carbon dioxide measurements. VD/VT values 0.50
at the time of extubation were associated with extubation
success, and VD/VT values 0.65 were associated with
the need for additional respiratory support following extubation.
A recent multicenter pediatric ALI trial used objective
criteria to determine extubation readiness by protocol ver-
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
450 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4
sus by physician judgment in the no-protocol arm of the
study. Objective criteria did no better than physician judgment
in determining which patients could be successfully
extubated.27 The average extubation failure rate was 19%
using an extubation readiness test and 17% using physician
judgment. These failure rates are consistent with some
previously reported rates in the literature.70,80,81 However,
other reports quote reintubation rates as low as 5%.69,71
Differences between inclusion and exclusion criteria in
these clinical studies, specifically the issues of upper airway
obstruction and minimal duration of ventilation, make
comparison of these reports difficult. Additional difficulties
are the somewhat subjective nature of the decision as
to whether a patient has failed extubation, the variable use
of NIV to help avoid reintubation, and the variable time
frame that patients are followed after extubation. Published
extubation failure rates in adult studies range from
1.8 to 18.6%.61,82–84
It should be noted that all of the extubation readiness
tests presented above for the pediatric and adult populations
test only the patient’s pulmonary status. The patient’s
overall clinical status must be considered before a patient
is extubated. Neurologic considerations include the patient’s
sedation status, ability to protect the airway, and
acceptable intracranial pressure. Cardiovascular considerations
include the degree of inotropic support, the presence
of hemodynamic stability, and the anticipated effects of
increased respiratory effort on cardiac function. Additional
considerations include the presence of an air leak around
the ETT and the resolution of the underlying process that
necessitated intubation.
Summary
The field of pediatric mechanical ventilation has advanced
dramatically over the last decade. During this period
many changes have occurred and continue to occur.
Noninvasive ventilation is being used at an increasing rate
to obviate invasive ventilation in a subgroup of patients
with impending respiratory failure. More data are needed
to help define which acute respiratory failure patients are
most likely to benefit from noninvasive ventilation.
The importance of monitoring the patient-ventilator interface
is more fully appreciated today than ever before.
Optimizing patient-ventilator interaction is essential to minimizing
adverse effects. The use of HFOV for pediatric
ALI is now commonplace. However, HFOV is still often
started late in the course of pediatric ALI, and earlier
initiation of HFOV may help minimize ventilator-induced
lung injury and improve outcomes. As the use of HFV
continues to increase, this mode of ventilation should be
considered another form of conventional ventilation, as its
use is no longer “nonconventional.”
Many pediatric patients can be “liberated” from mechanical
ventilation without a long weaning process. Although
protocol-guided weaning has been successful with
adults, this has not been demonstrated to be true for pediatric
ALI patients. Recent data support the view that
there may be objective extubation predictors and criteria
for pediatric patients.
The most important issue affecting the field of pediatric
mechanical ventilation is the need for multicenter, randomized,
prospective studies. In the past decade the field
of pediatric mechanical ventilation has progressed dramatically.
With increasing research efforts this progress should
be anticipated to continue.

REFERENCES
1. Galen C. On the functions of parts of the human body. Daremberg C,
translator. Paris: JB Bailliere; 1954.
2. Baker AB. Artificial respiration, the history of an idea. Med Hist
1971;15(14):336–351.
3. Emerson H. Artificial respiration in the treatment of edema of the
lungs: a suggestion based on animal experimentation. Arch Intern
Med 1909;3:368.
4. Engstrom CG. Treatment of severe cases of respiratory paralysis by
the Engstrom universal respirator. Br Med J 1954;2:665–669.
5. Lassen HC. A preliminary report on the 1952 epidemic of poliomyelitis
in Copenhagen with special reference to the treatment of acute
respiratory insufficiency. Lancet 1953;i:37.
6. Aubier M, Viires N, Syllie G, Mozes R, Roussos C. Respiratory
muscle contribution to lactic acidosis in low cardiac output. Am Rev
Respir Dis 1982;126(4):648–652.
7. Ellis ER, Bye PT, Bruderer JW, Sullivan CE. Treatment of respiratory
failure during sleep in patients with neuromuscular disease:
positive pressure ventilation through a nose mask. Am Rev Respir
Dis 1987;135(1):148–152.
8. Kerby GR, Mayer LS, Pingleton SK. Nocturnal positive pressure ventilation
via nasal mask. Am Rev Respir Dis 1987;135(3):738–740.
9. Padman R, Lawless ST, Kettrick RG. Noninvasive ventilation via
bilevel positive airway pressure support in pediatric practice. Crit
Care Med 1998;26(1):169–173.
10. Fortenberry JD, Del Toro J, Jefferson LS, Evey L, Haase D. Management
of pediatric acute hypoxemic respiratory insufficiency with
bilevel positive pressure (BiPAP) nasal mask ventilation. Chest 1995;
108(4):1059–1064.
11. Hertzog JH, Siegel LB, Hauser GJ, Dalton HJ. Noninvasive positivepressure
ventilation facilitates tracheal extubation after laryngotracheal
reconstruction in children. Chest 1999;116(1):260–263.
12. Friedman O, Chidekel A, Lawless ST, Cook SP. Postoperative bilevel
positive airway pressure ventilation after tonsillectomy and
adenoidectomy in children –a preliminary report. Int J Pediatr Otorhinolaryngol
1999;51(3):177–180.
13. Serra A, Polese G, Braggion C, Rossi A. Non-invasive proportional
assist and pressure support ventilation in patients with cystic fibrosis
and chronic respiratory failure. Thorax 2002;57(1):50–54.
14. MacIntyre NR. Mechanical ventilation: the next 50 years. Respir
Care 1998;43(6):490–493.
15. Alvarez A, Subirana M, Benito S. Decelerating flow ventilation
effects in acute respiratory failure. J Crit Care 1998;13(1):21–25.
16. MacIntyre N, Nishimura M, Usada Y, Tokioka H, Takezawa J, Shimada
Y. The Nagoya conference on system design and patientventilator
interactions during pressure support ventilation. Chest 1990;
97(6):1463–1466.
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4 451
17. MacIntyre NR, Ho LI. Effects of initial flow rate and breath termination
criteria on pressure support ventilation. Chest 1991;99(1):134–138.
18. MacIntyre NR, Leatherman NE. Ventilatory muscle loads and the frequency-
tidal volume pattern during inspiratory pressure-assisted (pressure-
supported) ventilation. Am Rev Respir Dis 1990;141(2):327–331.
19. Mang H, Kacmarek RM, Ritz R, Wilson RS, Kimball WP. Cardiorespiratory
effects of volume- and pressure-controlled ventilation at
various I/E ratios in an acute lung injury model. Am J Respir Crit
Care Med 1995;151(3 Pt 1):731–736.
20. Marini JJ, Ravenscraft SA. Mean airway pressure: physiologic determinants
and clinical importance—Part 1: Physiologic determinants
and measurements. Crit Care Med 1992;20(10):1461–1472.
21. Marini JJ, Ravenscraft SA. Mean airway pressure: physiologic determinants
and clinical importance—Part 2: Clinical implications.
Crit Care Med 1992;20(11):1604–1616.
22. Boros SJ. Variations in inspiratory:expiratory ratio and airway pressure
waveform during mechanical ventilation: the significance of
mean airway pressure. J Pediatr 1979;94(1):114–117.
23. Abraham E, Yoshihara G. Cardiorespiratory effects of pressure-controlled
ventilation in severe respiratory failure. Chest 1990;98(6):
1445–1449.
24. Tobin MJ. Respiratory monitoring in the intensive care unit. Am Rev
Respir Dise 1988;138(6):1625–1642.
25. Slutsky AS. Mechanical ventilation. American College of Chest Physicians’
Consensus Conference. Chest 1993;104(6):1833–1859.
26. Rosen WC, Mammel MC, Fisher JB, Coleman JM, Bing DR, Holloman
KK, Boros SJ. The effects of bedside pulmonary mechanics
testing during infant mechanical ventilation: a retrospective analysis.
Pediatr Pulmonol 1993;16(3):147–152.
27. Randolph AG, Wypij, D, Venkataraman ST, Hanson JH, Gedeit RG,
Meert KL, et al. Effect of mechanical ventilator weaning protocols
on respiratory outcomes in infants and children: a randomized controlled
trial. JAMA 2002;288(20):2561–2568.
28. Ventilation with lower tidal volumes as compared with traditional
tidal volumes for acute lung injury and the acute respiratory distress
syndrome. The Acute Respiratory Distress Syndrome Network. New
Engl J Med 2000;342(18):1301–1308.
29. Wilson BG, Kern FH, Cheifetz IM, Meliones JN. Direct measurement
via an inline pneumotach is necessary to determine effective
tidal volume in children (abstract). Respir Care 1995;40(11):1172.
30. Cannon ML, Cornell J, Tripp-Hamel DS, Gentile MA, Hubble CL,
Meliones JN, Cheifetz IM. Tidal volume for ventilated infants should
be determined with a pneumotachometer placed at the endotracheal
tube. Am J Respir Crit Care Med 2000;162(6):2109–2112.
31. Dreyfuss D, Soler P, Basset G, Saumon G. High inflation pressure
pulmonary edema: respective effects of high airway pressure, high
tidal volume, and positive end-expiratory pressure. Am Rev Respir
Dis 1988;137(5):1159–1164.
32. Papadakos PJ, Apostolakos MJ. High-inflation pressure and positive
end-expiratory pressure. Injurious to the lung? Yes. Crit Care Clin
1996;12(3):627–634.
33. Parker JC, Hernandez LA, Peevy KJ. Mechanisms of ventilatorinduced
lung injury. Crit Care Med 1993;21(1):131–143.
34. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from
experimental studies. Am J Respir Crit Care Med 1998;157(1):294–323.
35. Mead J, Takishima T, Leith D. Stress distribution in lungs: a model
of pulmonary elasticity. J Appl Physiol 1970;28(5):596–608.
36. Carlton DP, Cummings JJ, Scheerer RG, Poulain FR, Bland RD.
Lung overexpansion increases pulmonary microvascular protein permeability
in young lambs. J Appl Physiol 1990;69(2):577–583.
37. Hickling KG, Henderson SJ, Jackson R. Low mortality associated
with low volume pressure limited ventilation with permissive hypercapnia
in severe adult respiratory distress syndrome. Intensive Care
Med 1990;16(6):372–377.
38. Rippe B, Townsley M, Thigpen J, Parker JC, Korthuis RJ, Taylor
AE. Effects of vascular pressure on the pulmonary microvasculature
in isolated dog lungs. J Appl Physiol 1984;57(1):233–239.
39. Brunet F, Jeanbourquin D, Monchi M, Mira JP, Fierobe L, Armaganidis
A, et al. Should mechanical ventilation be optimized to blood
gases, lung mechanics, or thoracic CT scan? Am J Respir Crit Care
Med 1995;152(2):524–530.
40. Wilson BG. Using airway graphics to optimize mechanical ventilation
in neonates with respiratory distress syndrome. Neonatal Netw
1997;16(3):71–75.
41. Cheifetz IM, Craig DM, Quick G, McGovern JJ, Cannon ML, Ungerleider
RM, et al. Increasing tidal volumes and pulmonary overdistention
adversely affect pulmonary vascular mechanics and cardiac
output in a pediatric swine model. Crit Care Med 1998;26(4):
710–716.
42. Wetzel RC, Gioia FR. High frequency ventilation. Pediatr Clin North
Am 1987;34(1):15–38.
43. Lunkenheimer PP, Rafflenbeul W, Keller H, Frank I, Dickhut HH,
Fuhrmann C. Application of transtracheal pressure oscillation as modification
of “diffusion respiration.” Br J Anaesth 1972;44(6):627.
44. Kinsella JP, Gerstmann DR, Clark RH, Null DM Jr, Morrow WR,
Taylor AF, deLemos RA. High-frequency oscillatory ventilation versus
intermittent mandatory ventilation: early hemodynamic effects in
the premature baboon with hyaline membrane disease. Pediatr Res
1991;29(2):160–166.
45. Arnold JH, Hanson JH, Toro-Figuero LO, Gutierrez J, Berens RJ,
Anglin DL. Prospective, randomized comparison of high-frequency
oscillatory ventilation and conventional mechanical ventilation in
pediatric respiratory failure. Crit Care Med 1994;22(10):1530–1539.
46. Clark RH, Gerstmann DR, Null DM Jr, deLemos RA. Prospective
randomized comparison of high-frequency oscillatory and conventional
ventilation in respiratory distress syndrome. Pediatrics 1992;
89(1):5–12.
47. Froese AB. High-frequency oscillatory ventilation for adult respiratory
distress syndrome: let’s get it right this time! Crit Care Med
1997;25(6):906–908.
48. Arnold JH, Anas NG, Luckett P, Cheifetz IM, Reyes G, Newth CJL, et
al. High frequency oscillatory ventilation in pediatric respiratory failure:
a multicenter experience. Crit Care Med 2000;28(12):3913–3919.
49. Hall JB, Wood LDH. Liberation of the patient from mechanical
ventilation. JAMA 1987;257(12):1621–1628.
50. Manthous CA, Schmidt GA, Hall JB. Liberation from mechanical
ventilation: a decade of progress. Chest 1998;114(3):886–901.
51. Tobias JD, Deshpande JK, Gregory DF. Outpatient therapy of
iatrogenic drug dependency following prolonged sedation in the
pediatric intensive care unit. Intensive Care Med 1994;20(7):504–
507.
52. Orlowski JP, Ellis NG, Amin NP, Crumrine RS. Complications of
airway intrusion in 100 consecutive cases in a pediatric ICU. Crit
Care Med 1980;8(6):324–331.
53. Benjamin PK, Thompson JE, O’Rourke PP. Complications of mechanical
ventilation in a children’s hospital multidisciplinary intensive
care unit. Respir Care 1990;35(9):873–878.
54. Pierson DJ. Complications of mechanical ventilation. In: Simmons
DH, editor. Current pulmonology. Vol 9. Chicago: Yearbook Medical
Publishers; 1990:19–46.
55. Randolph AG. Weaning from mechanical ventilation. New Horiz
1999;7:374–385.
56. Harris TR, Wood BR. Physiologic principles. In: Karotkin EH, Goldsmith
JP, editors. Assisted ventilation of the neonate. Philadelphia:
WB Saunders; 1996:29–31.
57. Gluck EH, Barkoviak MJ, Balk RA, Casey LC, Silver MR, Bone
RC. Medical effectiveness of esophageal balloon pressure manom-
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
452 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4
etry in weaning patients from mechanical ventilation. Crit Care Med
1995;23(3):504–509.
58. Kollef MH, Shapiro SD, Silver P, St John RE, Prentice D, Sauer S,
et al. A randomized, controlled trial of protocol-directed versus physician-
directed weaning from mechanical ventilation. Crit Care Med
1997;25(4):567–574.
59. Torres A, Gatell JM, Aznar E, el-Ebiary M, Puig de la Bellacasa J,
Gonzalez J, et al. Re-intubation increases the risk of nosocomial
pneumonia in patients needing mechanical ventilation. Am J Respir
Crit Care Med 1995;152(1):137–141.
60. Epstein SK, Ciubotaru RL, Wong JB. Effect of failure extubation on
the outcome of mechanical ventilation. Chest 1997;112(1):186–192.
61. Esteban A, Inmaculada A, Gordo F, Fernandez R, Solsona JF, Vallverdu I,
et al. Extubation outcome after spontaneous breathing trials with T-tube or
pressure support ventilation. The Spanish Lung Failure Collaborative Group.
Am J Respir Crit Care Med 1997;156(2 Pt 1):459–465.
62. Daley BJ, Garcia-Perez F, Ross SE. Reintubation as an outcome
predictor in trauma patients. Chest 1996;110(6):1577–1580.
63. Epstein SK, Ciubotaru RL. Independent effects of etiology of failure
and time to reintubation on outcome for patients failing extubation.
Am J Respir Crit Care Med 1998;158(2):489–493.
64. Rady MY, Ryan T. Perioperative predictors of extubation failure and
the effect on clinical outcome after cardiac surgery. Crit Care Med
1999;27(2s):340–347.
65. Demling RH, Read T, Lind LJ, Flanagan HL. Incidence and morbidity
of extubation failure in surgical intensive care patients. Crit
Care Med 1988;16(6):573–577.
66. Yang KL, Tobin MJ. A prospective study of indexes predicting the
outcome of trials of weaning from mechanical ventilation. N Engl
J Med 1991;324(24):1445–1450.
67. Tahvanainen J, Salmenpera M, Nikki P. Extubation criteria after
weaning from intermittent mandatory ventilation and continuous positive
airway pressure. Crit Care Med 1983;11(9):702–707.
68. Sahn SA, Lakshminarayan S. Bedside criteria for discontinuation of
mechanical ventilation. Chest 1973;63(6):1002–1005.
69. Edmunds S, Weiss I, Harrison R. Extubation failure in a large pediatric
ICU population. Chest 2001;119(3):897–900.
70. Khan N, Brown A, Venkataraman ST. Predictors of extubation success
and failure in mechanically ventilated infants and children. Crit
Care Med 1996;24(9):1568–1579.
71. Hubble CL, Gentile MA, Tripp DS, Craig DM, Meliones JN, Cheifetz
IM. Dead space to tidal ventilation ratio predicts successful extubation
in infants and children. Crit Care Med 2000;28(6):2034–2040.
72. Burki NK. The dead space to tidal volume ratio in the diagnosis of
pulmonary embolism. Am Rev Respir Dis 1986;133(4):679–685.
73. Eriksson L, Wollmer P, Olsson CG, Albrechtsson U, Larusdottir H,
Nilsson RS, et al. Diagnosis of pulmonary embolism based upon
alveolar dead space analysis. Chest 1989;96(2):357–362.
74. Severinghaus JW, Stupfel M. Alveolar dead space as an index of
distribution of blood flow in pulmonary capillaries. J Appl Physiol
1957;10:335–348.
75. Manthous CA, Goulding P. The effect of volume infusion on dead
space in mechanically ventilated patients with severe asthma. Chest
1997;112(3):843–846.
76. Ashutosh K, Dev G, Steele D. Nonbronchodilator effects of pirbuterol
and ipratropium in chronic obstructive pulmonary disease.
Chest 1995;107(1):173–178.
77. Arnold JH, Bower LK, Thompson JE. Respiratory deadspace measurements
in neonates with congenital diaphragmatic hernia. Crit
Care Med 1995;23(2):371–375.
78. Fletcher R. Relationship between alveolar deadspace and arterial
oxygenation in children with congenital cardiac disease. Br J Anaesth
1989;62(2):168–176.
79. Arnold JH, Thompson JE, Benjamin PK. Respiratory deadspace measurements
in neonates during extracorporeal membrane oxygenation.
Crit Care Med 1993;21(12):1895–1900.
80. el Khatib MF, Baumeister B, Smith PG, Chatburn RL, Blumer JL.
Inspiratory pressure/maximal inspiratory pressure: does it predict
successful extubation in critically ill infants and children? Intensive
Care Med 1996;22(3):264–268.
81. Thiagarajan RR, Bratton SL, Martin LD, Brogan TV, Taylor D.
Predictors of successful extubation in children. Am J Respir Crit
Care Med 1999;160(5 Pt 1):1562–1566.
82. DeHaven CB, Hurst JM, Branson RD. Evaluation of two different
extubation criteria: attributes contributing to success. Crit Care Med
1986;14(2):92–94.
83. Ely EW, Baker AM, Dunagan DP, Burke HL, Smith AC, Kelly PT,
et al. Effect on the duration of mechanical ventilation of identifying
patients capable of breathing spontaneously. N Engl J Med 1996;
335(25):1864–1869.
84. Leitch EA, Moran JL, Brealy B. Weaning and extubation in the
intensive care unit: clinical or index-driven approach? Intensive Care
Med 1996;22(8):752–759.
Discussion
Donn: Very nice presentation, Ira. I
would mirror your comments about
weaning and extubation as they apply
to neonatal and mechanical ventilation.
I think if you look in the index of
either of the 2 leading textbooks on
neonatal/perinatal medicine, you don’t
find the word “weaning” appearing at
all. Maybe part of the issue with the
big trial that you presented is that it
was a trial.
What I have found is a parallel with
what we were all taught as pediatric
residents—if you think about a spinal
tap, you ought to do it. Weaning is the
same way. You have to think about it.
What we try to convey to our pediatric
trainees is that weaning begins immediately
after intubation. The idea is
to get the patient off the ventilator as
rapidly as possible, but, obviously, without
jeopardizing well-being in the postextubation
phase.
We’ve seen a very dramatic change
in our very-low-birth-weight babies;
in the past there was enormous reluctance
to extubate a baby who was
1,000 g, for reasons that totally baffle
me. But now we’re seeing 600–800 g
babies extubated very earlyin thecourse
of the disease and maintained
on continuous positive airway pressure
or nasal cannula oxygen, with surprisingly
good success, so I think it’s
still our last frontier. But the take-home
message is, you’ve got to think about
it to do it.
Cheifetz: I fully agree with you. In
the weaning study by Randolph et al1
no difference was found between protocol
weaning and non-protocol weaning.
Your point is excellent. There
were a substantial number of inclusion
and exclusion criteria, and the
subgroup of patients studied might be
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4 453
a relatively small subset of the total
group of patients. Additionally, upon
entry into the study the patients already
had resolution of the acute phase
of the illness. So, I agree, the results
of any study really depend on the details
of the specific population you are
investigating, how you are studying
the question, and how you extrapolate
data from one study to all of pediatrics.
REFERENCE
1. Randolph AG, Wypij D, Venkataraman ST,
Hanson JH, Gedeit RG, Meert KL, et al.
Effect of mechanical ventilator weaning
protocols on respiratory outcomes in infants
and children: a randomized, controlled
trial. JAMA 2002;288(20):2561–2568.
Kercsmar: You mentioned the importance
of using NIV, or at least trying
it, and that one advantage is the
possibility of using NIV outside of the
ICU; it might be less expensive, more
comfortable for the patient, and offer
more options.Onedifficultywe’ve had
is that that’s often easier said than
done. At our institution the rules require
that patients who might need various
forms of noninvasive mechanical
ventilation must go to the ICU unless
they are at the chronic and stable stage.
Would you expand a bit about NIV
criteria and what you mean by “sites
outside of the ICU” that would permit
safe and effective use of NIV?
Cheifetz: It is difficult to set exact
criteria of what can be done in the
various clinical care locations within
a hospital. Early in our NIV program
we did all of our NIV in the ICUs.
Now we also use NIV in our stepdown
unit, our pediatric wards, and in
our bone marrow transplant ward. So
we’ve expanded NIV out of the ICU
to more effectively utilize our resources.
And we do have objective criteria
for the use of NIV in these various
settings. Patients outside the ICU must
be clinically stable. They cannot be
requiring increasing noninvasive support.
Any increase in support beyond
minimal titrations warrants a trip to
the ICU. We also have FIO2
requirements.
Any patient who has an escalating
FIO2
requirement or an FIO2
50%
must be moved into the ICU. Beyond
the ICU, noninvasive ventilation must
be used as a respiratory assistance device
and not as a life support device.
Or stated differently, the non-ICU patient
receiving NIV must be able to
tolerate disconnection from the ventilator
for a reasonable period of time.
The use of NIV outside the ICU requires
protocols and guidelines to provide
safe and effective care.
Black: Regarding weaning criteria,
the rapid shallow breathing index
that’s commonly used with adults
seems to work very well with all different
situations where intubation and
mechanical ventilation are required,
including lung disease, trauma, closed
head injury, and others conditions. The
majority of intubated patients in our
pediatric ICU have closed head injuries
from motor vehicle accidents. Do
you think VD/VT will work with those
patients?
Cheifetz: With adult patients the
rapid shallow breathing index works
extremely well for predicting successful
extubation. In pediatrics it fails
miserably because there are so many
additional variables that affect respiratory
rate, including the patient’s fear
when awakening in a strange setting.
So I don’t think the rapid shallow
breathing index is useful in pediatrics.
In terms of the VD/VT one of the
key points concerning predicting the
success of extubation is that it only
considers the pulmonary process.
VD/VT simply provides an indication
of the resolution of the pulmonary disease.
In a trauma patient with a severe
pulmonary contusion, I believe VD/VT
will be an excellent marker for the
likelihood of extubation success.1
However, in a patient with a closed
head injury, in which the primary issue
is neurologic, the VD/VT will not
be useful at all.
REFERENCE
1. Hubble CL, Gentile MA, Tripp DS, Craig
DM, Meliones JN, Cheifetz IM. Deadspace
to tidal volume ventilation ratio predicts
successful extubation in infants and children.
Crit Care Med 2000;28(6):2034–
2040.
Myers: Referring to the measurement
of pressure, volume, and flow,
in some of the studies that we’ve done,
pumping gas from a calibrated syringe
through a pneumotachograph, adult
pneumotachographs seem to be fairly
accurate and have good precision. The
smaller, infant pneumotachographs,
while they’re very precise, they all
seem to have a built-in inaccuracy to
them, which scares me about using
volume-targeted ventilation in the neonatal
ICU.
The second issue is that in the majority
of our patients we’re using uncuffed
ETTs, so there is an air leak
between the ETT and trachea. Where
is the cutoff point at which we should
stop believing all the pulmonary mechanics
measurements (compliance
and resistance) with which we’re trying
to make treatment decisions? We
often have patients who look much
better from the perspective of pulmonary
mechanics, but if the system has
a 35% leak, then the pulmonary mechanics
monitor is practically a random
number generator!
Cheifetz: Those are important clinical
issues. The clinician must consider
the detailed specifications and
accuracy of the monitoring device.
Most of the pneumotachometers that
we use in the pediatric ICU have accuracy
and precision well within clinical
acceptability. With neonates and
small premature infants I don’t have
enough experience to comment on
whether the devices are accurate or
precise enough. In terms of air leak it
is a difficult question, a huge ques-
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
454 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4
tion. The data I presented about VT
measurements specifically represented
exhaled volumes, to avoid the
issue of air leak. The underlying question
is, what is an acceptable air leak?
I think if we ask everyone in this room,
“what is an acceptable air leak?” we
would probably have 20 different answers
concerning (1) clinical management
and (2) respiratory mechanics
measurements.
A question you did not mention is,
when do you monitor respiratory mechanics
in small neonates? When you
consider some of the variables in this
population, it becomes apparent that
the compliance of the ventilator circuit
can be greater than the compliance
of the patient’s lungs, so it becomes
a difficult question. There are a
huge number of questions and research
projects that need to be answered before
we can come to any kind of conclusion.
Hansell: With HFOV, especially
with larger patients, when we get into
high distending pressures, we increase
the potential to damage the lungs,
which releases many mediators that
actually increase the negative effect
of being septic. In septic patients we
may actually be increasing the morbidity
and mortality if we use HFOV
improperly. I think that is another reason
we ought to consider implementing
HFOV very early in the course of
disease.
The other factor is that P (the
change in pressure) is less attenuated
as we get into larger ETTs. The larger
the ETT, the more like conventional
ventilation HFOV becomes. Whether
that is important and whether we
should put them on HFOV, I don’t
know. Nevertheless, I think we need
to be aware that when we use large
P and high mean airway pressure we
may actually be closer to ventilating
them as we would if they were on a
conventional ventilator at a high respiration
rate.
Cheifetz: That is an important point.
You can’t really compare a large adult
patient who’s using a large ETT to a
small infant who’s using a small ETT,
because the changes in pressure can
be attenuated much more dramatically
in a smaller patient. I would say, and
we have investigated this in bench
studies, that even with large ETTs and
large amplitudes on HFOV, pressure
amplitude is still dramatically attenuated.
Although P is larger than in
very small patients, the P that is delivered
to the adult ARDS patient is
still going to be dramatically less than
the P on a conventional ventilator,
any way you look at it. So I still think
HFOV is a lung-protective strategy in
all patient populations—all sizes and
all ages.
Salyer: I think there is compelling
evidence from animal tests and growing
evidence from tests with humans
that what causes ventilator-induced
lung injury, at least the mechanical
injury, as opposed to biochemical injury,
is overdistention of the lung,
which is a volume-related phenomenon.
If you give the same VT with 2
different flow patterns, at the end of
inspiration you have the same volume
in the lung, and the differences in airway
pressure between the 2 flow patterns
are just a result of the resistance
to flow during the breath. So it’s unclear
to me why such a reduction in
peak pressure would offer any benefit
to the patient.
Cheifetz: There are a couple of issues
here. Yes, I agree that if you overdistend
the lungs, you can cause volutrauma.
But if you are attempting to
compare 2 different flow patterns at
the same VT, and that VT is within
acceptable limits (ie, the lungs are not
overdistended), then the issue is probably
different. If you can deliver the
same 6 mL/kg VT at a lower peak
inspiratory pressure, I think you are
less likely to cause barotrauma and
secondary lung injury. I must admit,
though, that I don’t have convincing
data to support this theoretical point,
and I do not know if there are data in
the literature that address the issue. It
begs further study.
Rotta: I second your enthusiasm for
HFOV for pediatric ARDS patients.
HFOV is not a fad and it is not a
nonconventional strategy. Centers that
have used HFOV for a while are comfortable
with it and are using it earlier
and earlier in the course of disease,
and are seeing good results, as you
see in your service at Duke and as I
see in Buffalo. I think the problem we
are seeing now in gaining more acceptance
of HFOV is that centers that
are just beginning to use it are going
through the problems of learning and
mastering the new technology—technology
that when not used properly
can give results that are interpreted as
bad outcomes, such as hypoxemia. In
addition, centers that are reluctant to
start HFOV until the patient is moribund
will continue to see bad results
because HFOV will not resurrect
someone who is near death.
Now, addressing the previous comment
on whether you still have lung
protection during HFOV with the bigger
patient, who has lower respiratory
rate and a larger ETT, there are now
good data suggesting that lung protection
persists in adult ARDS patients
ventilated with the SensorMedics
3100B ventilator.1 These adult patients
are being ventilated using the same
principles that have been applied to
neonatal and pediatric patients for
years.
REFERENCE
1. Derdak S, Mehta S, Stewart TE, Smith T,
Rogers M, Buchman TG et al; The Multicenter
Oscillatory Ventilation For Acute
Respiratory Distress Syndrome Trial
(MOAT) Study Investigators. High-frequency
oscillatory ventilation for acute respiratory
distress syndrome in adults: a randomized,
controlled trial. Am J Respir Crit
Care Med 2002;166(6):801–808.
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4 455
Cheifetz: I agree. The important
point, as you mentioned, is education.
The SensorMedics 3100B is not a new
device: it was originally FDA-approved
in 1991. However, with any
device new to a specific institution
there will be a learning curve. I would
say again that I think HFOV is today
a conventional ventilation mode, but
centers that don’t have substantial experience
with HFOV will have to learn
to apply it most efficiently.
Wagener: I totally agree that highfrequency
should be considered conventional
ventilation now; about the
terminology we could argue one way
or another. But I’d point out that it’s
not appropriate to extrapolate from a
limited-size study in a limited number
of centers and with a select population
and make the statement that no
further randomized, controlled study
needs to be done, especially knowing
that in the adult studies of HFOV there
has not been the success that we’ve
seen with babies. And in pediatrics
we’re covering the whole spectrum of
patients in between. It may be that we
have certain situations that were selected
for in that reasonably planned
study in which HFOV was effective,
and we also have other situations that
were not included in that study for
which it’s not going to be proven as
effective. So HFOV, whether you call
it conventional or something else, it’s
one standard form of ventilation.
Cheifetz: I need to clarify something
I said. I do believe it is important
to have another larger, multicenter,
randomized, prospective,
controlled trial investigating HFOV
for pediatric ARDS. I think such a
study would be important, especially
if both the intervention and control
groups used alveolar recruitment maneuvers.
My point is that I don’t believe
that will ever occur. I do not
think that there will be enough centers
with expertise in HFOV that would
agree to participate in a randomized,
controlled trial, without crossover. I
am not sure if a crossover trial would
fully address your concerns.Somecenters
might raise ethical concerns with
randomizing patients away from a
technique (ie, HFOV) that in clinical
practice and in the medical literature
is an approved, established therapy.
Whether that is right or wrong I do
not know.
Wiswell: The other side of the ethical
issue is that if we don’t do that
trial, many centers are not going to
start using HFOV, because they will
not believe there is adequate evidence
that it works, and so their patients will
not receive the benefit of our knowledge
that HFOV does work and ought
to be the standard of care. Though I’m
a firm believer in HFOV, I’m not sure
the existing data are going to convince
a lot of people thatHFOVworks. Steve
Donn and I are long-time New York
Yankee fans and remember the saying
of Yogi Berra, “It’s de´ja` vu all over
again.” I say that because embracing
new treatment technologies, such as
high frequency oscillation, without
validation by randomized, controlled
trials has happened all too frequently
in neonatology.
Cheifetz: This is a hard issue. How
do you perform a study with a technique
that has become an accepted
standard clinical practice? I am not
saying it is correct that it is an accepted,
standard clinical practice. But
once a technique is widely accepted,
it is hard to convince enough centers
to go back and study it.
Wagener: Maybe at your center that
is true, but remember that there are
other centers at which it’s not standard
practice and a study could be performed.
Donn: Aren’t we lucky that Alexander
Fleming’s first patient didn’t develop
anaphylaxis to penicillin?
Rotta: I think this is going to be
one of those cases when we pediatricians
are going to follow the results of
adult studies that learned from techniques
that were applied in pediatrics
first.
Cheifetz: Let me go out on a limb
here and ask a question of everyone in
the room. If there were a proposed
randomized, controlled, prospective
trial of HFOV versus conventional
ventilation in pediatric ALI/ARDS
without crossover (which is what the
study design probably would need),
who here would enroll patients, knowing
that your patient might be randomized
to the control group (ie, could
not use HFOV)? Who here would do
that?
Wiswell: If I were a pulmonologist
and not a neonatologist, I would.
Cheifetz: Who would like to answer
my question? One hand. Two? Just a
couple. It is a small minority of the
people in the room.
Wiswell: I’ve been involved in a lot
of large randomized, controlled trials
that have examined therapies that my
colleagues and I truly believed in,
but—damn!—the randomized, controlled
trial showed there was no difference
between the “magical” new
therapy and controls! The marvelous
thing about a large, randomized,
controlled trial is that if there is a
difference, you’re going to see it.
But if there’s not, you’re going to
see that there’s not. Equally important
is that potential complications
are going to rear their ugly heads
too. I was in Texas in 1984, in the
baboon lab helping develop the firstgeneration
high-frequency oscillator,
and I’m a firm believer in it as
an effective therapy. But you’ve got
to prove it is effective, and you’ve
got to prove it on a large scale.
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
456 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4
Cheifetz: Fine. But for HFOV there
are randomized, controlled, prospective
studies. There are studies in the neonatal
population. In John Arnold’s study,
admittedly, the number of patients was
relatively small, but it is a good, randomized,
controlled trial.1,2 And recently
published adult studies support
the use of HFOV in adults.3–6 It’s not as
if oscillation is being used without any
randomized, controlled studies. There
just has not been a large pediatric study.
The published pediatric investigation
was smaller, and we may have to extrapolate
data for pediatrics from the
neonatal and adult populations, as Dr
Rotta mentioned earlier.
REFERENCES
1. Arnold JH, Hanson JH, Toro-Figuero LO,
Gutierrez J, Berens RJ, Anglin DL. Prospective,
randomized comparison of highfrequency
oscillatory ventilation and conventional
mechanical ventilation in
pediatric respiratory failure. Crit Care
Med 1994;22(10):1530–1539.
2. Arnold JH, Anas NG, Luckett P, Cheifetz
IM, Reyes G, Newth CJ et al. High-frequency
oscillatory ventilation in pediatric
respiratory failure: a multicenter experience.
Crit Care Med 2000;28(12):3913–
3919.
3. Clark RH, Gerstmann DR, Null DM Jr,
deLemos RA. Prospective randomized
comparison of high-frequency oscillatory
and conventional ventilation in respiratory
distress syndrome. Pediatrics 1992;89(1):
5–12.
4. Gerstmann DR, Minton SD, Stoddard RA,
Meredith KS, Monaco F, Bertrand JM et
al. The Provo multicenter early high-frequency
oscillatory ventilation trial: improved
pulmonary and clinical outcome in
respiratory distress syndrome. Pediatrics
1996;98(6 Pt 1):1044–1057.
5. Derdak S, Mehta S, Stewart TE, Smith T,
Rogers M, Buchman TG, et al. High-frequency
oscillatory ventilation for acute respiratory
distress syndrome in adults: a randomized,
controlled trial. Am J Respir Crit
Care Med 2002;166(6):801–808.
6. Mehta S, Lapinsky SE, Hallett DC, Merker
D, Groll RJ, Cooper AB, et al. Prospective
trial of high-frequency oscillation in adults
with acute respiratory distress syndrome.
Crit Care Med 2001;29(7):1360–1369.
Rotta: Once you start studying the
effect of more than one lung-protective
strategy in a clinical trial, it is
even harder to show that one strategy
is better than another. That’s why, for
instance, liquid ventilation is not approved
and probably will not be approved,
since it has been studied in
the era of lung-protective conventional
ventilation. It is very hard to show
separation between 2 groups in clinical
trials when both are subjected to
some form of lung protection.
Just for illustration, in the successful
ARDS Network trial,1 although the
entry VT was approximately 10 mL/
kg, the low-VT group received 6 mL/
kg, whereas the VT in the conventional
treatment group was increased to 12
mL/kg. This was not done by chance,
but to provide separation between the
2 groups,1 which had not occurred in
a previous trial.2
In the laboratory HFOV does not
appear to be superior (purely from a
lung-injury standpoint3,4) to a conventional
ventilation strategy using the
open-lung approach used by Dr Amato
in Brazil,5 although animals treated
with HFOV have more stable hemodynamics.
3,4 Throw these 2 strategies
into a clinical trial and your differences
get even more diluted. Compare
HFOV with the conventional ventilation
(control) group of the Amato trial5
and HFOV would probably look
really good. But no one would do that
study now. It is all a matter of timing.
REFERENCES
1. Ventilation with lower tidal volumes as
compared with traditional tidal volumes for
acute lung injury and the acute respiratory
distress syndrome. The Acute Respiratory
Distress Syndrome Network. N Engl J Med
2000; 342(18):1301–1308.
2. Stewart TE, Meade MO, Cook DJ, Granton
JT, Holder RV, Lapinsky SE, et al. Evaluation
of a ventilation strategy to prevent
barotrauma in patients at high risk for acute
respiratory distress syndrome. Pressure- and
Volume-Limited Ventilation Strategy
Group. N Engl J Med 1998;338(6):355–
361.
3. Imai Y, Nakagawa S, Ito Y, Kawano T,
Slutsky AS, Miyasaka K. Comparison of
lung protection strategies using conventional
and high-frequency oscillatory ventilation.
J Appl Physiol 2001;91(4):1836–
1844.
4. Rotta AT, Gunnarsson B, Fuhrman BP, Hernan
LJ, Steinhorn DM. Comparison of lung
protective ventilation strategies in a rabbit
model of acute lung injury. Crit Care Med
2001;29(11):2176–2184.
5. Amato MB, Barbas CS, Medeiros DM, Magaldi
RB, Schettino GP, Lorenzi-Filho G,
et al. Effect of a protective-ventilation strategy
on mortality in the acute respiratory
distress syndrome. N Engl J Med 1998;
338(6):347–354.
Wagener: But you can’t say that
there’s one approach that has a clear
advantage over another until you have
tested your hypothesis.
Cheifetz: That’s correct, but what
I’m saying is that HFOV should be
considered as another mode of mechanical
ventilation. I think everyone
in the room would agree that there has
not been any published study that demonstrates
that a particular mode of ventilation,
whether it be volume-control,
pressure-support, pressure-regulated
volume-controlled, or pressure-control,
significantly affects outcome for
a given patient population. My point
is that the oscillator should be viewed
as another mode of standard ventilation,
not a nonconventional “rescue”
or “heroic” therapy. It should be
viewed as a conventional ventilation
therapy. You are correct, Dr Rotta: if
we performed a head-to-head comparison
of HFOV and the open-lung conventional
strategy, no one knows what
the results would reveal. But from the
available published studies and clinical
experience, I do not believe there
are any important adverse effects associated
with HFOV. There is obviously
a fair amount of debate and controversy
in this room, and I would then
go out on another limb and challenge
someone in this room to coordinate
the study. I am a little skeptical about
how many centers would participate and
how many patients would be enrolled.
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4 457
Donn: Maybe what we need to do
is just expunge the word “conventional.”
In the past it was used to
talk about things that were done conventionally.
What we need to do is
just talk about tidal ventilation versus
nontidal ventilation, and then we get
away from what’s conventional and
what isn’t.
Cheifetz: I guess the biggest problem
I have with the whole discussion
is the use of that term “nonconventional.”
The take-home message
I want to send is that HFOV is no
longer nonconventional ventilation.
It is conventional ventilation.
Black: I don’t think you’re ever
going to get away from the use of
the term “nonconventional” because,
let’s face it, when you use what we
call “conventional” ventilation, we
do everything we can to make that
conventional ventilation mimic our
own natural spontaneous ventilation.
And there is nothing about HFOV
(unless you want to talk about panting
dogs) that mimics normal spontaneous
ventilation. But that’s not to
say that it isn’t a superior therapeutic
technique in certain clinical situations.
I’m a very strong believer
in early use of HFV.
Cheifetz: Let me comment on that
point before you continue. My comment
is, if you think about spontaneous
normal breathing, everyone in
this room is breathing using what?
Negative-pressure ventilation! So, to
use your definition, positive-pressure
“conventional” ventilation is really
nonconventional!
Black: Well, you’re right—absolutely
correct! But it’s closer to conventional
ventilation than HFV. I
think a clinical trial of that nature,
without the potentialhttp://www.rcjournal.com/contents/04.03/04.03.0442.pdf for crossover,
may border on unethical today. The
study that you showed, even with
the crossovers, showed very clear
statistically significant results. My
point is that with sophisticated statistical
techniques, things like crossovers,
which obviously do muddy
the waters, can be gotten around.
There are also statistical techniques
that allow you to continuously analyze
the data as they are being gathered,
and when you reach the point
of significance, you can stop the
study. These techniques were widely
used by the pharmaceutical industry,
but they haven’t really made their
way into testing of ventilators.
INVASIVE AND NONINVASIVE PEDIATRIC MECHANICAL VENTILATION
458 RESPIRATORY CARE • APRIL 2003 VOL 48 NO 4