+ MODEL

Pediatrics and Neonatology (2017) xx, 1e7

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http://www.pediatr-neonatol.com

Original Article

A randomized pilot study comparing the role

of PEEP, O2 flow, and high-flow air for
weaning of ventilatory support in very low
birth weight infants
Chang-Yo Yang a, Mei-Chin Yang b, Shih-Ming Chu a,b,
Ming-Chou Chiang a,b, Reyin Lien a,b,*

a
Division of Neonatology, Department of Pediatrics, Chang Gung Medical Center, School of Medicine,
Chang Gung University, Taoyuan, 33305, Taiwan
b
Department of Respiratory Care, Chang Gung Medical Center, Taoyuan, 33305, Taiwan

Received Jul 13, 2016; received in revised form Jan 16, 2017; accepted Feb 24, 2017
Available online - - -

Key Words Background: There is a lack of evidence to guide step-wise weaning of positive pressure respira-
weaning; tory support for premature infants. This study sought to compare the efficacy of three weaning
nasal continuous protocols we designed to facilitate weaning of very low birth weight (VLBW, less than 1500 g) pre-
positive airway term infants from nasal continuous positive airway pressure (NCPAP) support.
pressure; Methods: This was a prospective, randomized, controlled trial of VLBW preterm infants who
very low birth weight; received positive pressure ventilatory support in our neonatal intensive care unit (NICU) from
nasal air flow April 2008 through March 2009. When these infants were weaned to CPAP as their last step of res-
piratory support, they would be randomly assigned to one of the following three groups as their
further weaning methods (M): (M1) CPAP group, (M2) O2 flow group, and (M3) air flow group. The
time period they needed to wean off any kind of respiratory support, as well as the likelihood of
developing relevant prematurity related morbidities, were compared among patients using
different weaning modalities.
Results: 181 patients were enrolled in the study. Their gestational age (GA) and birth weight (BW)
were 29.1
2.5, 28.7
2.4, 28.7
2.4 (mean
SD) weeks and 1142
232, 1099
234,
1083
219 g, in M1, M2 and M3, respectively. The time (period) needed to wean off support
was 16.0
10.0 days (M1), 11.6
6.4 days (M2), and 15.0
8.9 days (M3), respectively
(p Z .033). Incidence of retinopathy of prematurity (ROP) and bronchopulmonary dysplasia
(BPD) were both significantly higher in the O2 flow group (p Z .048).
Conclusions: Although using low oxygen flow significantly shortens CPAP weaning time, it may in-
crease risks of BPD and ROP, both known to be related to oxygen toxicity. Unless the infant has BPD

* Corresponding author. Division of Neonatology, Department of Pediatrics, No. 5, Fu-Shing Street, Kwei-Shan, Taoyuan, 33305, Taiwan.
Fax: þ886 3 328 8957.
E-mail address: reyinl@cgmh.org.tw (R. Lien).

http://dx.doi.org/10.1016/j.pedneo.2017.02.005
1875-9572/Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open access article under the
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Yang C-Y, et al., A randomized pilot study comparing the role of PEEP, O2 flow, and high-flow air for
weaning of ventilatory support in very low birth weight infants, Pediatrics and Neonatology (2017), http://dx.doi.org/10.1016/
j.pedneo.2017.02.005
 + MODEL
2 C.-Y. Yang et al
and is O2-dependent, clinicians should consider using air flow or just splinting with no support at
all when weaning NCPAP.
Copyright ª 2017, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/
4.0/).
1. Introduction
Respiratory support has become the mainstay of successful
care for the premature infants. We also made significant
progress in the understanding and management of neonatal
respiratory disorders. Nasal continuous positive airway
pressure (NCPAP) therapy is used as a primary support
modality for infants with respiratory distress syndrome and
apnea of prematurity. It is also used as the key post-
extubation respiratory support.1e3 Multiple studies have
demonstrated that NCPAP is a safe treatment modality with
no increase in short-term4e9 or long-term morbidities.10,11
It also has other beneficial effects, including the induc-
tion of lung growth.12
In contrast to the variable but well-examined strategies
in the initial management of respiratory distress in pre-
mature infants, there has been a lack of evidence-based
guidelines for the step-wise weaning of these patients. This
is especially true when the weaning process reaches the
last part of pressure support, which often is NCPAP, either
with low fraction of inspiratory oxygen (FiO2) or room air.
Further weaning techniques differ considerably from one
clinician to another. At this point, drastic weaning may lead
to hypoxemia or a rebound of respiratory distress, whereas
sluggish weaning may be the cause of a prolonged hospital
stay or even pose a risk of nosocomial infection.
After weaning from CPAP therapy, oxygen is often deliv-
ered through a nasal cannula (NC) at low-flow rate (up to
0.5e1 L/min). However, the preterm infant would more
often have been in 21% FiO2 with CPAP prior to this step, in
which case they were being treated for apnea of prematurity.
Previous studies have demonstrated that high-flow nasal
cannula can deliver positive distending pressure equivalent
to conventional forms of nasal CPAP.13,14 High flow was
considered as a gas flow greater than 1 L/min in these
studies. HF-CPAP has also been shown to be as effective as
conventional CPAP in managing apnea of prematurity.
Furthermore, it has been well illustrated in immature
mammals that stimulation of the trigeminal afferent can
trigger control of respiration.15 Moreover, researchers also
found that electrical stimulation of the sensory root of
trigeminal nerve can induce significant responses on the
pontine respiratory neurons in newborn rats, a finding
indicating presence of respiratory neurons in the rostral
pons approximately overlapping with the Kölliker-fuse nu-
cleus receiving trigeminal input.16 Based on the neuroana-
tomical distribution and pathophysiology of apnea of
prematurity, tactile stimulation by an air flow to the nos-
trils, even in a moderate flow rate, could also be used as a
modality in this circumstance.
Therefore, we performed this prospective, randomized
controlled trial to compare the efficacy of the three
Please cite this article in press as: Yang C-Y, et al., A randomized pilot study comparing the role of PEEP, O2 flow, and high-flow air for
weaning of ventilatory support in very low birth weight infants, Pediatrics and Neonatology (2017), http://dx.doi.org/10.1016/
j.pedneo.2017.02.005
weaning protocols we designed for the weaning of VLBW
preterm infants from positive airway pressure support.
With this study, we also aimed to set forth a standardized
weaning practice.
2. Methods
2.1. Subjects
This study was conducted in the NICU of Chang Gung Med-
ical Center in Lin Kou, Taiwan between August 2008 and
September 2009. All VLBW preterm infants (BW  1500 g
and GA < 34 weeks) admitted to our NICU who also required
positive pressure ventilation support were eligible for this
study. The study population was comprised of 189 patients
(106 males and 83 females), with a mean (
SD) gestational
age of 28.8
2.4 weeks, and BW of 1108
228 gms. The
exclusion criteria included the following: (1) CPAP use for
<24 h; (2) no parental consent for the study; (3) weaning
interrupted by a need for surgery; (4) transfer of the pa-
tient to other hospital; (5) FiO2 use >25% at the time of
CPAP weaning; (6) body weight less than 750 g at the time
of randomization; (7) congenital abnormalities affecting
cardiopulmonary performance; (8) intraventricular hemor-
rhage (IVH) Gr. III or above. Patient enrollment was started
when these preterm infants’ ventilator support was
decreasing down to CPAP and they were ready for further
weaning. This study was reviewed and approved by the
Institutional Review Board (IRB) of Human Investigation
Committee at the Chang Gung Memorial Hospital, and a
signed informed consent was obtained from the parent(s) or
guardian(s) of each participant before the study.
2.2. Measurements and protocol
Once the infant was stable and achieved target oxygen
saturation of 88%e93% on a CPAP pressure of 5e7 cm H2O
and an FiO2 no greater than 0.21 for 24 h, then the infant
would be checked to meet the following criteria before
starting weaning of CPAP: a body weight of 1250 g or larger;
a hemoglobin level of at least 10 gm/dL; no use of vaso-
active or sedative agents. In addition, the attending
physician agreed with the patient’s weaning. Once started,
the enrolled premature infants would be randomly assigned
to one of the three weaning protocols (Fig. 1).
(A) Method 1 (M1): The CPAP was continued at 4e6 cm
H2O for another five days, then the CPAP was taken
“OFF” and the premature infant remained in the crib
on oxygen by nasal cannula as needed or in room air
with a plan to remain off CPAP. If the infant failed
 + MODEL
Ventilator weaning in very low birth weight infants
Figure 1 Design of the three weaning protocols.
without CPAP support, he/she would be restarted
“ON” CPAP at the previous level for at least 48 h.
Also, the infant would not undergo another weaning
trial until the stability criteria were again satisfied.
(B) Method 2 (M2): The infant would be taken “OFF”
CPAP and placed on O2 flow at 0.2 L/min by NC for an
increasing period of time according to the weaning
protocol. Once the baby was able to tolerate 12 h
“OFF” CPAP with an O2 flow of 0.2 L/min, an attempt
was made to stop the CPAP during the next off period.
If the infant met the “failure criteria,” the baby
would be restarted “ON” CPAP for 6 h or until the
“stability criteria” were met.
(C) Method 3 (M3): The CPAP was weaned “OFF” in a
similar way to “M2” but the premature infants in this
group were given air flow at 1.5 L/min by NC when
CPAP was taken off.
Duration of time off the ventilator would be increased
daily in the same manner for all three groups. The weaning
process would be considered successful if the patient
remained stable off CPAP for five days.
The trial would be suspended if a patient had at least
two of the following signs of distress:
(1) Visible increase in the work of breathing (intercostal
recession and accessory muscles contributing for
respiration) with a respiratory rate of >75/min;
(2) An increase in apnea and/or bradycardia events and/
or >2 desaturation episodes in 1 h for the previous 6-
hr period.
(3) An increase in the FiO2 of more than 25% to maintain
an oxygen saturation of >86%.
(4) An arterial blood gas with pH of <7.2
(5) A PaO2/transcutaneous PaO2 ratio of >65 mmHg
(6) A major apnea or bradycardia episode requiring
resuscitation.
Please cite this article in press as: Yang C-Y, et al., A randomized pilot study comparing the role of PEEP, O2 flow, and high-flow air for
weaning of ventilatory support in very low birth weight infants, Pediatrics and Neonatology (2017), http://dx.doi.org/10.1016/
j.pedneo.2017.02.005
3
Patients who had none of these features at the end of
the trial were considered to be successfully weaned. If a
patient had signs of poor tolerance at any time during the
trial, the NCPAP ventilation would be reinstituted for
another five days before the next trial. If the infant met the
criteria of weaning failure for several times and was later
diagnosed with moderate-to-severe bronchopulmonary
dysplasia (BPD), the NCPAP ventilation would be reinsti-
tuted and these patients were taken off the study.
The following factors were analyzed and compared
among the three groups of patients: demographic data at
birth; length of weaning duration; body weight and post-
conceptional age (PCA) of successful weaning; causes of
failure; risks of relevant morbidities of prematurity [reti-
nopathy of prematurity (ROP) and BPD]; and the duration of
hospital stay.
2.3. Statistical analysis
Statistical analysis was performed using SPSS 17 for Win-
dows. Data were expressed as either mean
standard error
of the mean or standard deviation where appropriate.
Analysis of variance (ANOVA) was used for repeated mea-
sures. The post-hoc analysis used Fisher’s least significant
difference test. Continuous variables were analyzed by
Student’s t-test and by the one-way ANOVA-test for non-
parametric data. A p value of <0.05 was considered sta-
tistically significant.
3. Results
3.1. Study population and clinical outcomes
Of the 189 study patients, 8 patients with signs of poor
tolerance during the weaning procedure who were even-
tually diagnosed with severe BPD were excluded from the
 + MODEL
4 C.-Y. Yang et al

M1 study group. One hundred and eighty-one VLBW infants
were randomly assigned to one of the 3 groups: M1
(n Z 55), M2 (n Z 63), and the M3 (n Z 63) (Fig. 1). De-
mographic data are shown in Table 1 and they were similar
among the groups. The birth weight and GA were
1142
232 gms., 1099
234 gms., 1083
219 gms.,
(mean
SD); and 29.1
2.5 weeks, 28.7
2.4 weeks,
28.7
2 0.4 weeks for M1, M2, and M3 study groups,
respectively. There were no significant differences in cor-
rected GA or body weight among the groups at the time
of randomization or commencement of weaning either
(Table 1).
3.2. Outcome analysis
The effectiveness of weaning is summarized in Table 2. The
CPAP weaning success rate on the first trial was 53%, 56%,
and 43% (p Z .044) in M1, M2 and M3 groups, respectively;
the oxygen days in groups M1, M2, and M3 were 47
28 days
(mean), 53
39 days, and 55
28 days, respectively. The
mean body weight and PMA at successful weaning in the
three groups were 1924
454 g and 35.4
2.0 weeks,
1821
415 g and 34.9
1.9 weeks; and 1973
576 g and
35.5
2.2 weeks in M1, M2 and M3, respectively (Table 3).
The time (duration) needed to wean off pressure support
from commencement of weaning trial was 16.0
10.5 days
(mean
SD), 11.6
4.2 days, and 15.0
8.1 days
(p Z .033). Serum theophylline levels were measured at
the time infants failed and succeeded weaning, and there
were no differences among the 3 groups.
Table 4 lists the risks of relevant morbidities using the 3
weaning techniques. Significantly, the group of infants who
received O2 flow as weaning measure had higher occurrence
of ROP and BPD than those infants who received CPAP or
nasal air flow support.
4. Discussion
In this study we found that in preterm infants without se-
vere BPD, instead of using O2 flow, weaning of nasal CPAP
could be achieved alternatively by using air flow NC, or by
taking off CPAP and giving no support (splitting CPAP with
no support) gradually. We also found that although weaning
by O2 flow could achieve weaning faster than using air flow
or by splitting CPAP, infants using O2 flow to wean had a
higher risk of development of BPD and ROP. In our study,
weaning with O2 flow shortened the weaning duration, but
the length of hospital stay did not decrease.
NCPAP is used as respiratory support after extubation
and in the management of apnea of prematurity. It has been
shown to increase the functional residual capacity, improve
oxygenation, and decrease the rate of BPD.2 Despite the
many documented benefits of NCPAP, it is a form of respi-
ratory support that has its complications including trau-
matic injuries to the nose, pulmonary air leak syndromes
and gaseous distension of the stomach. Therefore, reducing
the duration of NCPAP application could decrease associ-
ated complications. However, premature discontinuation of
NCPAP has potential hazards including increased apnea,
oxygen demand and the need to restart NCPAP, and occa-
sionally mechanical ventilation. Oxygen delivery through
NC for neonates is preferred by caregivers because of the
ease of administration and the ability to feed and care for
the infant, while continuing oxygen administration. Infants
with NC also have increased mobility compared with those
who are weaned with intermittent CPAP off, which may
allow increased interactions with the parents, caregivers,
and environment and may be developmentally beneficial.17
These benefits are balanced by some drawbacks, including
the instability of oxygen administration in transition be-
tween oral and nasal breathing, the drying of the nasal
mucosa, and the lack of precise knowledge about the
delivered oxygen concentration. The last factor may
contribute to the highly variable weaning practices among
institutions and physicians and to the continued prescrip-
tion of oxygen through NC which delivers concentrations of
oxygen that are effectively similar to room air. Such prac-
tices may contribute to the unnecessary days of oxygen
delivery, hospitalization, and costs of care.18
The use of air and/or mixture with oxygen gas flow
through NC to deliver end-expiratory pressure to reduce the

Table 1 Characteristics of the study subjects.

M1 (n Z 55) (%) M2 (n Z 63) (%) M3 (n Z 63) (%)
Gestational age 29.1
2.5 28.7
2.4 28.7
2.4
(weeks)
Birth weight (g) 1142
232 1099
234 1083
219
Apgar 1 min 6
2 5
2 6
2
Apgar 5 min 8
1 7
2 7
2
SGA 10 (18%) 6 (10%) 9 (14%)
Male 31 (56%) 39 (62%) 27 (43%)
Initial OI 3.4
2.7 3.4
1.9 6.1
3.2
Surfactant dosage 0.5
0.6 0.7
0.8 0.5
0.8
Discharge PMA 39.1
3.8 39.2
3.1 39.8
2.9
Length of stay (d) 70
31 74
23 78
28
PMA start wean (weeks) 33.3
2.4 34.3
2.4 33.6
2.
Start weaning body weight (g) 1715
366 1610
284 1717
388
SGA, small for gestational age; OI, oxygen index; NEC, necrotizing enterocolitis; PMA, post menstrual age; GA, gestational age.

Please cite this article in press as: Yang C-Y, et al., A randomized pilot study comparing the role of PEEP, O2 flow, and high-flow air for
weaning of ventilatory support in very low birth weight infants, Pediatrics and Neonatology (2017), http://dx.doi.org/10.1016/
j.pedneo.2017.02.005
 + MODEL
Ventilator weaning in very low birth weight infants 5

Table 2 Respiratory data of the study subjects.

M1 (n Z 55) (%) M2 (n Z 63) (%) M3 (n Z 63) (%)
1st weaning success rate 29/55 (53%) 35/63 (56%) 27/63 (43%)*
Total pressure support days 47
28 48
24 55
28
Duration between weaned-off 0 5
2 0
CPAP and O2 independence
Oxygen days 47
28 53
39 55
28
*p < .05, significant.

Table 3 Comparison weaning results between NCPAP, low-flow O2, and high-flow air.
M1 (n Z 55) (%) M2 (n Z 63) (%) M3 (n Z 63) (%)
Success PMA (Off NCPAP) 35.4
2.0 weeks 33.4
1.1 weeks 34.8
2.0 weeks
Success PMA (Off any support) 35.4
2.0 weeks 34.9
1.9 weeks 35.5
2.2 weeks
Success BW 1924
454 g 1821
415 g 1973
576 g
Weaning duration 16.0
10.5 days 11.6
4.2 days* 15.0
8.1 days
Theophylline level e Failure 4.70 4.99 4.72
e Success 4.08 4.35 4.32
Cause of failure
-Apnea or bradycardia 31/52 (60%) 17/23 (74%) 33/57 (58%)
-SpO2 < 90% 21/52 (40%) 2/23 (8%)* 21/57 (37%)
-Resp. distress 14/52 (27%) 8/23 (35%)* 10/57 (18%)
*P < .05, significant.
PMA, post menstrual age; BW, body weight.

frequency of apnea and desaturation has not been tested
adequately, although such use has become increasingly
common. As stated previously, our study design of Method 3
to wean CPAP with air flow was based on the pathophysi-
ology of apnea of prematurity to provide tactile stimulation
by an air flow to the nostrils, and use it as a modality for
weaning. However, although some recent researchers tried
to use “high flow” by nasal cannulae for newborn infants
reaching flow rates as high as 6 L/min, or 4 L/min/Kg,
previous reports demonstrated that the air flow rate we
used for weaning (1.5 L/min) was high enough to generate a
positive distending pressure for the premature infants.
Locke et al.13 first described the ability of NC to generate
inadvertent positive end-distending pressure. Others later
exploited this possibility as a deliberate therapeutic inter-
vention. Sreenan et al.14 found that NC used at flow rates
between 1.0 and 2.5 L/min for preterm infants delivered
CPAP as high as 8 cm H2O and performed equivalently to the
conventional CPAP delivery among the infants who were
studied at a mean weight of 1260 g.
Table 4 Risk of complications.
M1 (n Z 55) (%) M2 (n Z 63) (%) M3 (n Z 63) (%)
BPD 26/55 (47%) 40/63 (63%)* 27/54 (50%)
ROP 21/55 (38%) 33/63 (52%)* 24/54 (44%)
*p < .05, significant.
BPD, bronchopulmonary dysplasia; ROP, retinopathy
prematurity.
Sixty-three infants in our study were randomized to
receive NC with room air at a flow rate of 1.5 L/min.
Twenty-seven (43%) of those infants were weaned suc-
cessfully to room air during the first weaning procedure,
but 57% were not. It would be difficult to delineate for
those who succeeded weaning through “high” air flow by
NC, whether the weaning was facilitated by the positive
distending pressure generated by the air flow, or by tactile
stimulation of the nostril to decrease apnea episodes.
It was among our aims for this study to eliminate un-
necessary oxygen exposure by replacing air flow or splinting
CPAP for low flow oxygen NC in the weaning process. It is
still alarming to us that although we achieved CPAP wean-
ing faster by using low flow O2, we also noted the fact that
infants using O2 flow to wean had higher risk to develop
significant BPD and ROP. Both BPD and ROP are known to be
complications of premature infants that are highly likely to
be associated with oxygen toxicity. In this study, all infants
should have had FiO2 of less than 25% before they qualified
for CPAP weaning. Considering our results, we want to un-
derscore that everyone caring for the preterm infants
should be aware that even a brief prolongation of such a
small amount of oxygen use may result in a significant dif-
ference in major complications.
The body weight and PMA at the time of successful
weaning showed no significant difference among the three
groups. While the body weight was still less than 1500 g,
about one-third of the infants were able to come off NCPAP in
groups using M1 and M3, whereas 50% of the infants were
successfully weaned in group M2. When the infants were
of stable, we were able to discontinue the positive pressure
support in two-thirds of the NCPAP group of patients, and only

Please cite this article in press as: Yang C-Y, et al., A randomized pilot study comparing the role of PEEP, O2 flow, and high-flow air for
weaning of ventilatory support in very low birth weight infants, Pediatrics and Neonatology (2017), http://dx.doi.org/10.1016/
j.pedneo.2017.02.005
 + MODEL
6 C.-Y. Yang et al
a small proportion required a restart of NCPAP within 48 h.
Therefore, we found that the clinical stability of the baby was
the key point to weaning pressure support, rather than body
weight or PMA. On the other hand, serum theophylline levels
were measured at the time of infants’ failure or success in
weaning, and all fell in a similar range with no difference
among the 3 groups. This indicates that theophylline prob-
ably has little contribution to the weaning of CPAP.
In this study, 8 patients were excluded from the study
when they met the diagnosis of severe BPD during weaning,
and all these infants were originally randomized to M1
group. Our study design was that M1 group infants were put
on 5 more days of CPAP with 21e25% FiO2 when they were
stable enough to be weaned, and the diagnosis of BPD in this
study was based on NICHD consensus, which defines oxygen
use (greater than 21% FiO2) for 28 days, and either further
oxygen dependency of >21% but less than 30% (moderate)
or further oxygen dependency of >30% and/or PEEP de-
pendency (severe).19 The reason why only patients from M1
group were excluded was due to “moderate-to-severe BPD”
was quite clear: according to our protocol when they failed
weaning, they would have been put back to the previous
weaning step for at least 48 h before they could go on
weaning again. In M1 group patients, this resulted in at least
2 more days of PEEP dependency, which qualified them for
severe BPD and meant that they were excluded from the
study (whereas M2 and M3 were on <25% FiO2 and no PEEP
when they started weaning.) We think this could also
explain why in our study those infants who were weaned by
using O2 flow were able to shorten weaning duration (to
take off CPAP), but the length of hospital stay did not
decrease. Because those infants who were qualified for the
diagnosis of moderate but not severe BPD by this definition
were the ones to lengthen oxygen dependency (of less than
30% FiO2) and hospital stay after CPAP was weaned off.
4.1. Limitations
The limitation of this study is that it was conducted more
than 6 years ago and before there was a consensus of
keeping target oxygen saturation at the range of 90e95%
for the preterm infants with RDS. During the study time
and, at that time for all patients with RDS, we were aiming
at a lower saturation of 88e93% which could be sub-optimal
from current evidence. Nevertheless, this should not have
affected the study results.
5. Conclusion
Although using the O2 flow NC may shorten the time to
wean off NCPAP, the length of hospital stay does not differ
from other weaning methods. Infants using low oxygen flow
to wean CPAP may have higher risk to develop significant
BPD and ROP. An essential factor to guarantee successful
weaning is clinical stability, and not body weight or age of
the patient. We suggest that weaning should be considered
early if the infant is stable on nasal CPAP of 4e6-cm H2O
and has a FiO2 of <0.25. Unless the infant has BPD and is O2-
dependent, clinicians should consider using air flow NC or
just splinting with no support to wean CPAP.
Please cite this article in press as: Yang C-Y, et al., A randomized pilot study comparing the role of PEEP, O2 flow, and high-flow air for
weaning of ventilatory support in very low birth weight infants, Pediatrics and Neonatology (2017), http://dx.doi.org/10.1016/
j.pedneo.2017.02.005
Conflict of Interest Statement
The authors declare that there is no conflict of interest.
Acknowledgments
We thank all of the nurses and respiratory therapists who
assisted with our work and, most of all, the families that
participated in the study.
References
1. Goldsmith JP, Karotkin EH. Continuous positive airway pres-
sure. In: Goldsmith JP, Karotkin EH, editors. Assisted ventila-
tion of the neonate. 4th ed. Philadelphia, PA: Saunders; 2003.
p. 127e47.
2. Polin RA, Sahni R. Newer experience with CPAP. Semin Neo-
natol 2002;7:379e89.
3. Sant’Anna G, Keszler M. Weaning from mechanical ventilation.
Clin Perinatol 2012;39:543e62.
4. Finer NN, Carlo WA, Duara S, Fanaroff AA, Donavan EF,
Wright LL, et al. Delivery room continuous positive airway
pressure/positive end-expiratory pressure in extremely low
birth weight infants: a feasibility trial. Pediatrics 2004;114:
651e7.
5. Aly H, Milner JD, Patel K, El-Mohandes AA. Does the experience
with the use of nasal continuous positive airway pressure
improve over time in extremely low birth weight infants? Pe-
diatrics 2004;114:697e702.
6. De Klerk AM, De Klerk RK. Nasal continuous positive airway
pressure and outcomes of preterm infants. J Paediatr Child
Health 2001;37:161e7.
7. Van Marter LJ, Allred EN, Pagano M, Sanocka U, Parad R,
Moore M, et al. Do clinical markers of barotrauma and oxygen
toxicity explain interhospital variation in rates of chronic lung
disease? The Neonatology Committee for the Developmental
Network. Pediatrics 2000;105:1194e201.
8. Morley CJ, Davis PG, Doyle LW, Brion LP, Hascoet JM, Carlin JB.
Nasal CPAP or intubation at birth for very preterm infants. N
Engl J Med 2008;358:700e8.
9. SUPPORT Study Group of the Eunice Kennedy Shriver NICHD
Neonatal Research Network, Finer NN, Carlo WA, Walsh MC,
Rich W, Gantz MG, et al. Early CPAP versus surfactant in
extremely preterm infants. N Engl J Med 2010;362:1970e9.
10. Hansen BM, Hoff B, Greisen G, Mortensen EL, Danish ETFOL
Study Group. Early nasal continuous positive airway pressure in
a cohort of the smallest infants in Denmark: neuro-
developmental outcome at five years of age. Acta Paediatr
2004;93:190e5.
11. Dahl M, Kamper J. Physical outcome and school performance of
very low birth weight infants treated with minimal handling
and early nasal CPAP. Acta Paediatr 2006;95:1099e103.
12. Zhang S, Garbutt V, McBride JT. Strain-induced growth of the
immature lung. J App Physiol 1996;81:1471e6.
13. Locke RG, Wolfson MR, Shaffer TH, Rubenstein SD,
Greenspan JS. Inadvertent administration of positive end-
distending pressure during nasal cannula flow. Pediatrics
1993;91:135e8.
14. Sreenan C, Lemke RP, Hudson-Mason A, Osiovich H. High- flow
nasal cannulae in the management of apnea of prematurity: a
comparison with conventional nasal continuous positive airway
pressure. Pediatrics 2001;107:1081e3.
15. Gluckman PD, Johnston BM. Lesions in the upper lateral pons
abolish the hypoxic depression of breathing in unanaesthetized
fetal lambs in utero. J Physiol 1987;382:373e83.
 + MODEL
Ventilator weaning in very low birth weight infants 7

16. Kobayashi S, Onimaru H, Inoue M, Inoue T, Sasa R. Localization
and properties of respiratory neurons in the rostral pons of the
newborn rat. Neuroscience 2005;134:317e25.
17. Dani C, Pratesi S, Migliori C, Bertini G. High flow nasal cannula
therapy as respiratory support in the preterm infant. Pediatr
Pulmonol 2009;44:629e34.
18. Walsh M, Engle W, Laptook A, Kazzi SN, Buchter S, Rasmussen M,
et al. Oxygen delivery through nasal cannulae to preterm in-
fants: can practice be improved? Pediatrics 2005;116:857e61.
19. Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am J
Respir Crit Care Med 2001;163:1723e9.

Please cite this article in press as: Yang C-Y, et al., A randomized pilot study comparing the role of PEEP, O2 flow, and high-flow air for
weaning of ventilatory support in very low birth weight infants, Pediatrics and Neonatology (2017), http://dx.doi.org/10.1016/
j.pedneo.2017.02.005