Mechanical Ventilation-Induced Reverse-Triggered CHEST 4 2013

CHEST
Original Research
CRITICAL CARE
Mechanical Ventilation-Induced
Reverse-Triggered Breaths
A Frequently Unrecognized Form of
Neuromechanical Coupling
Evangelia Akoumianaki, MD; Aissam Lyazidi, PhD; Nathalie Rey, MD;
Dimitrios Matamis, MD; Nelly Perez-Martinez, MD; Raphael Giraud, MD;
Jordi Mancebo, MD; Laurent Brochard, MD; and Jean-Christophe Marie Richard, MD, PhD
Background: Diaphragmatic muscle contractions triggered by ventilator insufflations constitute a
form of patient-ventilator interaction referred to as “entrainment,” which is usually unrecognized in critically ill patients. Our objective was to review tracings, which also included muscular
activity, obtained in sedated patients who were mechanically ventilated to describe the entrainment events and their characteristics. The term “reverse triggering” was adopted to describe the
ventilator-triggered muscular efforts.
Methods: Over a 3-month period, recordings containing flow, airway pressure, and esophageal
pressure or electrical activity of the diaphragm were reviewed. Recordings were obtained from a
series of consecutive heavily sedated patients ventilated with an assist-control mode of ventilation
for ARDS. The duration of entrainment, the entrainment ratio, and the phase difference elapsing
between the commencement of the ventilator and neural breaths were evaluated.
Results: The tracings of eight consecutive patients with ARDS were reviewed; they all showed
different forms of entrainment. Reverse triggering occurred over a portion varying from 12% to
100% of the total recording period. Seven patients had a 1:1 mechanical insufflation to diaphragmatic contractions ratio; this coexisted with a 1:2 ratio in one patient and 1:2 and 1:3 ratios in
another. One patient exhibited only a 1:2 ratio. The frequency of reverse-triggered breaths had a
mean coefficient of variability of , 5%, very close to the variability of mechanical breaths.
Conclusions: To our knowledge, this is the first time that the presence of respiratory entrainment
in sedated, critically ill adult patients who are mechanically ventilated has been documented.
The “reverse-triggered” breaths illustrate a new form of neuromechanical coupling with potentially important clinical consequences.
CHEST 2013; 143(4):927–938
Abbreviations: CV 5 coefficient of variation; dP 5 phase difference; EAdi 5 diaphragmatic electrical activity; IBW 5
ideal body weight; IQR 5 interquartile range; Paw 5 airway pressure; Pes 5 esophageal pressure; RASS 5 Richmond
Agitation-Sedation Scale; RRvent 5 ventilator frequency; Ttotmech 5 ventilator cycle duration; Ttotneu 5 neural
respiratory time; VAC 5 volume assist-control; Vt 5 tidal volume
physics, the term “entrainment” refers to the
Inalignment
of the phase and period of a nonlinear
oscillatory system to the phase and period of a periodic external input. In respiratory physiology, “respiratory entrainment” (also called “respiratory phase
locking”) refers to the establishment of a fixed repetitive temporal relationship between the neural and
mechanical respiratory cycles.1
Studies in animals and normal humans have shown
that the respiratory rhythm can be entrained or phase
locked to extrinsic periodic mechanical inflations
imposed during controlled mechanical ventilation.2-8
The pathogenesis involves the activation of vagally
mediated pulmonary reflexes, along with cortical and
subcortical influences.5-7,9-11
The current study arose from an accidental observation: In a patient with a continuous esophageal
pressure (Pes) recording, inspiratory efforts occurred
near the end of each mechanical inspiration in a
repetitive and consistent manner. Subsequently, we
journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
CHEST / 143 / 4 / APRIL 2013
927
reviewed available recordings of patients with an esophageal or a diaphragmatic electrical activity (EAdi)
catheter in place, searching for similar entrainment
phenomena.
We adopted the term “reverse triggering” to describe
these muscular efforts apparently triggered by the
ventilator. To our knowledge, this case series represents the first description of respiratory entrainment
in critically ill patients and introduces a new form of
patient-ventilator interaction: “reverse triggering.”
Because it has a large number of potential clinical
consequences, a better recognition of this phenomenon and a better knowledge of the mechanism at play
are worthy of attention.
Materials and Methods
This study was approved by the ethics committee of Geneva
University Hospital (Comité départemental d’éthique NAC,
NAC 12-032R). Because of the descriptive character of the study,
informed consent from the patients was not required.
Subjects
The study was conducted at the adult medicosurgical ICU of
the University Hospital of Geneva. Over a 3-month period (January
2012 to March 2012), available recordings of flow, airway pressure
(Paw), and Pes or EAdi of consecutive patients admitted to the
ICU and ventilated with an assist-control mode of ventilation
were inspected visually. We detected whether the patients were,
at some point during their recording time, entrained with the ventilator and, if so, the characteristics of this patient-ventilator relationship were identified. The patients’ demographic characteristics,
causes of admission, arterial blood gas levels, and ventilator features were collected from a computerized ICU database. Furthermore, the level of sedation, as assessed by the Richmond
Agitation-Sedation Scale (RASS), was recorded.12
Data Acquisition
Six patients had an esophageal balloon catheter placed for
measurements of respiratory mechanics. Differential pressure
transducers (Validyne MP 45; Validyne Engineering) and a pneumotachograph (Fleisch No. 2, Fleisch) connected to the respiratory circuit were used to record pressures (Paw, Pes) and flow,
Manuscript received July 20, 2012; revision accepted September 10,
2012.
Affiliations: From the Intensive Care Unit Division (Drs
Akoumianaki, Lyazidi, Rey, Matamis, Perez-Martinez, Giraud,
Brochard, Richard), Anesthesiology Pharmacology and Intensive
Care Department, and the School of Medicine (Drs Lyazidi,
Brochard, and Richard), University of Geneva, Geneva, Switzerland;
and Hospital Sant Pau (Dr Mancebo), Servei de Medicina Intensiva, Barcelona, Spain.
Funding/Support: The authors have reported to CHEST that no
funding was received for this study.
Correspondence to: Jean-Christophe M. Richard, MD, PhD,
Soins Intensifs, Hopitaux Universitaires de Genève, Rue GabriellePerret-Gentil 4, 1211 Geneva 14, Switzerland; e-mail: jcm.richard@
hcuge.ch
© 2013 American College of Chest Physicians. Reproduction
of this article is prohibited without written permission from the
American College of Chest Physicians. See online for more details.
DOI: 10.1378/chest.12-1817
respectively. The signals were acquired with an analog-digital
converter (MP 100; BIOPAC Systems, Inc) sampled at 200 Hz. In
two patients with a dedicated catheter (EAdi catheter; MAQUET
Holding GmbH & Co KG), EAdi, flow, and Paw were obtained
by means of a dedicated software (NAVA tracker SV 1.3; MAQUET
Holding GmbH & Co KG). All recordings were stored in a laptop
computer and were analyzed using commercially available software (Acknowledge 3.7.3; BIOPAC Systems, Inc).
Definition of Terms and Data Analysis
Entrainment was defined as a pattern in which the inspiratory
efforts of the patient occurred over a specific and repetitive phase
of the ventilator cycle, therefore, with a minimal variability of their
neural respiratory time (Ttotneu).7 To express the latter, the
coefficient of variation (CV) of the ventilator cycle duration
(Ttotmech) and the CV of the Ttotneu of entrained and nonentrained breaths were compared. In the case of entrainment periods
at some point during the recordings, the following descriptive
parameters were evaluated on a breath-by-breath basis: (1) the
duration of the entrainment, (2) the entrainment pattern or ratio,
and (3) the phase difference (dP) (Fig 1). The entrainment ratio
corresponded to the number of neural breaths within each ventilator breath. Hence, in a 1:1 entrainment ratio, one neural respiratory cycle was associated with one machine cycle; in a 1:2 ratio,
one neural effort occurred every other machine cycle, and so
forth. Of note, we reasoned that to be classified as respiratory
entrainment, the 1:1 ratio had to be apparent in more than five
consecutive cycles and in more than 10 cycles for the other ratios
(1:2 and 1:3).
The dP was defined as the time, in seconds, elapsing between
the commencement of the ventilator and the neural breath (Fig 1).
The onset of the patient’s neural respiratory activity was determined as the point at which a sudden decrease in Pes or a sudden
increase in EAdi was detected. The dP divided by the Ttotmech
and multiplied by 360° provided the phase angle (u), which is the
standard way of expressing the relationship between machine and
neural respiratory activity onset.3,7
The phase angle (u) 5 ([neural onset time 2 ventilator onset
time]/Ttotmech) 3 360° (Fig 1). Therefore, a u of 0° would correspond to neural and ventilator onset coincidence. The degree of
entrainment was assessed by the SD and the interquartile range
(IQR) (25th-75th quartile) of the different u, as well as by the
CV of Ttotneu. Phase angles were calculated solely in breaths
with a clear deviation of Pes or EAdi from the baseline value.
Recordings in which patient effort was absent were not included
in the analysis. Finally, whenever possible, the inspiratory pressure time product of the inspiratory muscles was calculated as the
product of breathing frequency and Pes time integral.13,14
Data were analyzed by descriptive statistical methods and are
expressed as mean ⫾ SD, medians, and the IQR, CV, and box
plots. Statistical analysis was carried out by SPSS software (version 16.0; IBM).
Results
Patient Characteristics
Recordings of Pes or EAdi activity were available
in eight patients. All had a diagnosis of ARDS and
were deeply sedated as indicated by the RASS. They
were ventilated with either volume assist-control
(VAC) or pressure assist-control modes. The ventilators used were Evita XL (Dräger) and Servo-I
(MAQUET Holding GmbH & Co KG). An esophageal
928
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
Original Research
Figure 1. Definition of variables based on flow and Pes tracings. The entrainment duration in this
patient was 32.17 s, and the entrainment ratio was 1:2 (one neural cycle every two mechanical cycles).
Dotted lines denote the commencement of the mechanical and neural cycles. Ttotmech is the duration, in
seconds, of the mechanical cycle, and dP is defined as the interval between the commencement of the
mechanical and the neural inspiration. In this example, dP was 0.66 s and Ttotmech was 2.29 s. The phase
angle (u) was calculated as u 5 dP/ Ttotmech 3 360°, resulting in a value of 104°. dP 5 phase difference;
Pes 5 esophageal pressure; Ttotmech 5 ventilator cycle duration.
balloon was inserted into six patients with the objective of selecting ventilator settings according to lung
mechanical properties. To optimize monitoring of
patient-ventilator synchronization, an EAdi catheter
was placed in two patients. The demographic characteristics, diagnosis on admission, ventilator settings,
and respiratory system mechanics are shown in Table 1.
Entrainment Duration, Observed Patterns,
and Phase Angles
All patients demonstrated entrainment for periods
varying from 12% to 100% of their recording time
(Table 2). The Ttotneu of 1:1 and 1:2 reverse-triggered
breaths had a mean (⫾ SD) CV of 2.2% (⫾0.7%) and
4% (⫾2.8%), respectively, which was very close to
the CV of Ttotmech (0.3% [⫾0.2%]). Periods of
nonentrained efforts (with a CV of Ttotneu of 24.1%
[⫾7.7%]) or absence of inspiratory activity could
intervene between entrainment epochs. Box plots
of the CV of the Ttotneu of 1:1 and 1:2 reversetriggered breaths, and the Ttotmech and Ttotneu
of nonentrained breaths, are shown in Figure 2.
The 1:1 pattern was the dominant entrainment
pattern and the more stable one, enduring without interruption for notably long time periods (Table 2). On the
other hand, the 1:2 pattern was commonly interrupted,
every 10 to 12 cycles, by nonentrainment epochs.
Figure 3 shows representative recordings of the
various entrainment patterns. It can be observed that
the appearance of a “reverse-triggered effort” encroaching upon the end of mechanical inflation makes its
visual recognition extremely difficult in most pressure
and flow waveforms recordings. The negative notch
in inspiratory Paw signal during pressure assist-control
(Fig 3A) and the elimination of the plateau phase
during VAC (Fig 3B) denote the commencement of
neural breath within the mechanical inflation. In
addition, careful inspection of the inspiratory or expiratory flow signal reveals a deviation from its expected
passive shape.
The SD of phase angles during 1:1 entrainment
periods varied from 4° to 10°, and the angles were
wider than those of 1:2 entrained periods (12° to 23°).
Box plots of the median and the IQR of phase angles
are presented in Figure 4. In general, neural efforts
started before the cycling off of the ventilator. In two
patients (patients 1 and 8), 63% and 11% of reversetriggered efforts, respectively, resulted in a mean
(⫾ SD) tidal volume (Vt) inflation of 293 ⫾ 161 mL
(4 ⫾ 2 mL/kg ideal body weight [IBW]) in the first
patient and 400 ⫾ 34 mL (6 ⫾ 0.54 mL/kg IBW) in
the second (Figs 5A, 5B). Pressure time product
of the inspiratory muscles of the entrained breaths
could be reliably estimated in three patients (patients
2, 3, and 8) and their mean values ( ⫾ SD) were
journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
CHEST / 143 / 4 / APRIL 2013
929
Crs 5 respiratory system compliance; F 5 female; IBW 5 ideal body weight; M 5 male; PAC 5 pressure assist-control; PEEP 5 positive end-expiratory pressure; P/F 5 Pao2 to Fio2 ratio; Pplat 5 plateau
airway pressure; RR 5 respiratory rate; Rrs 5 respiratory system resistance; SIRS 5 systemic inflammatory response syndrome; Timech 5 mechanical inspiratory time; VAC 5 volume assist-control; Vt 5 tidal
volume.
40
40
70
45
40
50
70
35
150
140
132
150
229
184
130
180
14
8
9
9
15
13
13
13
46
29
55
27
27
48
42
35
0.82
0.81
0.81
0.75
0.96
0.96
0.82
0.65
10
5
13
15
10
10
8
14
25
31
22
28
30
30
18
25
17
26
20
30
20
20
20
22
ARDS-pneumonia
ARDS-sepsis
ARDS-sepsis
ARDS-sepsis
ARDS-SIRS
ARDS-pneumonia
ARDS-polytrauma
ARDS-pneumonia
1
2
3
4
5
6
7
8
73
68
41
65
61
34
25
43
F
M
M
M
F
M
M
M
PAC
VAC
VAC
VAC
PAC
PAC
VAC
VAC
11
5
7
6
8
6
6
6
Fio2, %
P/F, mm Hg
Rrs, cm H2O/L/s
Crs, mL/cm H2O
Timech, s
PEEP, cm H2O
Pplat, cm H2O
Vt, mL/kg IBW
RR
Mode
Diagnosis
Sex
Patient No.
Age, y
Table 1—Demographic, Clinical, Ventilator, and Respiratory Parameters of the Eight Patients Examined
38 ⫾ 7.2 cm H2O 3 s/min, 20 ⫾ 4.1 cm H2O 3 s/min,
and 19 ⫾ 8.3 cm H2O 3 s/min, respectively.
Effects of Changing Ventilator Settings
The ventilator settings were modified in four
patients:
• In patient 1, the 1:3 entrainment ratio was apparent at an initial ventilator frequency (RRvent)
of 22 breaths/min. Lowering the machine’s rate
to 17 breaths/min, with no other change in ventilator settings, resulted in a 1:2 entrainment ratio
that rapidly changed to a 1:1 entrainment ratio.
• Patient 3 exhibited a stable 1:1 respiratory entrainment during ventilation with VAC at an RRvent of
20 breaths/min. The RRvent was increased to
27 breaths/min, and the neural efforts disappeared.
• In patient 4, the 1:1 patient-ventilator entrainment pattern was maintained when RRvent was
changed from 30 to 26 breaths/min, whereas at a
ventilator rate of 21 breaths/min, all cycles were
triggered by the patient. With an increase of
RRvent to 31 breaths/min, the neural effort ceased.
Other ventilator settings remained unchanged during the aforementioned alterations in RRvent.
We also examined, in the same patient, the effect
of Vt during a stable RRvent of 30 breaths/min.
At this RRvent, 1:1 entrainment was apparent
at a Vt of 380 mL, whereas an increase of Vt to
500 mL eliminated patient efforts.
• Finally, patient 5 was mostly recorded under
pressure assist-control with the settings shown in
Table 1. He exhibited a stable 1:1 entrainment
ratio throughout the whole 654 s of recording
time. Subsequently, the physician decided to switch
the mode to VAC, with an increased RRvent
(from 20 to 31 breaths/min) and a decreased
Vt (from 8 to 4 mL/kg IBW). During 230 s of
recording in VAC with the previous settings, only
short-lived 1:1 entrainment periods (six consecutive cycles) could be identified.
Irregularly Distributed Reverse-Triggered Efforts
Time periods elapsed among entrainment intervals
during which the neural efforts did not fulfill the previously defined entrainment ratios. These reversetriggered efforts always appeared at the same u. These
efforts, however, also seemed to be stimulated by the
ventilator, albeit in a chaotic manner (Fig 6).
Discussion
In this observational study, we documented the presence of respiratory entrainment in critically ill adults
who were mechanically ventilated. We consecutively
observed, in eight deeply sedated patients with ARDS,
930
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
Original Research
Table 2—Sedation Level, Total Recording Time, Entrainment Duration and Ratio, and Arterial Blood Gases in the
Eight Patients Tested
Ratio, % of
Entrainment Time
Patient No.
1
2
3
4
5
6
7
8
RASS
Recording Time, s
Entrainment,
s (% of Recording Time)
1:1
1:2
1:3
Pao2, mm Hg
Paco2, mm Hg
pH
25
25
24
25
25
24
24
25
875
439
467
1658
1538
619
365
1295
407 (46)
58 (13)
184 (39)
1,421 (86)
683 (44)
619 (100)
43 (12)
246 (19)
27
0
100
100
97
100
100
100
66
100
0
0
3
0
0
0
7
0
0
0
0
0
0
0
60
55
86
68
92
74
78
63
46
54
45
47
30
39
46
48
7.41
7.30
7.44
7.46
7.45
7.25
7.41
7.43
RASS 5 Richmond Agitation and Sedation Scale.
neural efforts entrained by the ventilator at three
different ratios: 1:1, 1:2, and 1:3. We defined these
neural efforts apparently triggered by the ventilator
as “reverse-triggered breaths.” They occurred mainly
around the transition phase from mechanical inspiration to expiration, remaining unnoticed by the treating physician. These “reverse-triggered breaths” form a
potentially common, yet unclassified, form of patientventilator interaction.
Pathophysiologic Mechanisms of Respiratory
Entrainment
The pathophysiologic mechanisms of “reverse triggering,” described in the current study, could be
related to the phenomenon of respiratory entrainment,
reported only in animals,4,5,9,11,15,16 healthy humans,3,6,7
and preterm infants.17,18 As in the recovery of rhythmic limb motor movements after peripheral stimulation in spinal cord-injured cats, afferent inputs seem
to play a critical role on entrainment generation.19,20
Failure in the reproduction of the phenomenon in
anesthetized animals after bilateral vagotomy has
emphasized that slowly adapting stretch receptors,
responsible for the Hering-Breuer reflexes, are essential
for respiratory entrainment mechanisms.5,11,15 Despite
its fundamental role, the Hering-Breuer reflex is not
the only factor implicated in entrainment phenomena.
Entrainment can be observed after vagal cooling
in animals and in healthy subjects who have undergone a lung transplant. In these scenarios, however,
there is a wider distribution around each phase angle
and a more limited ventilator rate range.6,9 Rapidly
adapting receptors and vagal C fibers, along with
cortical and subcortical influences, also seem to be
responsible for respiratory rhythm entrainment to
the ventilator.
Methodologic Aspects
We initially defined entrainment visually, based on
the inspection of available recordings, and, provided
that entrainment periods were evident, we proceeded
to phase angles calculation. The strength of the entrainment was assessed by the SD and the IQR of the
phase angles and, additionally, by the variability of the
Ttotneu. Because respiratory entrainment has not
yet been classified formally and various periods or
cycle-based definitions have been adopted, we reasoned that the 1:1 entrainment had to be present
for a minimum of five consecutive cycles and more
complex patterns (ie, 1:2 and 1:3) for a minimum of
10 cycles for entrainment to be characterized as
stable.4,7,9,16,17
Characteristics of Reverse Triggering
Entrainment periods of variable duration and ratios
were observed in all consecutive patients studied,
causing regular reverse triggering. Our recordings
were not specifically performed for this purpose, and
one may thus hypothesize that the phenomenon is
frequent in sedated patients, though scarcely recognized under mechanical ventilation. Indeed, respiratory entrainment could be reproduced in studies
conducted in animals and normal subjects. The true
incidence of the phenomenon in the ICU setting needs
to be studied.
We believe that the fact that all patients in the current study had ARDS possibly reflects (1) the physician’s decision to monitor respiratory mechanics and
neural efforts in this specific group of patients rather
than an association between ARDS and entrainment
phenomena, and (2) that this is a group of patients
often receiving high doses of sedatives. To what extent
the type of critical illness and the magnitude of respiratory mechanics’ compromise promote reverse triggering remains to be explored.
All patients were deeply sedated as suggested by
the low RASS score. Simon et al7 investigated the
effect of the wake state and found that wakefulness
and anesthesia, in contrast to non-rapid eye movement
sleep, broadened the range of machine frequencies at
journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
CHEST / 143 / 4 / APRIL 2013
931
Figure 2. CVs of the cycle durations. Box plots illustrating the median and interquartile range of the
CVs of T tot mech, T tot neural 1:1, T tot neural 1:2, and T tot no entrainment. CV is expressed as a
percentage. CV 5 coefficient of variation; T tot mech 5 cycle duration of mechanical breaths; T tot
neural 1:1 5 cycle duration of reverse triggered breaths during 1:1 entrainment; T tot neural 1:2 5 cycle
duration of reverse triggered breaths during 1:2 entrainment; T tot no entrainment 5 cycle duration of
neural nonentrained breaths.
which the 1:1 entrainment occurred. In this study, the
only one evaluating the effect of the wake state, the
authors suggested that cortical influences facilitate the
resetting of the respiratory rhythm to maintain 1:1 entrainment (the patient modified the rhythm to increase
comfort when ventilated). It is possible, however, that
stronger cortical influences such as pain or agitation
could lead to patient-triggered cycles through neural
efforts enhancement. In other words, the phenomenon
might be present but obscured in awake patients with
strong behavioral responses and more obvious in sedated
patients. Therefore, sedation in the current study could
either promote entrainment phenomena, as suggested
by Simon et al,7 and/or make them more apparent.
In accordance with the literature in healthy subjects, 1:1 was the most frequent and stable entrainment ratio recorded.3,4,10,21,22 Graves et al3 reported
that rarely noticed complex ratios (2:1, 3:1, and so
forth) were more transient, being disrupted every
seven to 15 cycles by irregular patterns. Indeed, in our
study, 1:2 entrainment was not steady, and nonentrainment periods were interspersed between 1:2
patterns. Moreover, the SDs of 1:2 entrainment phase
angles were considerably higher than those of 1:1,
implying a less tight entrainment pattern. Only one
patient exhibited 1:3 entrainment for a brief period
of 25 s before rapidly switching to 1:2.
Additionally, in our study, the ventilator rate was
modified in three patients without any other change
in ventilator settings and in one patient concomitantly
with other changes. In general, increasing the respiratory rate on the ventilator changed the entrainment
characteristics (from 1:1 to 1:3 ratios) or abolished
entrainment. In patient 5, the increase in RRvent
was associated with a reduction in Vt and probably in
alveolar ventilation; despite this, the increase in rate
was associated with less entrainment. A more systematic assessment would be necessary, however,
to understand the influence of the different settings.
We did not measure Paco2, and the aforementioned changes in entrainment characteristics were
observed within 1 to 2 min after each modification of
the ventilator settings. Thus, it is difficult to determine the precise role of Paco2. Simon et al7 evaluated
the effect of Paco2 in entrainment phenomena and
found that mild increases in respiratory drive caused
by CO2 stimulation did not affect entrainment
responses. Nevertheless, it is not clear whether greater
changes in Paco2 or acidosis could affect entrainment through powerful effects on the respiratory
drive.
In one patient, entrained breaths triggered the
ventilator at the end of each passive inflation, thereby
doubling the inspired Vt (Fig 5B). This appeared as
932
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
Original Research
Figure 3. Traces of flow, Paw, and Pes or EAdi during respiratory entrainment epochs.
A, 1:1 entrainment in a patient ventilated with pressure assist-control. Arrows illustrate the
notch on Paw in the presence of the patient’s effort. B, 1:2 entrainment during volume assistcontrol. Arrows illustrate the notch on flow in the presence of the patient’s effort.
C, 1:3 entrainment during pressure assist-control. Arrows illustrate the notch on Paw in
the presence of the patient’s effort. Dotted lines indicate the commencement of neural
breath. EAdi 5 diaphragmatic electrical activity; Paw 5 airway pressure. See Figure 1
legend for expansion of other abbreviations.
journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
CHEST / 143 / 4 / APRIL 2013
933
Figure 4. Phase angles. Box plots of u (median values ⫾ interquartile ranges) during 1:1 (white boxes)
and 1:2 (gray boxes) respiratory entrainment of all eight PTs examined. PT 5 patient.
a “double triggering,” although the triggering mechanism differed for the two breaths. How the physician
can abolish such an undesirable effect by modifying
the ventilator settings cannot be answered based
solely on our results. Previous studies have shown
that entrainment is affected by changes in RRvent
and Vt.3,17 In newborn infants, 1:1 entrainment has
been achieved at ventilator rates close to spontaneous
breathing.17,23-25 In anesthetized humans, 1:1 entrainment was maintained at a range of machine inflations ⫾ 40% of the subject’s spontaneous breathing
frequencies and at mechanical Vt between 40% and
140% of the patient’s spontaneous Vt.3 In vagally
intact animals and humans, when the machine frequency was reduced to below the spontaneous breathing rate, neural efforts preceded the onset of mechanical
inflations and the spontaneous rate decreased.3,5,7
Finally, as the RRvent decreased progressively to
well below the patient’s spontaneous rate, 1:1 entrainment bifurcated into 1:2 entrainment.3,7 Therefore,
in some cases, decreasing the machine rate in a patient
may reduce the spontaneous breathing rate and change
the entrainment ratio and the phase angle, thereby
abolishing high Vt inflations. The few observations
made in the patients suggest that entrainment occurs
within a certain range of frequencies, above which
the patient may be hyperventilated and below which
hypoventilation may increase respiratory drive. The
direct consequences of the reverse triggering will
then depend on the phase angle.
Another important observation is that among entrainment epochs, periods of irregular patient-ventilator
ratios occurred, whereas all neural efforts had a consistent coupling interval with the ventilator (approximately
the same u), despite their regular patient-ventilator
ratios. Such neural breaths have not been described
in the literature. Similar to regular reverse-triggered
efforts, they occurred before or around the transition
phase from mechanical inflation to passive expiration.
In most cases, the physician in charge regarded the
patient as entirely passively ventilated.
Clinical Relevance
The phenomenon of respiratory entrainment could
be clinically important in critically ill patients in many
ways. First, understanding the physiologic mechanisms
of entrainment is linked to comprehension of patientventilator interaction. A 1:1 entrainment (defined also
as harmonic entrainment)4,9 with a very low u, for
example, virtually resembles patient-ventilator synchronization (at least regarding the triggering phase).
934
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
Original Research
Figure 5. Tracings of flow, VT, Paw, and Pes in patients 1 and 8, ventilated with pressure assist-control
and VAC, respectively. Dotted lines illustrate the beginning of neural efforts. A, Patient 1. During
1:2 entrainment, the patient’s inspiratory efforts did not trigger the ventilator but managed to insert
flow into the lungs (arrows) and, thus, increased VT (by 180 mL in this cycle). B, Patient 8. 1:1 entrainment during which the patient’s efforts triggered the ventilator with a sort of “breath stacking” (arrows)
generating very high end-inspiratory VT (approximately 800 mL) during the entrained breaths. Double
arrow line represents the additional VT (400 mL) insufflated into the lungs by an entrained breath.
VT 5 tidal volume. See Figure 1 and 3 legends for expansion of other abbreviations.
On the other hand, a lack of respiratory entrainment,
a more complex entrainment pattern, or patients with
a higher u all constitute different forms of patientventilator asynchrony.
Second, whether the loss of respiratory entrainment is associated with the severity of critical illness
has never been explored. On the one hand, the ability
of the brain stem to entrain the respiratory rhythm to
periodic mechanical inflations is considered a normal
phenomenon steadily reproduced experimentally in
vagally intact humans and animals. It seems to reflect
the ability of the central controller to adapt its output
to afferent information. In addition, the positive impact
of wakefulness to 1:1 entrainment has been interpreted
journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
CHEST / 143 / 4 / APRIL 2013
935
Figure 6. Irregular reverse triggering. Phase angles of sequential neural efforts (gray diamonds) in
relation to sequential ventilator breaths (horizontal axis) in one patient during irregular reverse triggering. We can observe that ventilator cycles are accompanied by neural effort in an unstable ratio.
Despite their irregular relationship with the ventilator frequency, neural efforts start at around the same
phase angle of the mechanical cycle.
by Simon et al7 as an adaptive strategy to avoid discomfort during mechanical ventilation. On the other
hand, respiratory entrainment indicates a loss of
breathing variability.26 Preservation of breathing variability has been linked to improvement of oxygenation and weaning success.27-30 Prospective studies are
needed to investigate the prognostic significance of
entrainment in patients in the ICU.
Finally, although partially assisted ventilator modes
are commonly applied in critically ill patients with
the aim of minimizing respiratory muscle activity,
reverse-triggered efforts were evident in all patients
examined. The consequences of this asynchrony are
potentially large. This may indeed continuously induce
pliometric contractions of the diaphragm. These contractions are associated with muscle cytokine release
and muscle fiber damage. They also induce increased
respiratory muscle work and oxygen consumption,
may contribute to cardiovascular instability, and make
monitoring of the plateau pressure very misleading.30-34
Moreover, reverse-triggered efforts may generate
higher plateau pressure in VAC and large Vt and
transpulmonary pressure swings during pressure assistcontrol. In a study by Papazian et al,35 the administration of neuromuscular blocking agents early in the
course of severe ARDS was associated with improved
survival and more ventilator-free days. In this study,
all patients were ventilated with VAC, and a significantly higher number of patients in the placebo group
developed a pneumothorax. One hypothesis to explain
the results of this study is the potentially deleterious
effects of neural efforts despite the fact that lung protective ventilation was applied. Reverse triggering
could play a role in this setting.
Unresolved Issues
Several unresolved issues arise from our study,
whose small sample size and observational character
limit the analysis. Crucial questions include the incidence of entrainment, its various forms, its physiologic mechanism, and the clinical impact on patients
in the ICU. The impact of modifying the rate, amplitude, and timing of mechanical inflation, the influence
of chemical feedback (Pao2, Paco2, pH), the sedation
level and its type, and whether this is a form of peripheral receptor or central controller dysfunction are all
largely unknown. Finally, whether the physician can
take advantage of respiratory entrainment to optimize
patient-ventilator interaction has not been explored.
Conclusions
In this observational study, we documented, for the
first time to our knowledge, the presence of “reverse
triggering” in deeply sedated, critically ill adult patients
under volume or pressure assist-control mechanical
ventilation for ARDS. In the context of entrainment,
“reverse-triggered” breaths constitute an unclassified
and unrecognized form of patient-ventilator interaction. They can be totally overlooked, leading to erroneous assumptions regarding the patient’s respiratory
936
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
Original Research
status. The description of this phenomenon represents
a step forward in the comprehension of the complex
mechanisms dictating the patient-ventilator relationship. Further prospective studies are needed to explore
in detail the phenomenon and its consequences in
the ICU setting.
Acknowledgments
Author contributions: Dr Richard is the guarantor of the manuscript and takes responsibility for the integrity of the data and
the accuracy of the data analysis.
Dr Akoumianaki: contributed to the identification of the phenomenon, data recording, data analysis and interpretation, and writing
of the manuscript.
Dr Lyazidi: contributed to the data collection, statistical analysis,
creation of images and tables, and review of the manuscript.
Dr Rey: contributed to the recruitment of patients, data interpretation, and revision of the manuscript.
Dr Matamis: contributed to the data interpretation and revision of
the manuscript.
Dr Perez-Martinez: contributed the data collection, data analysis,
and revision of the manuscript.
Dr Giraud: contributed to the recruitment of patients and review
of the manuscript.
Dr Mancebo: contributed to the provision of feedback on the
manuscript.
Dr Brochard: contributed to the scientific guidance for the project,
study design, data interpretation, and revision of the manuscript.
Dr Richard: contributed to the identification of the phenomenon,
study design, results interpretation, and revision of the manuscript.
Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Brochard’s
research laboratory has received grants from several ventilator
companies for specific research projects over the last five years
(MAQUET, NAVA; Covidien AG, PAV; Dräger, SmartCare;
Philips Respironics, NIV; General Electric, FRC). The remaining author report no potential conflicts of interest exist with any
companies/organizations whose products or services may be discussed in this article.
References
1. Georgopoulos D. Effects of Mechanical ventilation on control of breathing. In: Tobin MJ, ed. Principles and Practice
of Mechanical Ventilation. New York, NY: Mc Graw-Hill;
2006:715-728.
2. Baconnier PF, Benchetrit G, Pachot P, Demongeot J. Entrainment of the respiratory rhythm: a new approach. J Theor Biol.
1993;164(2):149-162.
3. Graves C, Glass L, Laporta D, Meloche R, Grassino A. Respiratory phase locking during mechanical ventilation in
anesthetized human subjects. Am J Physiol. 1986;250(5 pt 2):
R902-R909.
4. Muzzin S, Baconnier P, Benchetrit G. Entrainment of respiratory rhythm by periodic lung inflation: effect of airflow rate
and duration. Am J Physiol. 1992;263(2 pt 2):R292-R300.
5. Petrillo GA, Glass L, Trippenbach T. Phase locking of the
respiratory rhythm in cats to a mechanical ventilator. Can
J Physiol Pharmacol. 1983;61(6):599-607.
6. Simon PM, Habel AM, Daubenspeck JA, Leiter JC. Vagal
feedback in the entrainment of respiration to mechanical
ventilation in sleeping humans. J Appl Physiol. 2000;89(2):
760-769.
7. Simon PM, Zurob AS, Wies WM, Leiter JC, Hubmayr RD.
Entrainment of respiration in humans by periodic lung inflations. Effect of state and CO2. Am J Respir Crit Care Med.
1999;160(3):950-960.
8. Pisarri TE, Yu J, Coleridge HM, Coleridge JC. Background
activity in pulmonary vagal C-fibers and its effects on breathing. Respir Physiol. 1986;64(1):29-43.
9. Muzzin S, Trippenbach T, Baconnier P, Benchetrit G.
Entrainment of the respiratory rhythm by periodic lung
inflation during vagal cooling. Respir Physiol. 1989;75(2):
157-172.
10. Petrillo GA, Glass L. A theory for phase locking of respiration in
cats to a mechanical ventilator. Am J Physiol. 1984;246(3 pt 2):
R311-R320.
11. Vibert JF, Caille D, Segundo JP. Respiratory oscillator entrainment by periodic vagal afferentes: an experimental test of a
model. Biol Cybern. 1981;41(2):119-130.
12. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond
Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002;
166(10):1338-1344.
13. Sassoon CS, Light RW, Lodia R, Sieck GC, Mahutte CK.
Pressure-time product during continuous positive airway
pressure, pressure support ventilation, and T-piece during
weaning from mechanical ventilation. Am Rev Respir Dis.
1991;143(3):469-475.
14. Aslanian P, El Atrous S, Isabey D, et al. Effects of flow triggering on breathing effort during partial ventilatory support.
Am J Respir Crit Care Med. 1998;157(1):135-143.
15. Mühlemann R, Fallert M. Hering-Breuer reflex during artificial respiration in the rabbit. II. Relfex blocking by stepwise
vagal cooling [in German]. Pflugers Arch. 1971;330(2):175-188.
16. MacDonald SM, Song G, Poon CS. Nonassociative learning
promotes respiratory entrainment to mechanical ventilation.
PLoS ONE. 2007;2(9):e865.
17. Bignall S, Kitney RI, Summers D. Use of the frequencytracking locus in estimating the degree of respiratory entrainment in preterm infants. Physiol Meas. 1993;14(4):441-454.
18. Bignall S, Summers D. Effects of intermittent mandatory
ventilation on respiratory timing in preterm infants. Early
Hum Dev. 1994;36(3):175-186.
19. Barrière G, Leblond H, Provencher J, Rossignol S. Prominent
role of the spinal central pattern generator in the recovery
of locomotion after partial spinal cord injuries. J Neurosci.
2008;28(15):3976-3987.
20. Duysens J, Van de Crommert HW. Neural control of locomotion; the central pattern generator from cats to humans. Gait
Posture. 1998;7(2):131-141.
21. Benchetrit G, Muzzin S, Baconnier P, Bachy JP, Eberhard A.
Entrainment of the respiratory rhythm by repetitive stimulation of pulmonary receptors: effect of CO2. In: vonEuler C,
Lagercrantz H, eds. Neurobiology of the Control of Breathing.
New York, NY: Raven Press; 1986:263-268.
22. Rodenstein DO, Stănescu DC, Cuttita G, Liistro G, Veriter C.
Ventilatory and diaphragmatic EMG responses to negativepressure ventilation in airflow obstruction. J Appl Physiol.
1988;65(4):1621-1626.
23. Amitay M, Etches PC, Finer NN, Maidens JM. Synchronous
mechanical ventilation of the neonate with respiratory disease.
Crit Care Med. 1993;21(1):118-124.
24. Greenough A, Morley CJ, Pool J. Fighting the ventilator—
are fast rates an effective alternative to paralysis? Early Hum
Dev. 1986;13(2):189-194.
25. South M, Morley CJ. Synchronous mechanical ventilation of
the neonate. Arch Dis Child. 1986;61(12):1190-1195.
26. Tobin MJ, Mador MJ, Guenther SM, Lodato RF, Sackner MA.
Variability of resting respiratory drive and timing in healthy
subjects. J Appl Physiol. 1988;65(1):309-317.
27. Brack T, Jubran A, Tobin MJ. Effect of elastic loading on
variational activity of breathing. Am J Respir Crit Care Med.
1997;155(4):1341-1348.
journal.publications.chestnet.org
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
CHEST / 143 / 4 / APRIL 2013
937
28. Brack T, Jubran A, Tobin MJ. Effect of resistive loading on
variational activity of breathing. Am J Respir Crit Care Med.
1998;157(6 pt 1):1756-1763.
29. Brack T, Jubran A, Tobin MJ. Dyspnea and decreased variability of breathing in patients with restrictive lung disease.
Am J Respir Crit Care Med. 2002;165(9):1260-1264.
30. Wysocki M, Cracco C, Teixeira A, et al. Reduced breathing
variability as a predictor of unsuccessful patient separation
from mechanical ventilation. Crit Care Med. 2006;34(8):
2076-2083.
31. Fridén J, Sjöström M, Ekblom B. Myofibrillar damage following intense eccentric exercise in man. Int J Sports Med.
1983;4(3):170-176.
32. Hirose L, Nosaka K, Newton M, et al. Changes in inflammatory mediators following eccentric exercise of the elbow
flexors. Exerc Immunol Rev. 2004;10:75-90.
33. Sorichter S, Puschendorf B, Mair J. Skeletal muscle injury induced by eccentric muscle action: muscle proteins as markers
of muscle fiber injury. Exerc Immunol Rev. 1999;5:5-21.
34. Thille AW, Rodriguez P, Cabello B, Lellouche F, Brochard L.
Patient-ventilator asynchrony during assisted mechanical
ventilation. Intensive Care Med. 2006;32(10):1515-1522.
35. Papazian L, Forel JM, Gacouin A, et al; ACURASYS Study
Investigators. Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med. 2010;363(12):
1107-1116.
938
Downloaded From: http://journal.publications.chestnet.org/ by a McMaster University User on 04/25/2013
Original Research