Eur J Pediatr (2016) 175:1343–1351

DOI 10.1007/s00431-016-2770-2

ORIGINAL ARTICLE

Effect of the I/E ratio on CO2 removal during high-frequency

oscillatory ventilation with volume guarantee in a neonatal
animal model of RDS
Manuel Sánchez-Luna 1 & Noelia González-Pacheco 1 & Martín Santos 2 & Ángel Blanco 3 &
Cristina Orden 2 & Jaques Belik 4 & Francisco J. Tendillo 2

Received: 11 May 2016 / Revised: 4 August 2016 / Accepted: 24 August 2016 / Published online: 5 September 2016
# Springer-Verlag Berlin Heidelberg 2016

Abstract The objective of this study was to analyze the effect fore and after a bronchoalveolar lavage (BAL). The effect of
of I/E ratio on carbon dioxide (CO2) elimination during high- an I/E ratio of 1:1 and 1:2 with (VG-ON) and without VG
frequency oscillatory ventilation (HFOV) combined with vol- (VG-OFF) on PaCO2, as well as delta and mean airway pres-
ume guarantee (VG). Five 2-day-old piglets were studied be- sures at the airway opening (ΔPhf-ao, mPaw-ao) and at the
tracheal level (ΔPhf-t, mPaw-t) were evaluated at frequencies
of 5, 8, 11, and 14 Hz. With the VG-ON, PaCO2 was signif-
icant lower with the I/E ratio of 1:2 at 5 Hz compared with the
Communicated by Patrick Van Reempts
1:1. mPaw-t was higher than mPaw-ao, with 1:1 I/E ratio, and
on VG-ON, this difference was statistically significant.
* Manuel Sánchez-Luna
msluna@salud.madrid.org
Conclusion: BIn this animal study and with this ventilator,
the I/E ratio of 1:1 compared to 1:2 in HFOV and VG-ON did
Noelia González-Pacheco not produce a higher CO2 lavage as when HFOV was used
ngpacheco@salud.madrid.org without the VG modality. Even more, a lower PaCO2 was
Martín Santos
found when using the lower frequency and 1:2 ratio compared
martin.santos@salud.madrid.org to 1:1. So in contrast to non-VG HFOV mode, using a fixed
tidal volume, no significant changes on CO2 elimination are
Ángel Blanco
angelblanco30@gmail.com
observed during HFOV when the I/E ratios of 1:1 and 1:2 are
compared at different frequencies.^
Cristina Orden
cristinaordenq@gmail.com

Jaques Belik
What is Known:
jaques.belik@sickkids.ca
•The tidal volume on HFOV is determinant in CO2 removal, and this is
Francisco J. Tendillo generated by delta pressure and the length of the inspiratory time.
franciscojavier.tendillo@salud.madrid.org What is New:
•HFOV combined with VG, an I/E ratio of 1:2 is more effective to remove
1
CO2, and this is not related to the tidal volume.
Neonatology Division, Instituto de Investigación Sanitaria Gregorio
Marañón, Hospital General Universitario Gregorio Marañón,
University Complutense of Madrid, Madrid, Spain
2
Keywords High-frequency ventilation . Carbon dioxide .
Medical and Surgical Research Unit, Instituto de Investigación
Sanitatia Puerta de Hierro-Majadahonda, Hospital Universitario
Neonatal respiratory distress syndrome . Prematurity . Lung
Puerta de Hierro-Majadahonda, Madrid, Spain injury
3
Anaesthesia and Critical Care Department, Instituto de Investigación
Sanitatia Puerta de Hierro-Majadahonda, Hospital Universitario
Puerta de Hierro-Majadahonda, Madrid, Spain Abbreviations
4
Division of Neonatology, Department of Pediatrics, Hospital for Sick BAL Bronchoalveolar lavage
Children, University of Toronto, Toronto, ON, Canada Cdyn Dynamic compliance
1344 Eur J Pediatr (2016) 175:1343–1351

CMV Conventional mechanical ventilation determinant factor by ultimately affecting the tidal volume
CO2 Carbon dioxide delivered by the ventilator [9].
DCO2 carbon dioxide diffusion coefficient Aside from the tidal volume, CO2 removal is also depen-
FiO2 Inspired oxygen fraction dent on the HFOV frequency (fR) via an independent effect on
fR Frequency the distribution of the gas within the airways [3, 11, 15, 28].
HFOV High-frequency oscillatory ventilation Most often, the frequency used and recommended by clini-
mPaw- Mean airway pressure cians is 10 Hz [6], as in most ventilators, tidal volume de-
ao at the airway opening creases when more than 10 Hz is used [9].
mPaw-t Mean airway pressure at the trachea The possibility of using various I/E ratios during HFOV
mPaw Mean airway pressure has been balanced against a more or less efficacy on CO2
PaCO2 Partial carbon dioxide pressure removal during ventilation, although it has been demonstrated
PaO2 Partial arterial oxygen pressure that an I/E ratio of 1:1 results in better pressure transmission to
PEEP Positive end-expiratory pressure the alveoli [21]. Although, the longer the inspiratory time
SaO2 Arterial oxyhemoglobin saturation allows for a higher tidal volume resulting in more effective
SD Standard deviation ventilation and lower risk of gas trapping under CMV [7], the
VG Volume guarantee use of an I/E ratio of 1:2 was frequently reported in clinical
VT Tidal volume studies HFOV [6] to give enough time for gas empting during
VThf High-frequency expired tidal volume exhalation.
ΔPhf-ao High-frequency delta pressure at the airway In the newer high-frequency ventilators generation, it is
opening possible to maintain VThf constant by selecting the volume
ΔPhf-t High-frequency delta pressure at the trachea guarantee modality (VG). In this mode, the ventilator will
ΔPhf High-frequency delta pressure modify ΔPhf to maintain VThf over a certain range of values.
Thus, VG enables evaluate the independent contribution of
VThf and the fR on CO2 removal [20, 26].
Still unanswered, but now possible to evaluate, is the inde-
Introduction pendent effect of I/E ratio changes on CO2 clearance in neo-
nates. Therefore, the aim of this study was to evaluate the
As compared with other ventilatory modes, high-frequency independent effect of I/E ratio changes under HFOV com-
oscillation (HFOV), facilitates gas exchange using tidal vol- bined with VG on CO2 elimination in a neonatal animal mod-
umes lower than the volume of the conducting airways at el. We hypothesize that, in contrast to non-VG mode, using a
supra-physiological frequencies [2, 30]. In neonatal animal fixed VThf, no significant changes on CO2 elimination are
models, HFOV was shown to have a protective effect, when observed during HFOV when the I/E ratios of 1:1 and 1:2
compared to conventional ventilation [31] and thus proposed are compared at different frequencies.
as superior to conventional mechanical ventilation (CMV) in
chronic lung disease of prematurity prevention [5]. Yet, recent
data indicate that it has a no clear benefit over conventional Materials and methods
mechanical ventilation, as an elective form of respiratory sup-
port in neonates with severe respiratory distress [6]. The extent Five healthy 2-day-old Landrace-large white piglets with a
to which this lack of superiority reflects the fact that clinical mean (±standard deviation (SD)) body weight of
comparative ventilation trials are difficult to interpret due to 2.97 ± 0.34 kg were used. The animals were handled accord-
the different devices and HFOV modalities used is difficult to ing to the European and national regulations for protection of
ascertain. experimental animals (2010/63/UE and RD 53/2013), and the
Multiple mechanisms are involved in gas exchange during study was approved by the institutional ethic committee for
HFOV. Among them, direct penetration of fresh gas into the animal research.
alveolar space might be responsible for the improved gas ex- General anesthesia was induced via facemask with
change efficiency [15, 16], generating most of the time by 8 % sevoflurane in oxygen. A 24-gauge polyethylene
variations on delta pressure (ΔPhf) [3], which is known to be catheter was placed into the ventrolateral auricular vein
close to the airway dead space [15]. High-frequency ventila- for continuous infusion of lactated Ringer’s solution
tion has been demonstrated to be more effective than conven- (10 mL/kg/h) and the administration of drugs. When
tional ventilation for carbon dioxide (CO2) removal [12], and an adequate level of anesthesia was achieved, a trache-
this is directly proportional to the tidal volume generated dur- ostomy was performed and a tracheal tube (3 mm inner
ing high-frequency ventilation (VThf) raised to the power of 2 diameter) was placed and sealed with a ribbon wrapped
[25]. The severity of the lung disease is an important around the trachea to prevent any leak. Anesthesia was
Eur J Pediatr (2016) 175:1343–1351
maintained with an intravenous constant infusion of
propofol as needed, and intraoperative analgesia was
achieved with intravenous morphine chlorohydrate
(1 mg/kg).
The ventilator (Babylog VN500, Dräger, Lübeck,
Germany) was connected to the tracheal tube and adjusted in
conventional mechanical ventilation mode with the following
settings: frequency of 30 breaths/min, inspired oxygen frac-
tion (FiO2) of 0.4, positive end-expiratory pressure (PEEP) of
5 cm H2O, and a tidal volume (VT) of 8 mL/kg. If necessary,
the ventilator settings were minimally adjusted to achieve a
target arterial partial pressure of carbon dioxide (PaCO2) of
25–45 mmHg. A 20-gauge polyethylene catheter was ad-
vanced to the carina, between the tracheal tube and the trache-
al wall, to allow for continuous measurement of mean and
ΔPhf tracheal airway pressure (mPaw-t, ΔPhf-t). Mean airway
pressure at the airway opening (mPaw-ao) and ΔPhf at the
airway opening (ΔPhf-ao) were also continuously measured
from the proximal end of the endotracheal tube and close to
the ventilator circuit end of the endotracheal tube. These air-
way pressures were recorded via a calibrated pressure trans-
ducer (referred to a zero level) connected to a monitor (Infinity
Delta XL, Dräger, Lübeck, Germany).
Self-adhering patches were applied to the skin for ECG and
heart rate recording. Pulse oximetry values were continuously
recorded by placing a pulse oximeter on the paw. The rectal
temperature was monitored and maintained between 36 and
38 °C by means of a total temperature management system.
Via surgical cut down, a 20-gauge polyethylene cath-
eter was inserted into right carotid artery. This access
allowed for arterial blood sampling and continuous
blood pressure measurement via a calibrated pressure
transducer. Temperature, pulse oximetry, heart rate, and
arterial blood pressure (referred to a zero level) were
continuously recorded on a cardiovascular monitor
(PM8060 Vitara, Dräger, Lübeck, Germany). Arterial
blood samples were withdrawn anaerobically and imme-
diately analyzed for pH, partial pressure of oxygen
(PaO2), PaCO2, and arterial oxyhemoglobin saturation
(SaO 2 ) (IL 1306 pH/Blood GasAnalyzer, Allied
Instrumentation).
After 30 min of instrumentation and stabilization of the
target blood gases, the ventilator was switched to HFOV and
adjusted to a mean airway pressure (mPaw) of 10.2 cm H2O, a
fR of 5 Hz, and an initial ΔPhf of 25–30 cm H2O for a new
target PaCO2 of 25–30 mmHg. Animals were studied under
all the ventilatory modes tested and randomly assigned to
begin with VG-ON or VG-OFF and to an I/E ratio of 1:1 or
1:2. After being assigned to a specific ventilation method, the
fR was increased from 5 to 8, 11, and 14 Hz for a period of
10 min each and variables were intermittently monitored un-
der each ventilatory mode before bronchoalveolar lavage (pre-
BAL). The VThf was set VG mode to reach a target PaCO2 of
1345
25–30 mmHg. After all the ventilatory modalities were tested,
the animals were subjected to BAL, as previously described
[26], in order to reduce the lung compliance. Lung recruitment
was described elsewhere (ref 26), briefly to ensure adequate
alveolar reexpansion; a sustained high-pressure recruitment
maneuver using repeated 40 cm H2O of pressure during 30 s
was undertaken following BAL. A period of 30 min was de-
termined to establish the surfactant-depleted situation, while
the ventilator was set on HFOV with a VG of 3 mL/kg, and
then arterial blood tests were done.
After the BAL and lung recruitment, measurement of the
lung dynamic compliance (Cdyn) using the same ventilatory
parameters as in pre-BAL conventional ventilation was done,
then ventilation was switched to HFOV and the settings were
adjusted to an mPaw of 14.7 ± 1.6 cm H2O, fR of 5 Hz, and an
initial ΔPhf of 25–30 cm H2O for a target PaCO2 of 25–
30 mmHg The same ventilation protocol as before the BAL
was repeated in all the animals under the low compliance
condition (post-BAL). Also since after the initial drop in com-
pliance, there was a trend toward improved compliance after
the lung lavage and recruitment to prevent any resulting bias
regarding the effect of changes in the frequency at different I/E
ratio and with or without the VG strategy, there was an alter-
nation between different animals. At the end of the entire
study, the Cdyn measurement was repeated under the same
setting on conventional ventilation.
HFOV was delivered with a Draeger VN500 ventila-
tor (VN500, Draeger, Lübeck, Germany). The Babylog
VN500 generates a sinusoidal pressure signal around a
set mean airway pressure and has an active
inspiration and active expiration to allow for this. The
VG mode is a volume-targeted ventilation where the
microprocessor compares the VThf of the previous breath,
using leak compensated VThf, and adjusts delta pressure
up or down to achieve the set VThf. A hot wire anemo-
meter is used to monitor volume and has been shown to
provide reliable measurements of tidal volume at the
airway opening [27] and also is more accurate than the
use of a pneumotachograph, as there was greater linearity
of response across the range of frequencies tested [4].
Based on the results observed of previous studies [26],
a sample size of five animals per group was considered
suitable for detecting a ventilation mode-dependent PaCO2
difference ≥15 mmHg with a power of 90 % in and two-
tailed analysis mode. Statistical analysis of data was per-
formed using SPSS 20.0 software program (IBM SPSS
Statistics, Chicago, IL, USA). Following confirmation of
normal distribution (Shapiro-Wilk test), physiological data
were analyzed using a two-way analysis of variance
(ANOVA) for repeated measures followed by Tukey’s test
for multiple comparisons. These data are presented as
mean ± standard deviation. Statistical significance was ac-
cepted when P < 0.05.
1346
Results
Initial baseline Cdyn was measured at a pre-set conventional
ventilation parameters and repeated post stabilization period,
pre- and post-BAL, as well as at the end of the study. As
shown in Fig. 1, BAL significantly reduced Cdyn and the
lower compliance remained unchanged until the end of the
experiment.
Effect on CO2 lavage (Table 1)
VG-OFF modality
During pre-BAL and post-BAL, a statistically significant
higher PaCO2 was found when using 1:2 I/E ratio compared
to 1:1. At any frequency, but 14 Hz in the pre-BAL, no other
statistically significant changes were found.
VG-ON modality
No statistically significant effect when comparing 1:1 I/E ratio
to 1:2 was found in this modality but a lower PaCO2 when
using 1:2 I/E ratio at 5 Hz.
When compared to the VG-OFF modality, the use of VG
produced a statistically significant lower PaCO2 at almost all
the frequencies, pre- and post-BAL when using 1:2 I/E ratio,
demonstrating a more effective CO2 lavage with the 1:2 I/E
ratio in the VG-ON modality compared to the VG-OFF
modality.
Effect on the VT (Table 1)
Compared to the VG-OFF modality, with the use of VG-ON, a
more stable and statistically significant no variation of the VT
was observed over all the frequencies studied demonstrating a
large effect of the frequency on the VT in the VG-OFF mo-
dality and no effect of the modification of the frequency on
VT in the VG-ON modality.
6.0
5.0
Cdyn (mL/cmH 2 O)
4.0
3.0
* PaCO2. During HFOV, CO2 removal is mostly related to the
2.0 * VThf [16, 25], which is normally generated by variations on
1.0
0.0
Start Pre-BAL Post-BAL Final
Fig. 1 Dynamic compliance (Cdyn) at the basal, pre-BAL, post-BAL,
and final. *P < 0.05 compared with pre-BAL
Eur J Pediatr (2016) 175:1343–1351
Effect on ΔP (Table 2)
At any frequency and compared to the VG-OFF modality, pre-
and post-BAL, ΔPhf-ao was statistically significant higher
when using the VG-ON modality, demonstrating the higher
ΔP used by the ventilator to maintain the VT set when the
frequency was increased. Also a higher, ΔPhf-ao was used
in the 1:2 I/E ratio in the VG-ON modality compared to the
VG-OFF. No significant transmission distally to the tracheal
level (ΔPhf-t) was found in both modalities at any frequency.
The ventilator was not able to maintain the VT set at 14 Hz
pre- and post-BAL.
Effect of the I/E ratio on mPaw
mPaw set at the ventilator in the pre-BAL situation was
10.2 ± 0.0 and 14.7 ± 1.6 cm H2O in the post-BAL situation
in all the animals. With the VG-OFF, for the pre- and the post-
BAL conditions, mPaw-t was higher than mPaw-ao, mostly
when the 1:1 I/E ratio was used (Table 2). Both pressures
remained stable through the experiment with the VG-OFF as
the set on the ventilator fix the pressure and not the VT as in
the VG-ON.
In contrast, with the VG-ON, this difference was more
evident, and present all the time the mPaw-t was higher than
the mPaw-ao, when the 1:1 I/E ratio was used compared to 1:2
(Table 2).
Discussion
We have demonstrated that the I/E ratio alters ventilation ef-
fectiveness under different frequencies, depending on whether
the VG mode is used. The use of an I/E ratio of 1:2 is more
effective reducing the PaCO2 at any lung condition, when
compared with the 1:1 ratio setting. With the VG-OFF modal-
ity, in the pre-BAL situation, the 1:2 I/E ratio compared to 1:1
decreases less the PaCO2 at frequencies of 5, 8, and 11 Hz, not
at 14 Hz (Table 1) and a similar effect was found in the
post-BAL situation at any frequency.
We speculate that these results can be partially explained
due to a higher, but not statistically significant VThf generated
with a longer inspiratory time, mostly at the lower frequencies.
Under these conditions, the I/E ratio of 1:1 generates a higher
VThf and it is probably more effective in decreasing the
ΔPhf [3]. The lower the frequency, the longer inspiratory time
is expected and at 1:1, a higher VThf is generated due to a
higher delta pressure and then a lower PaCO2, compared to
1:2 ratio. As ΔPhf-ao was similar at any frequency with 1:1 or
1:2 I/E ratio, the CO2 clearance difference is possibly related
to the longer inspiratory time achieved with 1:1 I/E ratio thus
Eur J Pediatr (2016) 175:1343–1351 1347

Table 1 Partial pressure of oxygen (PaO2), partial pressure of carbon dioxide (PaCO2), delta pressure at the airway opening (ΔPhf-ao), delta pressure
at the trachea (ΔPhf-t), tidal volume on HFOV (VThf), and carbon dioxide diffusion coefficient (DCO2) by frequency in Hz

VG-OFF pre-BAL VG-ON pre-BAL VG-OFF post-BAL VG-ON post-BAL

I/E ratio 1:1 1:2 1:1 1:2 1:1 1:2 1:1 1:2

PaO2 (mmHg)
5 Hz 172 ± 48 151 ± 33 148 ± 26 157 ± 30 104 ± 33 107 ± 32 104 ± 30 115 ± 30
8 Hz 146 ± 31 127 ± 31 156 ± 31 161 ± 34 79 ± 20 87 ± 14 115 ± 35 120 ± 27
11 Hz 110 ± 30 109 ± 22 169 ± 32** 161 ± 41 67 ± 10 68 ± 8 130 ± 33** 105 ± 41
14 Hz 92 ± 27 98 ± 21 167 ± 37** 57 ± 10 60 ± 5 125 ± 52**
PaCO2 (mmHg)
5 Hz 28.8 ± 3.4 37.2 ± 5.8* 49.2 ± 3.4** 35.8 ± 2.2* 31.2 ± 5.1 44.0 ± 7,5* 49.6 ± 5.1** 39.4 ± 3.8*
8 Hz 41.4 ± 4.6 52.2 ± 6.6* 39.6 ± 3.6 32.8 ± 2.6** 44.4 ± 9.8 57.8 ± 7.0* 40.0 ± 4.4 33.2 ± 6.9**
11 Hz 52.6 ± 7.5 62.8 ± 5.2* 29.0 ± 2.5** 24.4 ± 2.3** 60.8 ± 4.1 71.2 ± 7.7* 32.4 ± 4.4** 27.6 ± 5.4**
14 Hz 64.0 ± 9.6 69.2 ± 3.0 25.8 ± 1.5** 70.4 ± 10.9 80.0 ± 11.4* 31.0 ± 6.2**
ΔPhf-ao (cm H2O)
5 Hz 19.0 ± 1.0 16.9 ± 0.7 12.5 ± 2.4** 14.1 ± 1.6 20.4 ± 1.7 17.7 ± 1.4 15.5 ± 2.3** 16.6 ± 2.4
8 Hz 19.6 ± 1.6 16.9 ± 1.6 24.8 ± 3.2** 27.2 ± 3.5** 20.7 ± 1.8 17.7 ± 1.4 30.5 ± 4.1** 31.6 ± 4.4**
11 Hz 19.9 ± 1.2 18.0 ± 1.8 48.4 ± 2.8** 49.8 ± 4.8** 20.4 ± 1.0 18.5 ± 0.7 56.6 ± 5.2** 55.8 ± 5.0**
14 Hz 18.8 ± 1.1 18.0 ± 1.1 50.0 ± 4.1** 20.7 ± 1.1 19.0 ± 1.4 57.1 ± 3.6**
ΔPhf-t (cm H2O)
5 Hz 6.0 ± 1.2 4.9 ± 1.2 5.2 ± 1.1 4.6 ± 1.6 9.2 ± 2.6 7.6 ± 2.7 8.4 ± 2.0 8.4 ± 2.2
8 Hz 4.6 ± 1.2 4.1 ± 1.0 4.9 ± 1.2 4.6 ± 1.8 5.4 ± 1.0 5.7 ± 1.5 7.3 ± 2.5 7.6 ± 1.6
11 Hz 3.5 ± 1.2 3.5 ± 0.7 5.4 ± 1.0 5.4 ± 1.4 4.4 ± 1.1 4.1 ± 1.0 7.9 ± 2.0** 6.8 ± 1.4**
14 Hz 3.0 ± 0.6 3.5 ± 0.7 4.8 ± 0.8 4.1 ± 0.0 3.8 ± 1.1 5.9 ± 2.1
VThf (mL)
5 Hz 16.2 ± 1.7 13.9 ± 1.3 12.3 ± 1.3** 12.3 ± 1.3 14.3 ± 1.7 12.9 ± 1.2 12.3 ± 1.3 12.3 ± 1.3
8 Hz 10.3 ± 1.0 9.1 ± 0.9 12.3 ± 1.3 12.3 ± 1.3** 9.0 ± 1.1 8.6 ± 0.9 12.3 ± 1.3** 12.3 ± 1.3**
11 Hz 7.0 ± 0.8 6.6 ± 0.7 12.0 ± 0.8** 11.5 ± 0.4** 6.3 ± 0.8 6.3 ± 0.7 11.8 ± 0.6** 11.3 ± 0.4**
14 Hz 5.4 ± 0.6 5.4 ± 0.6 9.9 ± 0.2** 4.9 ± 0.7 5.2 ± 0.6 9.7 ± 0.3**
DCO2 (mL2/s)
5 Hz 1342 ± 277 984 ± 183 769 ± 172** 767 ± 171 1042 ± 233 849 ± 157 768 ± 171 750 ± 176
8 Hz 859 ± 169 670 ± 131 1214 ± 270 1216 ± 272** 655 ± 151 600 ± 121 1215 ± 270** 1213 ± 274**
1 Hz 556 ± 120 495 ± 105 1633 ± 221** 1496 ± 105** 447 ± 160 456 ± 99 1577 ± 144** 1440 ± 106**
14 Hz 412 ± 85 405 ± 80 1345 ± 57** 346 ± 98 390 ± 95 1307 ± 74**

Data as mean ± SD
*P < 0.05 (ANOVA) compared to 1:1 ratio of each group; **P < 0.05 (ANOVA) compared to the same I/E ratio of VG-OFF groups

allowing for a higher VThf. This VThf depends mainly on the
time constant of the respiratory system. With higher time con-
stant, the use of longer time affects the delivery of higher VT;
with lower time constant, the use of longer time affects less the
delivery of higher VT.
Yet, 1:2 ratio-related effect is negligible at the higher fre-
quencies where the VThf was similar mostly in the post-BAL
situation probably due the lower compliance situation [9].
The effect of the tidal volume as a function of the inspira-
tory time is related to the time constant of the respiratory
system, as compliance decreases the effect of the VT that is
less significant.
ΔPhf transmission from the airway opening to the trachea
(difference between ΔPhf-ao and ΔPhf-t) was similar when
compared 1:1 and 1:2 ratio in the pre-BAL and post-BAL
(Table 1). With both I/E ratios there was a drop in ΔPhf-t as
the frequency was increased pre- and post-BAL. This effect
on the delta pressure is also a well-known phenomenon of the
HFOV, but there was not any decrease in the mPaw-t. A drop
in the PaO2 was found in the VG-OFF with 1:1 and 1:2 I/E
ratio with the increase in the frequency, and this also can be
partially related to the decrease in ΔPhf-t, although mPaw at
the ventilator was constant. Due to a lower VT at the higher
frequency in the VG-OFF, certain derecruitment could be
expected.
The decrease in ΔPhf-t with the increase in the frequency
was more pronounced in the post-BAL situation where the
lung compliance is reduced, but this value was higher in the
low compliance situation compared with the high compliance
at any frequency, and also the difference between the ΔPhf-ao
and the ΔPhf-t was less at the lower frequency, which means a
better transmission of ΔPhf in the low compliance compared
with the physiological lung condition [24]. Probably as the
frequency increases over 10 Hz, ΔPhf transmission decreases.
An alternative explanation for the higher VThf and ΔPhf-t
for a lower PaCO2 when the 1:1 I/E ratio was used instead of
the 1:2 mostly at the higher frequencies is the better pressure
transmission to the alveoli [21]. In the VG-ON modality, it has
been previously demonstrated that the PaCO2 decreases when
the frequency increases [26]. Also, we have shown that a
1348 Eur J Pediatr (2016) 175:1343–1351

Table 2 Mean airway pressure at the airway opening (mPaw-ao) and mean airway pressure at the trachea (mPaw-t) by frequency in Hz

VG-OFF pre-BAL VG-ON pre-BAL VG-OFF post-BAL VG-ON post-BAL

I/E ratio 1:1 1:2 1:1 1:2 1:1 1:2 1:1 1:2

mPaw-ao (cm H2O)

5 Hz 8.2 ± 0.0 8.2 ± 0.0 9.2 ± 0.6 8.7 ± 0.7 13.1 ± 1.6 13.1 ± 1.6 13.1 ± 1.8 13.1 ± 1.6
8 Hz 8.2 ± 0.0 8.2 ± 0.0 7.6 ± 0.7 6.8 ± 1.0** 13.1 ± 1.6 12.8 ± 1.2 12.0 ± 1.8 11.2 ± 1.5
11 Hz 8.2 ± 0.0 8.2 ± 0.0 4.9 ± 1.6** 3.5 ± 0.7** 13.1 ± 1.6 12.8 ± 1.2 9.5 ± 1.9** 8.2 ± 1.9**
14 Hz 8.2 ± 0.0 8.2 ± 0.0 2.4 ± 0.7** 13.1 ± 1.6 12.5 ± 1.1 8.2 ± 0.0**
mPaw-t (cm H2O)
5 Hz 9.5 ± 0.0 8.4 ± 0.6 9.5 ± 0.0 9.2 ± 0.6 14.4 ± 1.6 13.1 ± 1.6 14.4 ± 1.6 13.1 ± 1.6
8 Hz 9.5 ± 0.0 8.4 ± 0.6 9.5 ± 0.0 7.6 ± 0.7 14.4 ± 1.6 13.1 ± 1.6 13.9 ± 2.2 11.7 ± 1.6
11 Hz 9.5 ± 0.0 8.7 ± 0.7 9.0 ± 0.7 5.2 ± 0.6*,** 14.4 ± 1.6 13.1 ± 1.6 13.9 ± 2.2 9.0 ± 1.2*,**
14 Hz 9.5 ± 0.0 8.7 ± 0.7 6.5 ± 0.7** 14.4 ± 1.6 13.1 ± 1.6 12.0 ± 0.0

Data as mean ± SD
*P < 0.05 (ANOVA) compared to 1:1 ratio of each group. **P < 0.05 (ANOVA) compared to the same I/E ratio of VG-OFF groups

lower PaCO2 occurs with the 1:2 I/E ratio, when compared to
the 1:1 I/E ratio at 5 Hz.
The PaCO2 was lower when the inspiratory time was
shorter, even being similar the VThf and this was statistically
significant at 5 Hz compared to higher frequency, where no
significant differences were found. No simple explanation can
be found for this effect.
As during HFOV, other mechanisms (direct alveolar venti-
lation through convection, convection by Bhigh-frequent^
pendelluft, convective dispersion due to asymmetric velocity
profiles, Taylor dispersion (transverse diffusion), and molec-
ular diffusion) are considered important factors regulating
CO2 removal, it is possible that other mechanisms not directly
related to the modification of the inspiratory time could be
involved. Also a higher risk of gas trapping when using a
longer inspiratory time at these lower frequencies can explain
this result. A higher mPaw-t than the mPaw-ao was found
mostly when the 1:1 I/E ratio was used.
It is possible that a longer inspiratory time can cause sub-
stantial gas trapping resulting in a dynamic increase of the
functional residual capacity [32]. If dynamic gas trapping is
involved [32] when a fixed tidal volume is used, as in the VG-
ON modality, then the higher the frequency, the lower the
effectiveness in decreasing PaCO2. But, although a higher
tracheal than at the airway opening mean airway pressure
was found, we documented in this study a decrease in the
PaCO2 as the frequency increased, but more with the 1:2 ratio.
This can be possibly explained by the fact that the expiratory
flow is aided by actively suction of the ventilator, in contrast
with a jet ventilator where the expiratory flow depends in the
elastic recoil of the lungs [24]. The Babylog VN500 does have
an active inspiration and expiration due to a potent Venturi
system to generate oscillatory pressure amplitudes and
facilitate active expiration, so a minimal dynamic gas trapping
effect is expected. A fall in the mPaw-ao was seen in the VG-
ON as frequency increases, probably due to the lower trans-
mission of the pressure as the frequency increases. The lesser
drop in mPaw-t can be clearly caused by some gas trapping.
As ΔPhf-ao increased with the VG-ON modality as the
frequency increases to maintain a constant tidal volume, there
is a decrease in the mPaw-t. Such results were also described
in another study mostly when an I/E ratio of 1:2 was used [18].
Using a low-mass alveolar capsule, it has been reported that at
frequencies of 10–20 Hz, the inspiratory time of 0.3 is unlike-
ly to cause any lung hyperinflation and alveolar pressure is
less than proximal pressure, but it is largely unpredictable with
increasing inspiratory time to 0.5 where the alveolar and tra-
cheal pressure increases, and this occurs because a decrease in
the expiratory volume compared to the inspiratory volume
[13].
Probably due to a more stable VT, a statistically significant
higher PaO2 was observed when 1:1 I/E ratio was used during
VG-ON compared with VG-OFF, this effect was also seen
with the 1:2 but with no statistically significance. A constant
VT can prevent derecruitment compared to VG-OFF.
As shown by others in immature animal models [8, 14, 17],
small tidal volumes and lung volume maintenance prevent
ventilation-induced lung injury in the preterm lung, and
HFOV has been proposed an ideal mode of ventilation for
preterm infants to prevent damage of the immature lung. For
this reason, it has been recommended that the highest possible
frequency be used to decrease alveolar trauma [10]. When
using the VG-ON modality, it is possible to maintain similar
CO2 washout while decreasing VThf and increasing the fre-
quency [20], but if a very high frequency is to be used to
decrease as much as possible the tidal volume while
Eur J Pediatr (2016) 175:1343–1351
maintaining similar CO2 washout, the effect of the I/E ratio
has to be taken in mind.
If the purpose of HFOV is to use the highest possible fre-
quency with the lower VThf to decrease trauma to the most
immature lungs [20], then the effect of the 1:2 I/E ratio chosen
could be more effective to decrease PaCO2 at any frequency,
and although some gas trapping can occur at the higher fre-
quencies, the 1:1 I/E ratio will have to be used as the device is
not able to maintain the VThf set at the higher frequencies. I/E
ratios of 1:1 and 1:2 were chosen in this study as they are the
most frequently used in clinical studies [6], and the low com-
pliance situation compared to the physiological lung condition
was used as the magnitude of pressure amplitudes transmitted
to the lung in the newborn increases exponentially with de-
creasing compliance [22, 23]. Also the stiffness of the lung
can decrease the tidal volume transmitted [9]. Previous studies
showed that compliant alveoli are effectively spared from ex-
cessive oscillatory pressures with the larger alveolar pressure
swings being directed to the more poorly compliant compart-
ments [22].
I/E ratio of 1:2 has been proposed to prevent gas trapping
during ventilation. The active expiratory mechanism of most
of the HFOV devices can prevent this and most of the trials
used a fixed ratio of 1:2. Only a few studies used a 1:1 ratio.
Moriette et al. compared HFOV with CV with the OHF1
piston oscillator (Dufour, France) at a frequency of 15 Hz
with an I/E ratio of 1:1 [19]. No clear relationship of the
higher risk of severe intraventricular hemorrhage in the
HFOV group compared to the conventional ventilation group
was done due to a 1:1 I/E ratio of instead of the most fre-
quently used 1:2. Although some laboratory data demonstrat-
ed a higher volume generated by using an I/E ratio of 1:1
compared to 1:2 [7], to our knowledge, no studies have been
done comparing the effect of using different I/E ratios on CO2
lavage when the tidal volume is fixed using the VG strategy.
When VG-OFF modality is used, then the use of a 1:1 I/E
ratio produced a higher CO2 removal compared to 1:2 at all
the frequencies, before and after the BAL, but is not seen with
the VG-ON modality, where although only statistically sig-
nificant at the lower frequency of 5 Hz, the effect of a higher
CO2 removal at I/E ratio of 1:2 instead of 1:1 was found and
can have some clinical implication. As VT is now fixed in the
VG modality, a lower PCO2 cannot be related to modifica-
tions in the VT but to a shorter inspiratory time and probably
to less gas trapping at the lower frequency of 5 Hz, so a more
effective CO2 removal is achieved with a lower risk for the
patient lung. At the higher frequency, this effect is not signif-
icant so this mechanism can be overlap by some dynamic
gastrapping. At 14 Hz, this ventilator was not able to maintain
the fixed VT with the 1:2 I/E ratio, probably due to a very
short inspiratory time, so 1:2 ratio would be recommended
for clinical use unless a too short inspiratory time prevents
reaching the VT set.
1349
Very high frequency can limit the inspiratory time to pro-
duce the VT set in the VG-ON modality. When no VG is used,
then increasing the frequency up to 14 Hz produces a drop in
ΔPhf-ao and of the VT and an increase in the PaCO 2
(Table 1), and because of that when using very high frequen-
cies, an increase in ΔPhf-ao is needed by the user to prevent
this drop in VT. When the VG modality is ON, then the ven-
tilator adjust automatically ΔPhf-ao to maintain VT, but at
14 Hz of frequency when the 1:2 ratio was used in this animal
model, the ventilator was not able to maintain the set VT, even
by increasing ΔPhf-ao. So at 14 Hz, the very short inspiratory
time limits the ventilator to maintain the VT in the set limits,
and this is probably important if a very high frequency is used
during clinical application. A higher risk of gas trapping then
is possible, but with an important drop in the transmission of
ΔPhf-ao to ΔPhf-t. Even using the 1:1 I/E ratio, VT was not
maintained by the ventilator in the pre-BAL and post-BAL
conditions at 14 Hz and this can be an important challenge
of this modality when large VT is set at very high frequencies.
Although the use of HFOV in clinical trials did not dem-
onstrate a protective effect to prevent lung damage, probably
due to large differences in design and ventilators [6], still
HFOV combined with VG has to demonstrate any clinical
benefit over the use of standard non-VG HFOV, and random-
ized clinical trials are needed to verify any benefits.
In summary, the efficacy on CO2 removal of different I/E
ratios in the VG-OFF modality was mostly related to the
changes in the VThf. When using the VG-ON, these changes
disappear, being more effective with a shorter inspiratory
time, a mechanism not related to the VThf generated by the
ventilator. With the VG-ON modality using 1:2 I/E ratio, a
lower PaCO2 was found mostly at the lower frequencies, mak-
ing this effect less important when the frequency was
increased.
Study limitations
In this study, no alveolar pressure was measured. Mean airway
pressure at the airway opening is a poor indicator of the pa-
renchymal pressure and can underestimate the alveolar pres-
sure at any frequency and tidal volume [29]. Yet, during
HFOV in the clinical setting, the airway pressure is distally
overdamped and the carinal distending pressure swings can be
expected to always be equal or greater than the pressure
swings distending the average distal alveoli [32]. Also, a high
regional heterogeneity was noted when alveolar pressure was
measured during HFOV due to the regional flow distribution,
the airway branching angle [1]. Alveolar pressure swings and
tidal volumes fell mostly due to pressure drop along the tra-
cheal tube with increasing the frequency [10]. Also the use of
a particular ventilator can prevent traslation to the use of a
diferent HFOV device.
1350
Conclusions
BIn this animal study and with this ventilator, the I/E ratio of
1:1 compared to 1:2 in HFOV and VG-ON did not produce a
higher CO2 lavage as when HFOV was used without the VG
modality. Even more, a lower PaCO2 was found when using
the lower frequency and 1:2 ratio compared to 1:1. So in
contrast to non-VG HFOV mode, using a fixed tidal volume,
no significant changes on CO2 elimination are observed dur-
ing HFOV when the I/E ratios of 1:1 and 1:2 are compared at
different frequencies.^
The results of the current study support the use of 1:2 I/E
with HFOV combined with VG for more effective ventilation,
until 1:1 is needed at very high frequencies with this
ventilator.
Acknowledgments The authors thank Dräger Medical GmbH for pro-
viding the Babylog VN500 for the present study.
Authors’ contributions Manuel Sánchez-Luna: design, development,
data interpretation, statistical analysis, and preparation of manuscript.
Noelia González-Pacheco: design, development, data management,
data interpretation, and preparation of manuscript.
Martín Santos: design, development, data management, statistical
analysis, and preparation of manuscript.
Ángel Blanco: development and data management.
Cristina Orden: development and data management.
Jaques Belik: data interpretation and preparation of manuscript.
Francisco J. Tendillo: design, development, and data management.
Compliance with ethical standards
Conflict of interest Manuel Sánchez-Luna declares receiving advisory
board consulting fees from Dräger. The remaining authors declare that
they have no conflict of interest.
Ethical approval All applicable international, national, and/or institu-
tional guidelines for the care and use of animals were followed. All
procedures performed in studies involving animals were in accordance
with the ethical standards of the institution or practice at which the studies
were conducted. This article does not contain any studies with human
participants performed by any of the authors. The ethical approval num-
ber for the study was CEEA 005/2013.
References
1. Allen JL, Frantz ID 3rd, Fredberg JJ (1987) Heterogeneity of mean
alveolar pressure during high-frequency oscillations. J Appl Physiol
62:223–228
2. Bohn DJ, Miyasaka K, Marchak BE, Thompson WK, Froese AB,
Bryan AC (1980) Ventilation by high-frequency oscillation. J Appl
Physiol Respir Environ Exerc Physiol 48:710–716
3. Boynton BR, Hammond MD, Fredberg JJ, Buckley BG, Villanueva
D, Frantz ID 3rd (1989) Gas exchange in healthy rabbits during
high-frequency oscillatory ventilation. J Appl Physiol 66:1343–
1351
Eur J Pediatr (2016) 175:1343–1351
4. Chan V, Greenough A, Milner AD (1993) The effect of frequency
and mean airway pressure on volume delivery during high-
frequency oscillation. Pediatr Pulmonol 15:183–186
5. Clark RH, Gerstmann DR, Null DM Jr, deLemos RA (1992)
Prospective randomized comparison of high-frequency oscillatory
and conventional ventilation in respiratory distress syndrome.
Pediatrics 89:5–12
6. Cools F, Offringa M, Askie LM (2015) Elective high frequency oscil-
latory ventilation versus conventional ventilation for acute pulmonary
dysfunction in preterm infants. The Cochrane database of systematic
reviews . doi:10.1002/14651858.CD000104.pub43:Cd000104
7. De Luca D, Piastra M, Pietrini D, Conti G (2012) Effect of ampli-
tude and inspiratory time in a bench model of non-invasive HFOV
through nasal prongs. Pediatr Pulmonol 47:1012–1018.
doi:10.1002/ppul.22511
8. deLemos RA, Coalson JJ, deLemos JA, King RJ, Clark RH,
Gerstmann DR (1992) Rescue ventilation with high frequency os-
cillation in premature baboons with hyaline membrane disease.
Pediatr Pulmonol 12:29–36
9. Dimitriou G, Greenough A, Kavvadia V, Laubscher B, Milner AD
(1998) Volume delivery during high frequency oscillation. Arch
Dis Child Fetal Neonatal Ed 78:F148–F150
10. Frantz ID 3rd, Close RH (1985) Alveolar pressure swings during
high frequency ventilation in rabbits. Pediatr Res 19:162–166.
doi:10.1203/00006450-198502000-00002
11. Fredberg JJ, Glass GM, Boynton BR, Frantz ID 3rd (1987) Factors
influencing mechanical performance of neonatal high-frequency
ventilators. J Appl Physiol 62:2485–2490
12. Froese AB, Bryan AC (1987) High frequency ventilation. Am Rev
Respir Dis 135:1363–1374
13. Gerstmann DR, Fouke JM, Winter DC, Taylor AF, deLemos RA
(1990) Proximal, tracheal, and alveolar pressures during high-
frequency oscillatory ventilation in a normal rabbit model. Pediatr
Res 28:367–373. doi:10.1203/00006450-199010000-00013
14. Hamilton PP, Onayemi A, Smyth JA, Gillan JE, Cutz E, Froese AB,
Bryan AC (1983) Comparison of conventional and high-frequency
ventilation: oxygenation and lung pathology. J Appl Physiol Respir
Environ Exerc Physiol 55:131–138
15. Isabey D, Harf A, Chang HK (1984) Alveolar ventilation during
high-frequency oscillation: core dead space concept. J Appl Physiol
Respir Environ Exerc Physiol 56:700–707
16. Jaeger MJ, Kurzweg UH, Banner MJ (1984) Transport of gases in
high-frequency ventilation. Crit Care Med 12:708–710
17. Jobe AH, Kramer BW, Moss TJ, Newnham JP, Ikegami M (2002)
Decreased indicators of lung injury with continuous positive expi-
ratory pressure in preterm lambs. Pediatr Res 52:387–392.
doi:10.1203/00006450-200209000-00014
18. Leipala JA, Sharma A, Lee S, Milner AD, Greenough A (2005) An
in vitro assessment of gas trapping during high frequency oscilla-
tion. Physiol Meas 26:329–336. doi:10.1088/0967-3334/26/3/016
19. Moriette G, Paris-Llado J, Walti H, Escande B, Magny JF,
Cambonie G, Thiriez G, Cantagrel S, Lacaze-Masmonteil T,
Storme L, Blanc T, Liet JM, Andre C, Salanave B, Breart G
(2001) Prospective randomized multicenter comparison of high-
frequency oscillatory ventilation and conventional ventilation in
preterm infants of less than 30 weeks with respiratory distress syn-
drome. Pediatrics 107:363–372
20. Mukerji A, Belik J, Sanchez-Luna M (2014) Bringing back the old:
time to reevaluate the high-frequency ventilation strategy. J
Perinatol 34:464–467. doi:10.1038/jp.2014.39
21. Pillow JJ, Neil H, Wilkinson MH, Ramsden CA (1999) Effect of I/E
ratio on mean alveolar pressure during high-frequency oscillatory
ventilation. J Appl Physiol 87:407–414
22. Pillow JJ, Sly PD, Hantos Z (2004) Monitoring of lung volume
recruitment and derecruitment using oscillatory mechanics during
Eur J Pediatr (2016) 175:1343–1351
high-frequency oscillatory ventilation in the preterm lamb. Pediatr
Crit Care Med 5:172–180
23. Pillow JJ, Sly PD, Hantos Z, Bates JH (2002) Dependence of
intrapulmonary pressure amplitudes on respiratory mechanics dur-
ing high-frequency oscillatory ventilation in preterm lambs. Pediatr
Res 52:538–544. doi:10.1203/00006450-200210000-00013
24. Pillow JJ, Wilkinson MH, Neil HL, Ramsden CA (2001) In vitro
performance characteristics of high-frequency oscillatory ventila-
tors. Am J Respir Crit Care Med 164:1019–1024. doi:10.1164
/ajrccm.164.6.2005008
25. Rossing TH, Slutsky AS, Lehr JL, Drinker PA, Kamm R, Drazen
JM (1981) Tidal volume and frequency dependence of carbon di-
oxide elimination by high-frequency ventilation. N Engl J Med
305:1375–1379. doi:10.1056/nejm198112033052303
26. Sanchez Luna M, Santos Gonzalez M, Tendillo Cortijo F (2013) High-
frequency oscillatory ventilation combined with volume guarantee in a
neonatal animal model of respiratory distress syndrome. Critical care
research and practice 2013:593915. doi:10.1155/2013/593915
27. Scalfaro P, Pillow JJ, Sly PD, Cotting J (2001) Reliable tidal vol-
ume estimates at the airway opening with an infant monitor during
1351
high-frequency oscillatory ventilation. Crit Care Med 29:1925–
1930
28. Schuster DP, Karsch R, Cronin KP (1986) Gas transport during
different modes of high-frequency ventilation. Crit Care Med 14:
5–11
29. Simon BA, Weinmann GG, Mitzner W (1984) Mean airway
pressure and alveolar pressure during high-frequency venti-
lation. J Appl Physiol Respir Environ Exerc Physiol 57:
1069–1078
30. Slutsky AS, Drazen FM, Ingram RH Jr, Kamm RD, Shapiro AH,
Fredberg JJ, Loring SH, Lehr J (1980) Effective pulmonary venti-
lation with small-volume oscillations at high frequency. Science
(New York, NY) 209:609–671
31. Truog WE, Standaert TA, Murphy JH, Woodrum DE, Hodson WA
(1984) Effects of prolonged high-frequency oscillatory ventilation
in premature primates with experimental hyaline membrane dis-
ease. Am Rev Respir Dis 130:76–80
32. Venegas JG, Fredberg JJ (1994) Understanding the pressure cost of
ventilation: why does high-frequency ventilation work? Crit Care
Med 22:S49–S57