Improving synchrony in young infant supported by noninvasive
ventilation for severe bronchiolitis: Yes we can… so we should!
C Milési, MD1, J Baleine, MD1, G
Cambonie, MD, PhD1
1Department of Neonatal Medicine and Pediatric
Intensive Care, Arnaud de Villeneuve Hospital, Montpellier University
Hospital Center, Montpellier, France
Correspondence: Christophe Milési, MD, Department of Neonatal Medicine
and Pediatric Intensive Care Unit, Arnaud de Villeneuve Hospital,
Montpellier University Hospital, 371 Avenue du Doyen Gaston Giraud,
34295 Montpellier Cedex 5, France
Tel: +33 467 336 556, Fax: +33 467 336 228, e-mail:c-milesi@chu-montpellier.fr
Word count: 1391; references: 43
Keywords: asynchrony, bronchiolitis, noninvasive ventilation,
neurally adjusted ventilatory assist, infant
Admission to a pediatric intensive care unit (PICU) is required for
9-14% of infants with acute viral bronchiolitis (AVB) and evolving
respiratory distress.1,2 In this context, AVB
generally presents as severe obstructive lung disease, which causes an
increased load on the respiratory muscles.3,4 As no
pharmaceutical treatment currently in use is able to rapidly reduce
airway obstruction, the management of these patients is focused on
providing respiratory support to reduce respiratory muscle fatigue and
prevent intubation. Noninvasive ventilation (NIV), delivered by
continuous positive airway pressure (CPAP) or bilevel positive airway
pressure (BiPAP), has traditionally been applied and is associated with
reductions in intubation rates, ventilation-associated complications,
and duration and cost of hospitalization.5,6 More
recently, a third device was introduced to administer a heated and
humidified mixture of air and oxygen with high-flow nasal cannulae
(HFNC). PICU clinicians thus currently have at their disposal several
respiratory assistance modalities for infants with moderate to severe
AVB, but few high-grade evidence studies to guide their
choice.7 Indeed, most of the studies carried out in
this field have been observational, with comparisons with historical
cohorts,5,6,8-10 or physiological, assessing
differences with and without noninvasive respiratory
support.4,11-13
In this issue of Pediatric Pulmonology , Delacroix et al.
re-evaluates the use of BiPAP as the first-line respiratory support in
less-than-6-months patients with bronchiolitis.14 In
their single-center retrospective study, they report longer durations of
noninvasive support and longer PICU stays in the patients supported with
BiPAP compared with CPAP and HFNC. The authors should be congratulated
for this analysis covering more than 250 infants, one of the largest
cohorts treated with this device. This work usefully complements the
information provided by two recent observational studies that focused on
the comparison of these three techniques in this specific group of
patients.15,16 The inherent limitation of these
retrospective studies, whether monocentric14,16 or
database-driven,15 is the presence of confounders,
which influence both the choice of the initial respiratory support and
the outcome. It was particularly interesting to note in Delacroix et
al.’s study that the clinicians’ preferred choice was BiPAP, although no
local written protocol required it in this situation. However, the
BiPAP-treated group also included a higher rate of premature infants, a
condition associated with the immaturity of immune defenses and airway
development and ventilation-induced airway injury that predisposes to
more severe bronchiolitis.17
Several national and multinational surveys have demonstrated that
pediatric intensivists currently select HFNC for initial respiratory
care in cases of severe bronchiolitis.18,19 This
popularity among caregivers appears to be associated with the perception
of a technique that is easy to implement, with comparable effectiveness
and fewer complications than CPAP.20 The physiological
background for using CPAP in this instance is that the application of
nearly constant pressure support is associated with rapid unloading of
respiratory muscles, increased expiratory time, and concomitantly
improved respiratory distress.4,11,21 Reduced
respiratory effort and a change in breathing pattern suggest that CPAP
improves the work of breathing by offsetting the patient’s inspiratory
effort to overcome intrinsic end-expiratory pressure (PEEPi). In
addition, positive airway pressure helps maintain airway patency and
alleviate bronchiolar obstruction, a ‘stenting’ effect that in turn
reduces respiratory system resistance. HFNC also generates some degree
of airway distenting pressure, which supports inspiratory effort. The
reduced diaphragmatic electrical activity and decreased esophageal
pressure swings also confirm the effectiveness of HFNC to reduce the
work of breathing in AVB.12,13
Randomized controlled trials, however, have found that neither CPAP nor
HFNC reduces the need for intubation in infants with bronchiolitis,
probably due to the current low occurrence of this
event1,2,21-24 In practice, CPAP and HFNC are
introduced early in the course of the disease—even as a preemptive
measure in some cases—in infants generally not exhausted. While NIV is
widely used to treat bronchiolitis,25 most clinicians,
unlike in Delacroix et al.’s study,14 consider BiPAP
the next step for patients failing with HFNC or
CPAP.19 Failure rates vary widely, from 10% to 50%
in the major randomized controlled studies,1,2,22-24and depend on multiple factors, the most important probably related to
the criteria and delays in defining failure. According to the TRAMONTANE
study, the main causes of failure are worsening respiratory distress,
especially in patients supported by HFNC; patient discomfort, the
leading cause in patients treated with CPAP; and the occurrence of apnea
in a minority of cases in both groups.22 The very
large cohort of almost 6500 patients collected by Clayton et
al.15 in more than 90 PICUs in North America and Saudi
Arabia gives credit to clinicians who turn to NIV in the event of
failure, since this strategy seems to avoid escalation to invasive
mechanical ventilation (IMV) in more than 70% of cases. There is a
rationale for using BiPAP in infants with bronchiolitis and worsening
respiratory failure, but is this technique being used optimally?
Delacroix et al. point out that the unfavorable results in their BiPAP
group may have resulted from a suboptimal patient-ventilator
interaction.14 Indeed, use of pressure support in
spontaneous/timed modes requires inspiratory synchrony, expiratory
synchrony, and rapid compensation for leaks in order to reach
pre-established pressure values during inspiration.26Infants, especially when exhausted, have a higher respiratory rate,
lower tidal volume, and weaker inspiratory efforts, making
synchronization with their ventilator more complex.27Patient-ventilator asynchrony is frequent during IMV or NIV with
pressure support in infants and children.28,29 In an
elegant physiological study performed in infants with AVB, Baudin et al.
characterized the main inspiratory asynchronies with noninvasive
pressure assist control ventilation from diaphragmatic electrical
activity recordings.30 Autotriggering, double
triggering, and above all non-triggered breaths were observed for nearly
40% of the respiratory cycles, highlighting difficulties in detecting
inspiratory effort in patients younger than 6 months, the targeted
population in the study by Delacroix et al.14 These
triggering asynchronies are associated with leaks, notably when BiPAP is
performed with a nasal interface.31 This issue is
explained by the insufficient sensitivity of the triggers with regard to
the modest volumes and flows generated during inspiration at this young
age.32 In addition, airway obstruction and dynamic
hyperinflation may increase the frequency of ineffective respiratory
efforts.30 The asynchrony index could be even higher
if premature and late cycling are considered, i.e., asynchronies related
to excessively long or short ventilator inspiratory times in relation to
the neural command.28 These expiratory asynchronies
are influenced by the ventilator’s mode and algorithm and may be
improved by adjustments of the cycling-off criterion, which remains a
difficult bedside challenge.28
In adults, patient-ventilator asynchrony has been associated with
increased duration of mechanical ventilation, sleep disorders, prolonged
ICU stay, and increased mortality.33 Such a
demonstration has not been made in pediatrics, but recognition of this
phenomenon and the analysis of its risk factors and consequences are
much more recent.34 Currently, technological advances
in ventilators have opened new horizons regarding synchronization, even
in this group of patients. NIV software, management of leaks, and
turbines specifically dedicated to NIV are indisputable
advances.35 Neurally adjusted ventilatory assist
(NAVA), initially developed for intubated patients, offers another
option. In the field of severe AVB, an early report highlighted the
interest of NAVA in providing less aggressive IMV and more comfort to
the child.36 Discomfort during NIV is common in
infants,22 and the prescription of sedatives is
systematically considered by some teams.37 The
discomfort may have multiple origins, including intolerance of the
interface, skin breakdown, conjunctivitis, and gastric
distension.38 Patient-ventilator asynchrony is another
important source,39 which can be significantly
improved with NAVA. Indeed, the direct analysis of diaphragmatic
depolarization reduces the trigger delay, leading to more effective
synchronization than with conventional NIV, even after careful
optimization of the expiratory trigger
setting.28,30,40 The asynchrony index may be reduced
to 2%-8%, i.e., lower than the critical threshold of 10% defined in
the adult population,41 with a nasal interface and in
the presence of large leaks. One of the restraints on using NAVA is the
extra cost it entails. However, the targeted population is limited to
HFNC or CPAP failure, corresponding to 10-15% of moderate to severe
AVB.15 A recent physiological study in severe AVB
infants found that, compared to CPAP, NAVA was associated with a
decreased work of breathing, lower neural drive and lower Ti/Ttot
ratio.42 The promising results of this study suggest
that pediatric intensivists must be as ambitious in combating asynchrony
as they have been in combating pain and discomfort.43The impact on patient outcome will be judged in randomized controlled
trials targeting severe forms of the disease.
Disclosure: The authors declare no conflict of interest.