N a s a l H i g h - F re q u e n c y

Ven t il ati o n
a,b, a,b
Daniele De Luca, MD, PhD *, Roberta Centorrino, MD

KEYWORDS

Neonate
Noninvasive
Oscillation
Percussive
Rescue
Respiratory support

Interface
Newborn infant

KEY POINTS

The role of interface during noninvasive high-frequency ventilatory modes is very impor-
tant and physical characteristics of different interfaces must be known to optimize their
use.

NHFOV needs to be used within a physiology-driven protocol with accurate mini-invasive
multimodal monitoring and adequate nurse training. A protocol proposal is enclosed.

NHFOV may be useful to reduce PaCO2 and spare intubation and invasive ventilation in ne-
onates with CPIP (ie, evolving BPD).

Future trials about NHFOV need to be more explanatory and physiology-based.

There is less experience with NHFPV, although it might be useful for TTN.

INTRODUCTION

The term “noninvasive high-frequency ventilation” designates a nonconventional

ventilatory technique with supraphysiologic frequencies applied through an external
noninvasive interface (nasal prongs, helmet, or various types of mask), thus without
endotracheal intubation or tracheostomy. It is out of our scope to discuss whether
conventional or nonconventional modalities are generally preferable but we will review
the data regarding noninvasive high-frequency ventilations available in neonatology
and their possible benefits.
Within this technique, we may recognize 2 modalities:

Noninvasive high-frequency oscillatory ventilation (NHFOV),

Noninvasive high-frequency percussive ventilation (NHFPV).

a
Division of Pediatrics and Neonatal Critical Care, “A.Beclere” Medical Center, Paris Saclay
University Hospitals, APHP, Paris - France; b Physiopathology and Therapeutic Innovation Unit-
INSERM U999, Paris Saclay University, Paris - France
* Corresponding author. Service de Pédiatrie et Réanimation Néonatale, Hôpital “A. Béclère”-
GHU Paris Saclay, APHP, 157 rue de la Porte de Trivaux, Clamart Paris-IDF 92140, France.
E-mail address: dm.deluca@icloud.com

Clin Perinatol 48 (2021) 761–782

https://doi.org/10.1016/j.clp.2021.07.006 perinatology.theclinics.com
0095-5108/21/ª 2021 Elsevier Inc. All rights reserved.
762 De Luca & Centorrino

Only scanty exist about NHFPV, whereas NHFOV is quite often used in some coun-
tries.1 The diffusion of NHFOV is likely due to the wide experience about the use of
endotracheal high-frequency oscillatory ventilation (HFOV) and nasal continuous pos-
itive airway pressure (CPAP) in preterm neonates. The experience in HFOV and CPAP
has pushed clinicians to combine them to maximize their advantages (such as nonin-
vasive interface, increase in functional residual capacity determining oxygenation
improvement, no need for synchronization, efficient CO2 removal).

PHYSIOLOGY OF NON-INVASIVE HIGH-FREQUENCY OSCILLATORY VENTILATION

General Characteristics
NHFOV is based on the application of a continuous flow, generating a constant dis-
tending positive pressure with superimposed oscillations, delivered all over the spon-
taneous breathing cycle. NHFOV could be applied either in a restrictive (eg,:
respiratory distress syndrome [RDS]) or in a mixed (eg, evolving bronchopulmonary
dysplasia [BPD] or BPD plus acute-on-chronic respiratory failure) respiratory insuffi-
ciency. NHFOV has the same basic principles and peculiar physiology in both cases:
Fig. 1 shows similar flow, volume, and pressures tracings recorded from active
neonatal lung models of restrictive or mixed pattern ventilated with NHFOV.
Oscillations have constant frequency and may be seen as somehow similar to those
of bubble CPAP, which provides a positive pressure with oscillations although the
latter are smaller, irregular, and with inconstant amplitude.2 Furthermore, NHFOV
can produce much higher mean airway pressure (Paw) than bubble CPAP, as it is
generated by a ventilator rather than a simple water valve.
A first interesting characteristic of NHFOV is that it can be easily used for alveolar
recruitment increasing Paw, without the risk of gas trapping-induced CO2 retention,

A B

Fig. 1. NHFOV during spontaneous breathing in an active lung model of neonatal restrictive
(model A) and mixed (model B) respiratory failure. Blue, green, red, and orange lines repre-
sent flow, volume, airway pressure (measured at the lung), and inspiratory muscle pressure
(spontaneously generated by the patient), respectively. Data have been generated using a
bench model modified from adult setting consisting of a neonatal mannequin ventilated
through a nasal mask and whose trachea had been connected to an electronic active test
lung (ASL5000; Ingmar Medical, Pittsburgh, Pennsylvania, USA). A Sensormedics SM3100A
oscillator (Vyaire, San Diego, California, USA) was used. Data were filtered at 100 Hz and
measured at the lung simulator using a specific software (ICU Lab rel.2.3; KleisTEK Advanced
Electronic System, Bari, Italy). Model A mimics a preterm neonate with RDS (birth weight:
1.5 kg, resistances: 100 cmH2O/L/s, compliance: 0.5 mL/cmH2O/kg, respiratory rate: 40
breaths/min, Paw: 8 cmH2O, amplitude 30 cmH2O; frequency 9 Hz, IT 50%). Model B mimics
an infant with BPD and acute worsening of respiratory function (acute-on-chronic respira-
tory failure) (birth weight: 2 kg, resistances: 300 cmH2O/L/s, compliance: 0.4 mL/cmH2O/kg,
respiratory rate: 50 breaths/min, Paw: 10 cmH2O, amplitude 50 cmH2O; frequency 6 Hz, IT
50%); notice how the inspiratory effort is weaker in this example, as the patients are expe-
riencing relevant work of breathing. A single spontaneous breath is shown in both panels.
Paw, airway pressure measured at the lung; Pmus, negative spontaneously generated inspi-
ratory muscle pressure.
 Nasal High-Frequency Ventilation 763

as this is avoided by the superimposed oscillations. Alveolar recruitment in NHFOV

can be performed in a patient with restrictive respiratory failure in the same manner
as it is performed in endotracheal HFOV, following the well-known principles of the op-
timum lung volume strategy.3 Compared to endotracheal HFOV, however, NHFOV is
generally accompanied by relevant pressure leaks that should be considered.4
If the effects on oxygenation are quite well known, the mechanisms of gas exchange
during NHFOV are incompletely understood. They partially correspond to the ones occur-
ring in endotracheal HFOV5 albeit with some peculiarities. During NHFOV, a tidal volume
is spontaneously generated, but at the same time, a small oscillatory volume is provided
by the cyclic pressure oscillations and delivered all along the respiratory cycle.1 These os-
cillations may be variably transmitted along the respiratory tree and this adds to the
complexity of gas exchange, which is based on several physical phenomena.6
Oscillation transmission seems to be the most important variable influencing ventila-
tion, although both tidal and oscillatory volumes actually contribute to gas exchange.7,8
This dual contribution has been initially hypothesized in in vitro measurements, but
recent in vivo studies in preterm infants have provided consistent results.9 Bench
data have also initially shown that NHFOV is able to washout CO2 from the upper air-
ways’ dead space.10 Subsequently, CO2 clearance has been also demonstrated at
lung level in similar bench models.11 However, recent in vivo data demonstrated that
during binasal prongs-delivered NHFOV, oscillations are not only transmitted to the up-
per airways but also to the alveolar tissue, especially in the nongravity dependent and
right-sided lung regions; this effect was noticed at relatively low oscillatory amplitude
values.9 Nonetheless, these data may be significantly influenced by several factors.
For instance, oscillation transmission is more efficient through stiff structures,12 thus
the amount of oscillatory volume reaching the alveoli may be different between babies
with mainly restrictive and homogeneous (ie, RDS) and those with mixed and nonhomo-
geneous patterns (ie, evolving BPD). Furthermore, type, size, and material of interfaces
may significantly influence the oscillation transmission and the resulting volume delivery
(see below).1 These can also be influenced by alveolar recruitment which changes the
regional compliance.12 Patients’ position could also have an effect, as neonates are
often turned and this changes the non–gravity-dependent lung zones. In adults, this shift
has been associated with increased regional compliance and pulmonary perfusion and
subsequent effects on volume delivery and oxygenation.13,14 An useful tool to summa-
rize the interplay of factors influencing oscillation transmission is the oscillatory pressure
ratio, that is, the ratio between the oscillation amplitude set at the ventilator and that
actually measured at a given level (eg, at the interface or the pharynx).12

Non-invasive High-Frequency Oscillatory Ventilation and Patient-Ventilator

Interaction
Noninvasive ventilation is difficult to synchronize in neonates because of their high res-
piratory rate, low tidal volume, and irregular breathing pattern. NHFOV bypasses this
problem as all ventilations at supraphysiological frequencies, by definition, do not
require synchronization. Furthermore, NHFOV could provide benefits over conven-
tional noninvasive ventilations, because it does not induce phasic inspiratory glottal
constriction, or decrease inspiratory glottal dilatation in newborn lambs.15 This may
allow a more constant pressure/volume transmission to the distal airways.16 However,
animal data also showed suppression of respiratory drive when nasal mask-delivered
NHFOV was applied with a very low frequency (4 Hz).15 This effect has been confirmed
in various models and is not mediated by hypocarbia, but rather by an increased vagal
pulmonary stretch receptor or thoracic wall afferent activity.17,18 However, such low
frequencies are not used in neonatology and an increased respiratory drive with
764 De Luca & Centorrino

consequent diaphragmatic activation has also been observed in animals: this de-
pends on the ventilation parameters and is mediated by pulmonary rapidly adapting
receptors.19,20 Conversely, in adults with central sleep apnea, nasal mask-delivered
high-frequency oscillations stimulate respiratory effort in adult patients.21 In neonates,
other mechanisms also influence the spontaneous respiratory drive, such as inflam-
mation, pain, or discomfort and the choice of NHFOV interface may play a relevant
role (see below).22 Finally, as patients are spontaneously breathing during NHFOV,
an increment in their work of breathing (WOB) could be observed, although this is
lower in smaller patients.23 WOB increment depends on many factors such as lung
compliance and resistances, patients’ size, ventilator type (see below), and parame-
ters. Regarding these latter, lower frequencies seem to be associated with lower addi-
tional WOB24: this should be considered for long-lasting NHFOV, but also balanced
with the need to deliver adequate ventilation. Therefore, the interactions between
high-frequency oscillations and spontaneous respiratory drive are complex and oppo-
site effects might be observed in different patients or in different moments: tailoring
ventilation with close patient monitoring is crucial.

Effect of Different Interfaces for Non-invasive High-Frequency Oscillatory

Ventilation
NHFOV can be provided using different interfaces and each has its own mechanical
characteristics and multiples effects on NHFOV physiology. Moreover, interfaces may
significantly affect patients’ comfort and the combination of these mechanisms can
considerably change the effect of NHFOV in terms of oxygenation and gas exchange.
The first clinical experiences on NHFOV used single, long, high-resistive nasopharyn-
geal tubes.25 As demonstrated for CPAP, these interfaces were unsuitable and should
not be used; in fact, they are associated with large leaks occurring through the contra-
lateral nostril and with a relevant resistive load increasing the patients’ WOB.26 As short
binasal prongs should be preferred over nasopharyngeal tubes,27 and nasal masks
seem even better than short prongs,28 we have investigated them in dedicated NHFOV
bench studies finding efficient oscillation transmission and volume delivery.4,29,30
Finally, the use of prongs occluding only a small portion of the nostril cross-sectional
area and connected via a long and resistive tubing (RAMCannula) is currently spreading.
These interfaces are particularly comfortable; however, they provide high resistance,
which increases the patients’ WOB,31 and significant leaks when used to deliver
CPAP32,33 or conventional noninvasive ventilation.34 Despite these negative mechanical
characteristics, a case series described the use of RAMCannula-delivered NHFOV in 3
neonates ventilated with relatively low Paw.35
Our knowledge on the different interfaces for NHFOV and their effects on physiology
can be resumed as follows:
1. The diameter of binasal prongs is important to guarantee an efficient ventilation (ie,
the larger the probe, the greater the volume delivery); for a given amplitude and fre-
quency, increasing the inspiratory time from 33% to 50%, allows a greater volume
delivery, but increasing the amplitude beyond 50 to 60 cmH2O does not signifi-
cantly increase ventilation.29,30 Thus, when binasal prong-delivered NHFOV is pro-
vided with maximal parameters (ie, with amplitude of approximately 50–60 cmH2O,
a frequency of 8–10 Hz, and 50% inspiratory time), a suitable oscillatory volume
might be provided to neonates up to 1 to 1.5 kg29,30 (if we consider 1–2 mL/kg
as an ideal target alike in invasive HFOV36).
2. Nasal masks can efficaciously deliver NHFOV but provide lower oscillation trans-
mission, due to the dampening occurring on the skin tissue and the mask soft
 Nasal High-Frequency Ventilation 765

material.4 This seems consistent with what happens during full-face mask-deliv-
ered NHFOV in infants beyond neonatal age.7 More aggressive parameters (partic-
ularly lower frequency) may be needed to deliver the same oscillatory volume
provided through binasal prongs.4 Nasal masks are associated with lower pressure
leaks compared to nasal prongs33 and these leaks (z30%-35%) seem similar dur-
ing NHFOV and other types of noninvasive support.4 Bench data have demon-
strated that moderate leakage may increase CO2 clearance during NHFOV,
probably facilitating the washout from the upper airways dead space and reducing
gas trapping; thus, moderate leaks may be allowed, on a case-by-case
evaluation.37,38
3. RAMCannula should not be used to deliver NHFOV, if it is applied in cases of severe
respiratory failure (for instance, when intubation is pending) or for long periods or
when the added resistance may have negative consequences (for instance, in
extremely low birth weight neonates).
These mechanical data do not advocate for universal use of a single interface. In
fact, patients’ comfort, ventilatory parameters, integrity of skin, and also nonrespira-
tory factors should be considered, as well. Moreover, the severity of respiratory failure
may vary from one patient to the other and between different moments during the clin-
ical course; thus, sometimes less aggressive parameters may be sufficient to
compensate respiratory failure. Mechanical characteristics of interfaces, patients’
comfort, and severity have a complex interplay on NHFOV physiology; therefore,
the choice of NHFOV interface should be based on all these aspects and aim to
find the best compromise between ventilation efficiency and patients’ comfort.1 This
latter remains to be evaluated in specifically dedicated studies and, therefore, inter-
faces should be tailored on a case-by-case basis evaluation and interchanged to
reduce the risk of skin lesions and according to patients’ needs.39 Table 1 resumes
the factors influencing gas exchange during NHFOV.

Different Devices Producing Non-invasive High-Frequency Oscillatory Ventilation

NHFOV may be applied with any ventilator able to provide the HFOV mode: several
technologies are available.1 From a formal point of view, an actual “oscillatory”

Table 1
Factors influencing gas exchange during NHFOV

Variable Effect on Gas Exchange

Oscillation amplitude [
Inspiratory time [
Frequency Y
Leaks Ya
Interface Variable
Mean airway pressure Variable
Gravity (patient positioning) Variable

Effect of interface is variable because different interfaces may facilitate or reduce oscillation trans-
mission through an improved patients’ comfort and/or changing pressure leaks and/or oscillation
dampening.
a
Leaks are generally reducing gas exchange through decreased oscillation transmission, but
moderate leaks have been demonstrated to increase CO2 clearance under certain experimental
conditions.37,38 The effects of Paw or gravity are variable because increasing constant distending
pressure, or positioning the infant prone or supine may change regional compliance and affect
oscillation transmission.
766 De Luca & Centorrino

ventilation should have an active expiratory phase produced by a vibrating piston or

membrane over a continuous gas flow or by an electronically controlled, cyclic flow
reversal. These 2 ways to produce active oscillations have not been compared with
respect to the application of NHFOV. Other technologies to produce HFOV (although
without an active expiratory phase) are represented by the flow interruption because
of the cyclic opening-closure of one or more pressure valves. Some neonatal ventila-
tors are technically able to provide NHFOV using this technology, but their perfor-
mance to provide invasive HFOV can be suboptimal at extreme settings or for late
preterm/term neonates.40,41 As NHFOV is usually proposed for preterm infants, and,
as bench studies have demonstrated an adequate ventilation for neonates up to 1
to 1.5 kg,29,30 this is not likely to represent a significant problem. Another ventilator
produces oscillations based on the inertia of gas in the circuit when the pressure at
airway opening is rapidly changing: this technology is combined with fast responding
inspiratory valves and high-flow capability but it has not been formally tested for
NHFOV yet. There are also hybrid systems based on positive pressure generated
by high-flow nasal cannula with superimposed high-frequency oscillations provided
by a solenoid valve: they have been shown to provide efficacious CO2 clearance in
bench models.42,43 Finally, a new technology based on electronically controlled
blower and valve has been specifically proposed for NHFOV.44 So far, hybrid high
flow or blower and valve technologies have not been incorporated in any commercially
available ventilator.
The active oscillation is considered important for the CO2 clearance; however, the
wide experience accumulated on invasive HFOV seems to indicate that this is not
actually affecting clinical outcomes.45 Nonetheless, at least in some patients, the
kind of NHFOV-producing device may have an impact on its short-term effi-
ciency.40,41 This problem may be less relevant in NHFOV, as this is supposed to
be used in neonates below 1 to 1.5 kg. It is also important to note that, as patients
are spontaneously breathing during NHFOV, a certain WOB may theoretically be
superimposed by NHFOV application. This WOB increment is relatively low for pre-
term infants, but seems significantly different between ventilators based on the
above-described technologies, with tendency to a lower WOB for ventilators with
an active expiration.24

Humidification During Non-invasive High-Frequency Oscillatory Ventilation

Heating and humidification during noninvasive respiratory support seem to improve
comfort in adults, although we do not have specific neonatal data.46 An European
survey identified viscous secretions and consequent upper airway obstructions as
specific side effects of NHFOV47 and this seems logical as NHFOV is usually applied
as rescue, when other noninvasive respiratory techniques have failed. The American
Association for Respiratory Care suggests to use active humidification during nonin-
vasive ventilation, although there are still open questions about the type of active hu-
midifier to prefer.46 Ullrich and colleagues have studied humidification during
NHFOV and found that aggressive NHFOV settings (ie, low frequency, high ampli-
tude, and IT) significantly reduced oropharyngeal gas conditioning.48 This is consis-
tent with data on CO2 pharyngeal washout10; thus, it seems reasonable to think that
aggressive NHFOV removes water through physical mechanisms similar to those of
HFOV gas exchange.5 The presence of leakage might also contribute, as gas parti-
cles can be mixed by thermodiffusion, and heat may be lost by thermal conduction or
with the entry of cool dry gas particles from the room air. The clinical relevance of
these phenomena is unknown, but probably limited if NHFOV is not used for a
long time.
 Nasal High-Frequency Ventilation 767

BIOLOGY OF NON-INVASIVE HIGH-FREQUENCY OSCILLATORY VENTILATION

Reddy and colleagues showed that superimposing oscillations over tidal volume ex-
cursions in a surfactant bubble lowers surface tension significantly more than tidal vol-
ume excursion alone.49 Minimum surface tension decreased with increasing
frequencies and reached a value of z7 mN/m at extreme frequencies (70–80 Hz),
not attainable in clinical care. Conversely, minimum surface tension of 15 to 30 mN/
m was measured with frequencies usually applied when using NHFOV. Similar values
have been measured in neonates and infants with neonatal or pediatric ARDS.50,51
Invasive HFOV improves lung mechanics and histology in surfactant-depleted rab-
bits.52 Consistent findings, as well as larger surfactant aggregates, have been re-
ported in animal models mimicking different types of lung injury.53,54 These data
allow to hypothesize that NHFOV could improve surfactant function, although this
only remains a working hypothesis.

EVIDENCE-BASED REVIEW OF CLINICAL DATA ON NON-INVASIVE HIGH-FREQUENCY

OSCILLATORY VENTILATION
Uncontrolled Studies
In 2016, we have analyzed the clinical data on NHFOV available at that time and these
were mainly represented by uncontrolled small case series, showing globally prom-
ising results.1 Because of the neonatal experience on HFOV and wide availability of
this ventilatory mode, NHFOV spread in the last 5 years led to the publication of other
similar studies. These latest uncontrolled studies were consistent with the earlier data,
reporting that: (1) NHFOV may reduce the extubation failure or the need of invasive
ventilation in infants with pending intubation; (2) NHFOV may improve gas exchange;
(3) NHFOV may reduce the number of apneic spells; and (4) NHFOV did not cause any
severe adverse event.55–58 The absence of these had also been suggested by a Euro-
pean survey of physicians using NHFOV.47

Randomized Controlled Trials

After these studies, randomized controlled trials finally started to be published. We
performed a systematic review of these trials. A literature search was performed on
PubMed (on November 7, 2020), using “nasal” or “noninvasive high-frequency oscil-
latory ventilation” or “NHFOV”, as words or MeSH terms, without year or language re-
strictions. We also hand-searched references cited in the studies identified through
the initial search, review articles, and the authors’ personal archives. We excluded
“gray” literature, unpublished or non–peer-reviewed reports. Non-English manu-
scripts were translated using Google translator. We used a data extraction sheet
based on the Cochrane Review Group template adapted from our previous work.59
Data from included trials were extracted and cross-verified independently by the 2 au-
thors. We analyzed data applying the Sidik-Jonkman method60 with random-effects
models. Consistency was evaluated using the I2 statistics and c2 test for heterogene-
ity. Meta-regressions were performed adjusting for gestational age and prenatal
steroids as confounders. We inserted one covariate in each model to reduce false-
positive results.59 Analyses were performed with Open-MetaAnalyst 10.1.61

Meta-Analysis of Randomized Controlled Trials

To date, there are 11 trials comparing NHFOV against single-level or biphasic CPAP,
either with parallel or crossover design, mainly having short-term events or gas ex-
change as primary outcomes.62–72 Table 2 shows essential trials’ characteristics.
All but 2 trials63,68 recruited relatively small populations and 3 enrolled extremely
 768
Table 2
Randomized clinical trials comparing NHFOV versus single-level or bilevel CPAP

De Luca & Centorrino

No. of Prenatal Maximum Paw
Author/Year Patients GA (wk) Steroids (%) Primary Outcomes Secondary Outcomes and Amplitude Enrolled Population
Bottino et al,62 30 26.4 N.A. PaCO2 N.A. 8/10a Stable preterm neonates
2018
Chen et al,63 206 32.6 89.8 Need for IMV Complications of 16/40 Preterm neonates with RDS
2020 and PaCO2 prematurity or NARDS as postextubation
support
Iranpour et al,64 68 33 26.4 Duration of CPAP Need for reintubation; 8/20a Preterm neonates with RDS
2019 or NHFOV Complications of as primary support
prematurity
Klotz et al,65 26 26.7 100 PaCO2 Apneas, bradycardia, 8/N.A.a Stable preterm neonates
2018 and safety data
Lou & Zhang,66 65 32.4 38.4 Need for IMV Oxygenation and PaCO2, N.A. Preterm neonates with RDS
2017 complications of as postextubation support
prematurity
Lou et al,67 2018 65 33.8 35.3 Oxygenation IMV duration, apneas, 12/35 Preterm neonates with RDS
and PaCO2 complications of as postextubation support
prematurity
Malakian et al,68 124 31.1 53.9 Need for IMV IMV duration, 8/7a Preterm neonates with RDS
2018 complications of as primary support
prematurity
Mukerji et al,69 39 28.8 80.9 Feasibility IMV duration, PaCO2, 10/N.A.a Preterm neonates with RDS
2017 complications of as postextubation support
prematurity
Rüegger et al,70 40 26.5 90 Bradycardia and/ Vital parameters and 7/40a Stable extremely preterm
2018 or desaturation safety data neonates
Zhu et al,71 2017 76 31.8 36.7 Need for IMV Complications of 6/N.A.a Preterm neonates with RDS
prematurity as postextubation support
Zhu et al,72 2017 38 31.8 36.6 Oxygenation Complications of 10/10a Preterm neonates with RDS
and PaCO2 prematurity as postextubation support

Gestational age and prenatal steroids were considered as the weighted mean of the 2 trial arms. Three studies had a crossover design62,65,70; the remaining were
parallel trials. Values have been rounded to the closest decimal.
Abbreviations: GA, gestational age; IMV, invasive mechanical ventilation; N.A., not available; NARDS, neonatal acute respiratory distress syndrome; PaCO2, car-
bon dioxide levels; Paw, mean airway pressure; RDS, respiratory distress syndrome.
a
Asterisks indicate that patients in trial arms have equivalent Paw.
 Nasal High-Frequency Ventilation 769

preterm neonates.62,65,70 Most of the primary outcomes were represented by short-

term need for intubation and invasive ventilation (IMV)63,66,68,71 or PaCO2/
oxygenation.62,63,65,67,72 Although authors should be commended for the efforts, there
are important problems behind these outcomes’ choice:
1. The need for IMV can be a sensible outcome; however, some studies investigated
NHFOV in preterm neonates with RDS as primary respiratory support (ie, before
surfactant administration, if any)64,68 and others as secondary support after extu-
bation or surfactant administration.63,66,67,69,71,72 This lack of homogeneity pre-
vents to draw any conclusion: although it seems logical that a higher Paw
would reduce the risk of extubation failure as demonstrated for other noninvasive
respiratory support techniques,73,74 we need larger trials focused on the postex-
tubation phase, comparing NHFOV with other noninvasive techniques. More and
above this, we need to actually use higher Paw during NHFOV; in the majority of
trials, Paw was equivalent in the 2 arms,62,64,65,68–72 and this would prevent
NHFOV to provide an actual alveolar recruitmenet. Furthermore, it is unclear
what might be the advantage of NHFOV in the early phase of RDS. In fact, it is
known that CPAP works very well for most of the patients in this phase75 and,
when CPAP fails, that is usually for worsening oxygenation. Oxygenation impair-
ment in a purely restrictive and homogeneous condition (such as RDS) is easily
overcome by alveolar recruitment. However, as trials investigating NHFOV as a
primary mode always used an equivalent Paw in the 2 arms,64,68 no alveolar
recruitment was provided and it was logical to observe no difference. Moreover,
if alveolar recruitment through NHFOV would be applied in this phase, this might
delay surfactant replacement, reducing its efficacy, which is optimal only within
the first 3 hours of life.76,77
2. HFOV is known to be very powerful in washing out CO2. Because NHFOV has some
peculiar physiologic characteristics, it was interesting to evaluate its carbon diox-
ide clearance capacity, although it would have been unlikely to see NHFOV failing in
this regard. In fact, face mask-delivered NHFOV has been found to effectively
washout CO2 also in a small crossover trial enrolling adults.78 However, CO2 clear-
ance has been tested in trials enrolling stable preterm neonates or anyway with
relatively low PaCO2 levels and no respiratory acidosis.62,63,65–67,69,72 This choice
has led to less meaningful and possibly biased results, since, in the daily NICU
care, no one would shift a patient from CPAP to NHFOV if there is no hypercarbia.
Moreover, having CO2 clearance as outcome also presents a problem similar to the
aforementioned issue about Paw and oxygenation. In fact, some trials used a flow
interruption device, generating very low amplitudes, which are unlikely to be trans-
mitted downstream62,68,71,72; the generation of a very little oscillatory volume and
its actual contribution to gas exchange is doubtful.
Two large well-designed physiology-driven multicenter trials are currently ongoing.1
These trials aim to verify if NHFOV provides any benefit, compared to CPAP or nonin-
vasive positive pressure ventilation, either as primary respiratory support or in the
postextubation phase for preterm neonates with RDS.79,80
The trials published so far had a panoply of secondary outcomes, among which,
there are some difficult to improve, but also several vital parameters and safety
data. NHFOV reduced the number of desaturations and bradycardia in one trial,70
although there was no difference in any safety data in the other trials.62–69,71,72
Therefore, we can reasonably state that NHFOV is essentially safe or can even be
beneficial in reducing bradycardia, desaturations, and/or apneas, at least in some
patients. However, due to the complex effects of NHFOV on the spontaneous
770 De Luca & Centorrino

breathing (see above), these results cannot be generalized as they can change ac-
cording to the patient’s clinical condition, cointerventions, NHFOV interfaces, and
parameters.
We present here the meta-analysis of trials focusing on the 2 more commonly stud-
ied outcomes: (1) need for intubation and mechanical ventilation; (2) PaCO2 levels after
NHFOV application (Fig. 2). NHFOV significantly reduces the risk or intubation and
need of IMV (odds ratio: 0.29; 95% confidence interval: 0.2–0.4; P<.001) compared
to single-level or biphasic CPAP. These results are confirmed if we only analyze the
trials using NHFOV as postextubation support (odds ratio: 0.3; 95% confidence inter-
val: 0.18–0.5; P<.001). NHFOV also tends to reduce CO2 compared to single-level or
biphasic CPAP (mean difference: 4.6 mm Hg; 95% confidence interval: 9.3 to 0.08;
P 5 .05); significant heterogeneity is seen for this outcome and this may be related to
the different times and techniques to measure PaCO2 and to the different ventilatory
strategies described earlier.

Fig. 2. Meta-analysis of NHFOV trials: Forrest plots for the more commonly studied out-
comes. Panels A and B show the need for intubation and mechanical ventilation (681 pa-
tients) and PaCO2 levels after NHFOV application (662 patients), respectively. NHFOV and
the single-level or biphasic CPAP are considered as treatment (Trt) and control (Ctrl) arm,
respectively; events per arm and odds ratio or mean difference (95% confidence interval)
are reported in panels A and B, respectively. Square size is proportional to trial weight. Dia-
mond width indicates the 95% confidence interval of the final effect size. The need for intu-
bation and invasive ventilation was considered at any timepoint after intervention (some
trials defined this outcome within 72 hours, others within a 7-day time-window). PaCO2
levels were considered at any timepoint after intervention (trials defined this outcome by
measuring PaCO2 at various times after the intervention). Trials weight for the outcome intu-
bation were as follows: Chen: 36.138%, Iranpour: 1.636%, Lou: 10.087%, Lou-2: 12.530%,
Malakian: 9.406%, Mukerji: 8.153%, Zhu: 14.882%, Zhu-2: 7.167%. Trials weight for the
outcome PaCO2 levels were as follows: Bottino: 10.820%, Chen: 11.968%, Iranpour:
11.454%, Klotz: 8.461%, Lou: 11.471%, Lou-2: 11.762%, Malakian: 11.740%, Rüegger:
10.651%, Zhu-2: 11.672%. 95% CI, 95% confidence interval; Ctrl, control arm (ie, single-
level or biphasic CPAP); PaCO2, carbon dioxide levels; Trt, treatment arm (ie, NHFOV).
 Nasal High-Frequency Ventilation 771

We further studied the effect of possible confounders: for the outcome intubation,
neither gestational age (coefficient: 0.104 [95% confidence interval: 0.2; 0.4];
P 5 .527), nor prenatal steroids (coefficient: 0.007 [95% confidence interval:
0.02; 0.007]; P 5 .327), were associated with the effect size; same results were found
for PaCO2 levels, regarding gestational age (coefficient: 1.1 [95% confidence interval:
2.7; 0.4]; P 5 .137) and prenatal steroids (coefficient: 0.07 [95% confidence interval:
0.1; 0.2]; P 5 .424).
These results, and particularly those issued by subgroup analyses and metaregres-
sions, should be cautiously seen also in light of the above-described problems in trial
design and outcome choice. The NHFOV trials published so far have been affected
by significant intrinsic biases and these have been reported in comment letters.81,82
We do not analyse here all the biases, as this would be beyond our scope. Nonetheless,
future trials shall investigate NHFOV with a physiology-driven management, in homoge-
neous populations, with a clearly defined lung mechanics and restricted,

Fig. 3. Proposal for a tailored protocol to apply NHFOV in extremely preterm infants with
developing BPD (ie, chronic pulmonary insufficiency of prematurity).86 This is the protocol
in use at Paris Saclay University Hospitals NICU. Different types of noninvasive respiratory
techniques are used after the first week of life in extremely preterm infants if they experi-
ence a worsening of their respiratory function. NHFOV is integrated into the strategy with
the other techniques based on a physiology-driven approach. Hypoxic respiratory failure
(blue lines) is defined with an increased work of breathing (Silverman score >4) without hy-
percarbia and with FiO2 greater than 0.4 to achieve peripheral saturation between 90% and
95% and is treated with synchronized conventional noninvasive ventilations (NIV-NAVA or
sNIPPV); NHFOV is used if these fail. Hypercapnic respiratory failure (red lines) is defined
with hypercarbia (CO2 >65 mm Hg) and acidosis (pH<7.20), irrespective of the oxygenation
deficit, and treated with NHFOV as first line. NHFOV is managed by applying alveolar
recruitment maneuvers and with a close multimodal monitoring. Definition criteria should
be fulfilled for at least 4 to 6 h before instigating NIV-NAVA, sNIPPV, or NHFOV and patients
are monitored over time with several noninvasive techniques (see text for more details). As
the patient is improving, the respiratory support can be de-escalated. Full and hatched lines
indicate deterioration and improvement of respiratory conditions, respectively. aThe choice
between NIV-NAVA and sNIPPV depends on the availability of ventilators. EDIN, “Echelle et
Inconfort du Nouveau-né” score; FiO2, inspired oxygen fraction; LUS, lung ultrasound score;
NIV-NAVA, noninvasive ventilation with neurally adjusted ventilator assist; PI, perfusion in-
dex; SatO2, peripheral hemoglobin saturation; sNIPPV, synchronized noninvasive positive
pressure ventilation; WOB, work of breathing.
772 De Luca & Centorrino

physiopathologically sound objectives. NHFOV shall be compared to a well-defined

“control” technique and both shall be applied with a strict protocol. In other words,
future trials should have an explanatory design and tend to recruit only from well-
experienced sites. More pragmatic inclusive approaches are unsuitable because they
seek a “real world” answer for a more widely used and well-known intervention.83 On
the contrary, NHFOV is a relatively new technique and, by mixing different populations
or leaving ventilatory management too free, we risk to lose important information.84
To date, according to the available clinical data and the physiology background,
NHFOV can be considered as an additional technique for infants with severe respira-
tory failure. It may be suitable in preterm patients with pending reintubation or in those
with evolving BPD, where one may want to spare oxygen exposure and invasive venti-
lation as much as possible. In these cases, NHFOV can be used, if there is enough
expertise, after careful evaluation on a case-by-case scenario and with accurate pa-
tients’ monitoring and physiology-based management.

PERSONAL EXPERIENCE AND PROTOCOL TO MANAGE NON-INVASIVE HIGH-

FREQUENCY OSCILLATORY VENTILATION

As NHFOV represents another “brick in the wall” of the noninvasive respiratory support,85
we have been using it for extremely preterm infants with evolving BPD to reduce invasive
ventilation as much as possible. These patients are comprised under the definition of
chronic pulmonary insufficiency of prematurity (CPIP), recently issued by the International
Neonatal Consortium, which spans as a continuum from the end of the first week of life to
36 weeks’ postconceptional age.86 During this period, in our experience, some extremely
preterm infants show a worsening of their respiratory function around 14 days of post-
natal age and this can be easily visualized with semiquantitative lung ultrasound.87
Our NHFOV protocol follows a physiology-based approach, with alveolar recruit-
ment maneuvers alike in endotracheal HFOV3 and close multimodal monitoring, based
on semiquantitative lung ultrasound,88 transcutaneous blood gas measurements, pe-
ripheral saturation, and perfusion index.89 Lung ultrasound is used to assess lung
aeration and guide the alveolar recruitment in real-time, as described in critically ill
adults.90 Nurses are specifically trained to care for these infants who are considered
at high risk: nonpharmacological sedation is widely given, hydrocolloid gels are
used, and interfaces are swapped to change the pressure points and reduce the
risk of skin injuries. COMFORT91 and/or EDIN92 scores are serially used to evaluate

Table 3
Suggested parameter boundaries of NHFOV for extremely preterm infants with developing
BPD (ie, chronic pulmonary insufficiency of prematurity)

Minimum Maximum
Mean airway pressure (cmH2O) 10 18
Amplitude (cmH2O) 30 55
Frequency (Hz) 8 12
1
These suggestions have been modified from those previously proposed, based on accumulated
clinical experience. Inspiratory time should be fixed at 50%. Parameters may require serial adjust-
ments according to patients’ monitoring. Paw should be titrated on oxygenation and/or ultra-
sound assessed lung aeration. Oscillation amplitude and frequency should be titrated according
to transcutaneous CO2 levels. Interfaces might also impact the NHFOV performance and patients’
comfort needing to be changed and requiring parameters adjustments.
Data from Steinhorn R, Davis JM, Göpel W, et al.Chronic Pulmonary Insufficiency of Prematurity:
Developing Optimal Endpoints for Drug Development.J Pediatr 2017;191:15-21.e1.
Nasal High-Frequency Ventilation 773
774 De Luca & Centorrino

patients’ comfort. Our proposal also integrates different respiratory techniques and
respiratory support in personalized fashion.
As shown in Fig. 3, the respiratory management is initially based on gas exchange
traits. Owing to its capability to washout CO2, NHFOV is used as first intention in
extremely preterm infants experiencing hypercapnic respiratory failure. Conventional
noninvasive respiratory support is initially used in infants with hypoxemic respiratory
failure and NHFOV is regarded as rescue intervention in case of failure; conventional
noninvasive ventilation is synchronized, either using neurally adjusted ventilator
assist93 or flow/pressure-sensors to increase its efficacy and optimize patient-
ventilator interaction (more details in the figure legend). Point-of-care echocardiogra-
phy is also performed according to international guidelines94: when there are signs of
pulmonary hypertension and this significantly influences hypoxia, nebulized iloprost is
started,95 using modern vibrating-mesh nebulizers inserted on the inspiratory limb.96
Thus, intubation and inhaled nitric oxide are only considered as last resource. When
the monitoring shows consistent signs of improvement, the respiratory support is
de-escalated and can go back to CPAP, which is usually weaned between 33 and
340 weeks postconceptional age. An illustrative case of a patient managed with this
respiratory strategy has been described in our previous review on NHFOV.1
This is obviously just a proposal for a respiratory management protocol integrating
NHFOV for neonates with evolving BPD. Table 3 shows suggested boundaries for
NHFOV in our strategy. Other possible strategies exist and, for example, NHFOV
has been proposed also as first-line technique in neonates with RDS.1 However, the
use of NHFOV later in life for neonates with CPIP seems to us more reasonable and
well-grounded. It is actually difficult to design randomized controlled trials for these
patients, but in absence of these studies, the respiratory care should be tailored to
the patients’ characteristics as much as possible.

EXPERIENCES WITH NON-INVASIVE HIGH-FREQUENCY PERCUSSIVE VENTILATION

High-frequency percussive ventilation is a pneumatic, pressure-limited, time-cycled,

high-frequency ventilation providing subphysiological volumes generated by Venturi’s
effect through a sliding device (called Phasitron) powered by high-flow compressed
gas inlet. Thus, the high-frequency volume delivery is provided as gas “percussions”
into the airways. Between 90 and 650 percussions per minute can be provided. These
percussions are superimposed to a pressure-limited, conventional respiratory support
with physiologic rate and volume: conventional frequency, IT, and positive end-
expiratory pressure need to be set as usual. An end-expiratory pressure for the gas
percussions must be set, while the peak pressure is decided through the value of pul-
satile gas flow (the greater the flow, the higher is the peak pressure reached all along
the conventional respiratory cycle). A typical pressure waveform during this modality is
shown in Fig. 4: conventional breaths are drawn with superimposed percussions. As
the percussions are generated through the Venturi effect, the ventilator circuit has an

=
Fig. 4. Illustrative time-pressure waveform during neonatal NHFPV. Conventional breaths
are drawn with superimposed percussions. Ventilatory parameters to be decided by clini-
cians are indicated in the figure. Pressure rates for conventional breaths and flow rates
for the pulsatile percussions must also be set to decide the maximum delivered pressures.
ET, conventional breath expiratory time; i/e ratio, inspiratory/expiratory ratio for the gas
percussions; IT, conventional breath inspiratory time; PEEP, positive end-expiratory pressure;
PIP, peak inspiratory pressure.
 Nasal High-Frequency Ventilation 775

open expiratory limb and the patient may spontaneously breathe without any added
WOB. Only one ventilator can provide this modality, which can be delivered both
endotracheally or as NHFPV. This modality has the physical capability to improve se-
cretions clearance and to move secretions toward upper airways. Because of these
characteristics, this modality has been mainly used for aspiration-induced lung injuries
and acute RDS, both in adults and children.97
In neonatology, NHFPV has been investigated in a randomized controlled trial to treat
transient tachypnoea of the neonate (TTN): NHFPV was superior to CPAP in improving
oxygenation and reducing the duration of TTN.98 In a second work, the same authors
showed that NHFPV is safe in terms of cerebral oxygenation in neonates with TTN or
moderate RDS.99 As TTN is due to a lack of lung fluid reabsorption, these results
seem physiopathologically plausible as NHFPV may have facilitated the lung fluid clear-
ance. Given its physical characteristics, NHFPV might also be theoretically useful in
meconium aspiration, alike for other inhalation syndromes in older patients. Interest-
ingly, endotracheal high-frequency percussive ventilation compared to HFOV resulted
in a better oxygenation in the animal model of meconium aspiration,100 while the two
techniques resulted equivalent in a model of lung injury caused by depleting lung la-
vages.101 Furthermore, 2 other animal studies compared the long-term effect of NHFPV
and invasive ventilation in preterm lambs mimicking infants with CPIP (ie, evolving BPD).
The animals ventilated with NHFPV for 3 weeks showed improved alveolarization with
increased surfactant protein-B expression and better oxygenation.102,103 These findings
may be at least partially explained by an enhanced PTHrP-PPARg-mediated epithelial/
mesenchymal signaling of alveolarization.102 These results allow to hypothesize that
long-term respiratory support with NHFPV, or a strategy integrating different noninva-
sive nonconventional respiratory supports, might be useful to improve long-term respi-
ratory outcomes in preterm infants. In conclusion, the use of NHFPV for TTN seems
interesting, but given its complexity, the mildness of TTN and the effectiveness of
CPAP, it is unclear if NHFPV may be really useful.

ACKNOWLEDGMENTS

Authors are grateful to Alejandro Alonso for the artwork and to Prof. Giorgio Conti, for
the modeling of NHFOV.

CONFLICTS OF INTEREST

Prof. D. De Luca has received research grants, technical assistance, and travel grants
from Vyaire Inc. He also served as a lecturer for Getinge Inc. Dr R. Centorrino received
a travel grant from Vyaire Inc. These companies produce ventilators that are able to
provide noninvasive high-frequency ventilations but had no role in the conception,
writing, or decision to submit this article.

REFERENCES

1. De Luca D, Dell’Orto V. Non-invasive high-frequency oscillatory ventilation in ne-

onates: review of physiology, biology and clinical data. Arch Dis Child Fetal
Neonatal Ed 2016;101:F565–70.
2. Pillow JJ, Hillman N, Moss TJM, et al. Bubble continuous positive airway pres-
sure enhances lung volume and gas exchange in preterm lambs. Am J Respir
Crit Care Med 2007;176:63–9.
776 De Luca & Centorrino

3. De Jaegere A, van Veenendaal MB, Michiels A, et al. Lung recruitment using

oxygenation during open lung high-frequency ventilation in preterm infants.
Am J Respir Crit Care Med 2006;174:639–45.
4. Centorrino R, Dell’Orto V, Gitto E, et al. Mechanics of nasal mask-delivered
HFOV in neonates: a physiologic study. Pediatr Pulmonol 2019;54:1304–10.
5. Pillow JJ. High-frequency oscillatory ventilation: mechanisms of gas exchange
and lung mechanics. Crit Care Med 2005;33(3 Suppl):S135–41.
6. Boynton BR, Hammond MD, Fredberg JJ, et al. Gas exchange in healthy rabbits
during high-frequency oscillatory ventilation. J Appl Physiol 1989;66:1343–51.
7. De Luca D, Costa R, Visconti F, et al. Oscillation transmission and volume deliv-
ery during face mask-delivered HFOV in infants: bench and in vivo study: nonin-
vasive HFOV Feasibility in Infants. Pediatr Pulmonol 2016;51:705–12.
8. De Luca D, Costa R, Spinazzola G, et al. Oscillation transmission and ventilation
during face mask-delivered high frequency oscillatory ventilation in infants: a
bench study with active lung simulator. Arch Dis Child 2012;97(Suppl 2):
A117–8.
9. Gaertner VD, Waldmann AD, Davis PG, et al. Transmission of oscillatory vol-
umes into the preterm lung during noninvasive high-frequency ventilation. Am
J Respir Crit Care Med 2020. https://doi.org/10.1164/rccm.202007-2701OC.
10. Mukerji A, Finelli M, Belik J. Nasal high-frequency oscillation for lung carbon di-
oxide clearance in the newborn. Neonatology 2013;103:161–5.
11. Sivieri EM, Eichenwald EC, Rub DM, et al. An in-line high frequency flow inter-
rupter applied to nasal CPAP: improved carbon dioxide clearance in a prema-
ture infant lung model. Pediatr Pulmonol 2019;54:1974–81.
12. van Genderingen HR, Versprille A, Leenhoven T, et al. Reduction of oscillatory
pressure along the endotracheal tube is indicative for maximal respiratory
compliance during high-frequency oscillatory ventilation: a mathematical model
study. Pediatr Pulmonol 2001;31:458–63.
13. Prisk GK, Yamada K, Henderson AC, et al. Pulmonary perfusion in the prone and
supine postures in the normal human lung. J Appl Physiol 2007;103:883–94.
14. Nyrén S, Radell P, Lindahl SGE, et al. Lung ventilation and perfusion in prone
and supine postures with reference to anesthetized and mechanically ventilated
healthy volunteers. Anesthesiology 2010;112:682–7.
15. Hadj-Ahmed MA, Samson N, Nadeau C, et al. Laryngeal muscle activity during
nasal high-frequency oscillatory ventilation in nonsedated newborn lambs.
Neonatology 2015;107:199–205.
16. Owen LS, Morley CJ, Dawson JA, et al. Effects of non-synchronised nasal inter-
mittent positive pressure ventilation on spontaneous breathing in preterm in-
fants. Arch Dis Child Fetal Neonatal Ed 2011;96:F422–8.
17. Kohl J, Freund U, Koller EA. Reflex apnea induced by high-frequency oscillatory
ventilation in rabbits. Respir Physiol 1991;84:209–22.
18. England SJ, Onayemi A, Bryan AC. Neuromuscular blockade enhances phrenic
nerve activity during high-frequency ventilation. J Appl Physiol 1984;56:31–4.
19. Kohl J, Scholz U, Glowicki K, et al. Discharge of pulmonary rapidly adapting
stretch receptors during HFO ventilation. Respir Physiol 1992;90:115–24.
20. Kohl J, Koller EA. Blockade of pulmonary stretch receptors reinforces diaphrag-
matic activity during high-frequency oscillatory ventilation. Pflugers Arch 1988;
411:42–6.
21. Henke KG, Sullivan CE. Effects of high-frequency pressure waves applied to up-
per airway on respiration in central apnea. J Appl Physiol 1992;73:1141–5.
 Nasal High-Frequency Ventilation 777

22. Di Fiore JM, Martin RJ, Gauda EB. Apnea of prematurity-Perfect storm. Respir
Physiol Neurobiol 2013;189:213–22.
23. van Heerde M, van Genderingen H, Leenhoven T, et al. Imposed work of breath-
ing during high-frequency oscillatory ventilation: a bench study. Crit Care 2006;
10:R23.
24. Bordessoule A, Piquilloud L, Lyazidi A, et al. Imposed work of breathing during
high-frequency oscillatory ventilation in spontaneously breathing neonatal and
pediatric models. Respir Care 2018;63:1085–93.
25. van der Hoeven M, Brouwer E, Blanco CE. Nasal high frequency ventilation in
neonates with moderate respiratory insufficiency. Arch Dis Child Fetal Neonatal
Ed 1998;79:F61–3.
26. De Paoli AG, Lau R, Davis PG, et al. Pharyngeal pressure in preterm infants
receiving nasal continuous positive airway pressure. Arch Dis Child Fetal
Neonatal Ed 2005;90:F79–81.
27. De Paoli AG, Davis PG, Faber B, et al. Devices and pressure sources for admin-
istration of nasal continuous positive airway pressure (NCPAP) in preterm neo-
nates. Cochrane Database Syst Rev 2002;4:CD002977.
28. Jasani B, Ismail A, Rao S, et al. Effectiveness and safety of nasal mask versus
binasal prongs for providing continuous positive airway pressure in preterm
infants-A systematic review and meta-analysis. Pediatr Pulmonol 2018;53:
987–92.
29. De Luca D, Carnielli VP, Conti G, et al. Noninvasive high frequency oscillatory
ventilation through nasal prongs: bench evaluation of efficacy and mechanics.
Intensive Care Med 2010;36:2094–100.
30. De Luca D, Piastra M, Pietrini D, et al. Effect of amplitude and inspiratory time in
a bench model of non-invasive HFOV through nasal prongs. Pediatr Pulmonol
2012;47:1012–8.
31. Green EA, Dawson JA, Davis PG, et al. Assessment of resistance of nasal
continuous positive airway pressure interfaces. Arch Dis Child Fetal Neonatal
Ed 2019;104:F535–9.
32. Gerdes JS, Sivieri EM, Abbasi S. Factors influencing delivered mean airway
pressure during nasal CPAP with the RAM cannula: factors Affecting MAP dur-
ing NCPAP with RAM Cannula. Pediatr Pulmonol 2016;51:60–9.
33. Sharma D, Murki S, Maram S, et al. Comparison of delivered distending pres-
sures in the oropharynx in preterm infant on bubble CPAP and on three different
nasal interfaces. Pediatr Pulmonol 2020;55:1631–9.
34. Mukerji A, Belik J. Neonatal nasal intermittent positive pressure ventilation effi-
cacy and lung pressure transmission. J Perinatol 2015;35:716–9.
35. Aktas S, Unal S, Aksu M, et al. Nasal HFOV with binasal cannula appears effec-
tive and Feasible in ELBW newborns. J Trop Pediatr 2016;62:165–8.
36. Ventre KM, Arnold JH. High frequency oscillatory ventilation in acute respiratory
failure. Paediatr Respir Rev 2004;5:323–32.
37. Klotz D, Schaefer C, Stavropoulou D, et al. Leakage in nasal high-frequency
oscillatory ventilation improves carbon dioxide clearance-A bench study. Pe-
diatr Pulmonol 2017;52:367–72.
38. Schäfer C, Schumann S, Fuchs H, et al. Carbon dioxide diffusion coefficient in
noninvasive high-frequency oscillatory ventilation. Pediatr Pulmonol 2019;54:
759–64.
39. De Luca D, Servel AC, de Klerk A. Noninvasive ventilation interfaces and equip-
ment in neonatology. In: Noninvasive mechanical ventilation and difficult
778 De Luca & Centorrino

weaning in critical care. New York: Springer-Verlag International; 2015.

p. 393–400. https://doi.org/10.1007/978-3-319-04259-6.
40. Tingay DG, John J, Harcourt ER, et al. Are all oscillators created equal? In vitro
performance characteristics of Eight high-frequency oscillatory ventilators.
Neonatology 2015;108:220–8.
41. Grazioli S, Karam O, Rimensberger PC. New generation neonatal high fre-
quency ventilators: effect of oscillatory frequency and working principles on per-
formance. Respir Care 2015;60:363–70.
42. Sivieri EM, Eichenwald E, Bakri SM, et al. Effect of high frequency oscillatory
high flow nasal cannula on carbon dioxide clearance in a premature infant
lung model: a bench study. Pediatr Pulmonol 2019;54:436–43.
43. Rub DM, Sivieri EM, Abbasi S, et al. Effect of high-frequency oscillation on pres-
sure delivered by high flow nasal cannula in a premature infant lung model. Pe-
diatr Pulmonol 2019;54:1860–5.
44. Yuan Y, Sun J, Wang B, et al. A noninvasive high frequency oscillation ventilator:
achieved by utilizing a blower and a valve. Rev Sci Instrum 2016;87:025113.
45. Cools F, Offringa M, Askie LM. Elective high frequency oscillatory ventilation
versus conventional ventilation for acute pulmonary dysfunction in preterm in-
fants. Cochrane Database Syst Rev 2015;3:CD000104.
46. Restrepo RD, Walsh BK. Humidification during invasive and noninvasive me-
chanical ventilation:2012. Respir Care 2012;57:782–8.
47. Fischer HS, Bohlin K, Bührer C, et al. Nasal high-frequency oscillation ventilation
in neonates: a survey in five European countries. Eur J Pediatr 2015;174:
465–71.
48. Ullrich TL, Czernik C, Bührer C, et al. Nasal high-frequency oscillatory ventilation
impairs heated humidification: a neonatal bench study. Pediatr Pulmonol 2017;
52:1455–60.
49. Reddy PI, Al-Jumaily AM, Bold GT. Dynamic surface tension of natural surfac-
tant extract under superimposed oscillations. J Biomech 2011;44:156–63.
50. Autilio C, Echaide M, De Luca D, et al. Controlled hypothermia may improve sur-
factant function in asphyxiated neonates with or without meconium aspiration
syndrome. PLoS One 2018;13:e0192295.
51. De Luca D, Lopez-Rodriguez E, Minucci A, et al. Clinical and biological role of
secretory phospholipase A2 in acute respiratory distress syndrome infants. Crit
Care 2013;17:R163.
52. Simma B, Luz G, Trawöger R, et al. Comparison of different modes of high-
frequency ventilation in surfactant-deficient rabbits. Pediatr Pulmonol 1996;22:
263–70.
53. Kerr CL, Veldhuizen RAW, Lewis JF. Effects of high-frequency oscillation on
Endogenous surfactant in an acute lung injury model. Am J Respir Crit Care
Med 2001;164:237–42.
54. Aspros AJ, Coto CG, Lewis JF, et al. High-frequency oscillation and surfactant
treatment in an acid aspiration model. Can J Physiol Pharmacol 2010;88:14–20.
55. Ali YAH, Seshia MM, Ali E, et al. Noninvasive high-frequency oscillatory ventila-
tion: a Retrospective chart review. Am J Perinatol 2020. https://doi.org/10.1055/
s-0040-1718738.
56. Thatrimontrichai A, Sirianansopa K, Janjindamai W, et al. Comparison of endo-
tracheal Reintubation between nasal high-frequency oscillation and continuous
positive airway pressure in neonates. Am J Perinatol 2020;37:409–14.
 Nasal High-Frequency Ventilation 779

57. qoniewska B, Tousty J, Michalczyk B, et al. The use of noninvasive ventilation

with high frequency in newborns—a single-center experience. Am J Perinatol
2019;36:1362–7.
58. Binmanee A, el Helou S, Shivananda S, et al. Use of high noninvasive respiratory
support pressures in preterm neonates: a single-center experience. J Matern
Fetal Neonatal Med 2017;30:2838–43.
59. Tridente A, De Martino L, De Luca D. Porcine vs bovine surfactant therapy for
preterm neonates with RDS: systematic review with biological plausibility and
pragmatic meta-analysis of respiratory outcomes. Respir Res 2019;20:28.
60. Inthout J, Ioannidis JP, Borm GF. The Hartung-Knapp-Sidik-Jonkman method
for random effects meta-analysis is straightforward and considerably outper-
forms the standard DerSimonian-Laird method. BMC Med Res Methodol
2014;14:25.
61. Wallace BC, Schmid CH, Lau J, et al. Meta-Analyst: software for meta-analysis
of binary, continuous and diagnostic data. BMC Med Res Methodol 2009;9:80.
62. Bottino R, Pontiggia F, Ricci C, et al. Nasal high-frequency oscillatory ventilation
and CO2 removal: a randomized controlled crossover trial. Pediatr Pulmonol
2018;53:1245–51.
63. Chen L, Wang L, Ma J, et al. Nasal high-frequency oscillatory ventilation in pre-
term infants with respiratory distress syndrome and ARDS after extubation.
Chest 2019;155:740–8.
64. Iranpour R, Armanian AM, Abedi AR, et al. Nasal high-frequency oscillatory
ventilation (nHFOV) versus nasal continuous positive airway pressure (NCPAP)
as an initial therapy for respiratory distress syndrome (RDS) in preterm and
near-term infants. BMJ Paediatr Open 2019;3:e000443.
65. Klotz D, Schneider H, Schumann S, et al. Non-invasive high-frequency oscilla-
tory ventilation in preterm infants: a randomised controlled cross-over trial.
Arch Dis Child Fetal Neonatal Ed 2018;103:F1–5.
66. Lou W, Zhang W. Noninvasive high-frequency oscillatory ventilation versus nasal
continuous positive airway pressure in premature infants with respiratory
distress syndrome after weaning: a randomized controlled trial. Guangdong
Med J 2017;38:2037–40.
67. Lou W, Zhang W, Yuan L, et al. Comparative study of noninvasive high-
frequency oscillatory ventilation and bilevel positive airway pressure ventilation
for preterm infants with respiratory distress syndrome. Chin Gen Pract 2018;21:
1983–8.
68. Malakian A, Bashirnezhadkhabaz S, Aramesh MR, et al. Noninvasive high-
frequency oscillatory ventilation versus nasal continuous positive airway pres-
sure in preterm infants with respiratory distress syndrome: a randomized
controlled trial. J Mater Fetal Neonatal Med 2020;33:2601–7.
69. Mukerji A, Sarmiento K, Lee B, et al. Non-invasive high-frequency ventilation
versus bi-phasic continuous positive airway pressure (BP-CPAP) following
CPAP failure in infants<1250g: a pilot randomized controlled trial. J Perinatol
2017;37:49–53.
70. Rüegger CM, Lorenz L, Kamlin COF, et al. The effect of noninvasive high-
frequency oscillatory ventilation on desaturations and bradycardia in very
preterm infants: a randomized crossover trial. J Pediatr 2018;201:
269–73.e2.
71. Zhu XW, Zhao JN, Tang SF, et al. Noninvasive high-frequency oscillatory venti-
lation versus nasal continuous positive airway pressure in preterm infants with
780 De Luca & Centorrino

moderate-severe respiratory distress syndrome: a preliminary report. Pediatr

Pulmonol 2017;52:1038–42.
72. Zhu X, Yan J, Ran Q, et al. Noninvasive high-frequency oscillatory ventilation
versus for respiratory distress syndrome in preterm infants: a preliminary report.
Chin J Neonatol 2017;32:291–4.
73. Buzzella B, Claure N, D’Ugard C, et al. A randomized controlled trial of two nasal
continuous positive airway pressure levels after extubation in preterm infants.
J Pediatr 2014;164:46–51.
74. Lemyre B, Davis PG, De Paoli AG, et al. Nasal intermittent positive pressure
ventilation (NIPPV) versus nasal continuous positive airway pressure (NCPAP)
for preterm neonates after extubation. Cochrane Database Syst Rev 2017;2:
CD003212.
75. Subramaniam P, Ho JJ, Davis PG. Prophylactic nasal continuous positive airway
pressure for preventing morbidity and mortality in very preterm infants. Co-
chrane Database Syst Rev 2016;6:CD001243.
76. Raschetti R, Yousef N, Vigo G, et al. Echography-guided surfactant therapy to
improve Timeliness of surfactant replacement:A Quality improvement project.
J Pediatr 2019;212:137–43.
77. Bahadue FL, Soll R. Early versus delayed selective surfactant treatment for
neonatal respiratory distress syndrome. Cochrane Database Syst Rev 2012;
11:CD001456.
78. Esquinas A, Mantellini E, Benitez Leon F, et al. International study use of high
frequency chest wall oscillation (HFCWO) in secretion management in the me-
chanically ventilated patient. Proceedings of ESICM 2010 Congress. Barcelona
(Spain): Intensive Care Med 2010;36(S2):105(n.0080). https://doi.org/10.1007/
s00134-010-1999-x.
79. Zhu XW, Shi Y, Shi LP, et al, NHFOV Study Group. Non-invasive high-frequency
oscillatory ventilation versus nasal continuous positive airway pressure in pre-
term infants with respiratory distress syndrome: study protocol for a multi-
center prospective randomized controlled trial. Trials 2018;19:319.
80. Shi Y, De Luca D. NASal OscillatioN post-Extubation (NASONE) study group.-
Continuous positive airway pressure (CPAP) vs noninvasive positive pressure
ventilation (NIPPV) vs noninvasive high frequency oscillation ventilation
(NHFOV) as post-extubation support in preterm neonates: protocol for an
assessor-blinded, multicenter, randomized controlled trial. BMC Pediatr 2019;
19:256.
81. De Luca D. Noninvasive high-frequency ventilation and the errors from the past:
designing simple trials neglecting complex respiratory physiology. J Perinatol
2017;37:1065–6.
82. Fischer HS, Rimensberger PC. Early noninvasive high-frequency oscillatory
ventilation in the primary treatment of respiratory distress syndrome. Pediatr Pul-
monol 2018;53:126–7.
83. Thorpe KE, Zwarenstein M, Oxman AD, et al. A pragmatic–explanatory contin-
uum indicator summary (PRECIS): a tool to help trial designers. J Clin Epidemiol
2009;62:464–75.
84. De Luca D, Harrison D, Peters MJ. ‘Lumping or splitting’ in paediatric acute res-
piratory distress syndrome (PARDS). Intensive Care Med 2018;44:1548–50.
85. Gregoretti C, Cortegiani A, Maggiore SM. Noninvasive oscillatory ventilation
(NHFOV) in infants: another brick in the wall of paediatric noninvasive ventila-
tion? Pediatr Pulmonol 2016;51:663–4.
 Nasal High-Frequency Ventilation 781

86. Steinhorn R, Davis JM, Göpel W, et al. Chronic pulmonary insufficiency of pre-
maturity: developing optimal Endpoints for Drug development. J Pediatr 2017;
191:15–21.e1.
87. Loi B, Vigo G, Baraldi E, et al. LUSTRE study group.Lung ultrasound to monitor
extremely preterm infants and predict BPD: multicenter longitudinal cohort
study. Am J Respir Crit Care Med 2020. https://doi.org/10.1164/rccm.202008-
3131OC.
88. Brat R, Yousef N, Klifa R, et al. Lung ultrasonography score to evaluate oxygen-
ation and surfactant need in neonates treated with continuous positive airway
pressure. JAMA Pediatr 2015;169:e151797.
89. De Luca D, Romain O, Yousef N, et al. Monitorages physiopathologiques en ré-
animation néonatale. J Pediatrie Puericulture 2015;28:276–300.
90. Tusman G, Acosta CM, Costantini M. Ultrasonography for the assessment of
lung recruitment maneuvers. Crit Ultrasound J 2016;8:8.
91. van Dijk M, Peters JWB, van Deventer P, et al. The COMFORT Behavior Scale: a
tool for assessing pain and sedation in infants. Am J Nurs 2005;105:33–6.
92. Debillon T, Zupan V, Ravault N, et al. Development and initial validation of the
EDIN scale, a new tool for assessing prolonged pain in preterm infants. Arch
Dis Child Fetal Neonatal Ed 2001;85:F36–41.
93. Piastra M, De Luca D, Costa R, et al. Neurally adjusted ventilatory assist vs pres-
sure support ventilation in infants recovering from severe acute respiratory
distress syndrome: nested study. J Crit Care 2014;29:312.e1–5.
94. Singh Y, Tissot C, Fraga MV, et al. International evidence-based guidelines on
point of care ultrasound (POCUS) for critically ill neonates and children issued
by the POCUS working group of the European society of paediatric and
neonatal Intensive care (ESPNIC). Crit Care 2020;24(1):65.
95. Piastra M, De Luca D, De Carolis MP, et al. Nebulized iloprost and noninvasive
respiratory support for impending hypoxaemic respiratory failure in formerly
preterm infants: a case series. Pediatr Pulmonol 2012;47:757–62.
96. DiBlasi RM, Crotwell DN, Shen S, et al. Iloprost Drug delivery during infant con-
ventional and high-frequency oscillatory ventilation. Pulm Circ 2016;6:63–9.
97. Allan PF, Osborn EC, Chung KK, et al. High-frequency percussive ventilation
Revisited. J Burn Care Res 2010;31:510–20.
98. Dumas De La Roque E, Bertrand C, Tandonnet O, et al. Nasal high frequency
percussive ventilation versus nasal continuous positive airway pressure in tran-
sient tachypnea of the newborn: a pilot randomized controlled trial
(NCT00556738): NHFPV versus NCPAP in Transient Tachypnea of the Newborn.
Pediatr Pulmonol 2011;46:218–23.
99. Renesme L, Dumas de la Roque E, Germain C, et al. Nasal high-frequency
percussive ventilation vs nasal continuous positive airway pressure in newborn
infants respiratory distress: a cross over clinical trial. Pediatr Pulmonol 2020;55:
2617–23.
100. Renesme L, Elleau C, Nolent P, et al. Effect of high-frequency oscillation and
percussion versus conventional ventilation in a piglet model of meconium aspi-
ration: Hi-Frequency vs. Conventional Ventilation in MAS. Pediatr Pulmonol
2013;48:257–64.
101. Messier SE, DiGeronimo RJ, Gillette RK. Comparison of the Sensormedics
3100A and Bronchotron transporter in a neonatal piglet ARDS model. Pediatr
Pulmonol 2009;44:693–700.
782 De Luca & Centorrino

102. Rehan VK, Fong J, Lee R, et al. Mechanism of reduced lung injury by high-
frequency nasal ventilation in a preterm lamb model of neonatal chronic lung
disease. Pediatr Res 2011;70:462–6.
103. Null DM, Alvord J, Leavitt W, et al. High-frequency nasal ventilation for 21d main-
tains gas exchange with lower respiratory pressures and promotes alveolariza-
tion in preterm lambs. Pediatr Res 2014;75:507–16.