Edi the vital sign of respiration, educational video about Edi in PS to avoid VIDD
In this video Dr Stefano Muttini at Vimercate Hospital, Italy, explains how his team uses Edi as a vital sign in Pressure Support to avoid VIDD, on …
What can the diaphragm tell us about the clinical
use of Pressure Support or any other modes
of mechanical ventilation?
Dr Stefano Muttini works in the medical/surgical
Intensive Care Department of Vimercate, Italy,
together with eight other physicians.
They met the challenge of treating over 390
patients during 2014 in their 8 bed ICU.
In the past, Dr Muttini and colleagues were
looking to work more efficiently with better
Our aim in the ICU is to gain time restoring
old parameters using most of the time systematic
therapy in addition to causal therapy in order
to help the patient.
Those parameters are provided by vital signs.
Such as physical examination for the hemodynamic
vital signs are pressure, like central venous
pressure, invasive blood pressure, pulmonary
For the kidney function a vital sign is a
In respiratory terms we can consider two vital
signs: Respiratory rate, Tidal Volumes and
Systematic therapy carries side effects that
should be minimized.
We know that ventilator therapy may initiate
or exacerbate lung injury.
Protective ventilation has become standard
of care to prevent ventilator associated lung
More recently, another side effect involving
diaphragmatic weakness has been investigated.
It is known as ventilator-induced diaphragmatic
We need more vital signs from the respiratory
system and the scientific research in recent
years, is focusing, among others, on diaphragm
It seems important to allow as much diaphragmatic
activity as possible in order to minimize
the risk of VIDD and to determine how to promote
that the automatic effort during mechanical
The process from controlled to assisted ventilation
should be synchrony between the patient’s
own respiratory effort (from the breathing
center) and ventilator delivery of inspiratory
The many advantages of synchrony according
to Dr Muttini are: normalized work of breathing,
avoidance of diaphragm atrophy, decreased
need for sedation, patient comfort, improved
patient-ventilator interaction, early weaning
and improved hemodynamics.
Traditionally in pressure support ventilation,
a low respiratory rate in response to increased
pressure support level was seen as an indication
But how can we see if this is true?
As well as if tidal volumes of 6, 8 ml per
kilo predicted body weight are good indicators
of ventilator assistance level.
Do we have more vital signs that can be useful
to explore the adequacy of ventilatory assistance
during pressure support ventilation?
For the respiratory system we have a few respiratory
vital signs such as respiratory rate, Tidal
volume and SpO2.
We also have other valuable information such
as blood gas analysis, imaging and all parameters
that derive from mechanical respiratory maneuvers
like occlusion maneuvers, pressure / volume
curves, weaning index.
But these are not true respiratory vital signs.
So what can the diaphragm tell us about clinical
use of Pressure Support? In Dr Muttini’s
experience, the electrical activity of the
diaphragm, or Edi in Pressure Support may
inform us that:
increasing levels of pressure support may
depress the electrical activity of the diaphragm,
may illustrate trigger delay and autotrigger
in Pressure Support, or
may reveal ineffective patient efforts to
In terms of timing asynchrony in supported
ventilation, the timing asynchrony that is
discrepancy between the timing of the neural
respiratory cycle of the patient and that
of the ventilator, such as auto triggering
and ineffective efforts and the flow asynchrony
that is discrepancy between the amplitude
of neural respiratory output and the level
of assist from the ventilator, they both result
in lack of proportionality, which is very
hard to see without Edi.
Asynchrony means increased need for sedation,
worsening of gas exchange, and prolonged ventilation.
It can lead to higher morbidity and mortality
because of prolonged ICU stay.
According to Dr Muttini, the Edi curve reveals
common pitfalls with Pressure Support ventilation
A small patient effort only to trigger the
ventilator may result in flow assist asynchrony
– where the patient is not in control of
their breathing pattern.
Timing asynchrony in pressure support between
patient and ventilator is greater at higher
levels of assistance, leading to reduced respiratory
drive and overinflation.
If you want to know, and use Edi you need
to know the physiological basis.
All muscles generate electrical activity to
excite muscle contraction, including diaphragm
and respiratory muscles.
Electrical excitation is controlled by nerve
stimulus and controlled in magnitude by adjusting
stimulation frequency or nerve fiber recruitment.
Edi is related to the phrenic nerve activity,
which is present in healthy subjects as well
as patients with acute respiratory failure
and chronic respiratory insufficiency.
In the experience of Dr Muttini, the Edi is
a Vital Sign – the electrical activity of
the diaphragm is related to global inspiratory
efforts and can indicate anatomical issues
such as diaphragmatic hernia, central causes
such as no respiratory drive due to sedation,
hyperventilation or brain injury, and peripheral
causes such as diseases with phrenic nerve
or neuromuscular junction involvement.
If the patient didn’t have any central nervous
or peripheral dysfunction that could influence
the Edi generation or its detection, the Edi
value variation tells us something about the
Increased Edi maybe be related to respiratory
load or exercise. It is related to the increased
Reduced Edi values may be related to increasing
unloading with mechanical ventilation or decreased
central respiratory drive due to sedation.
In monitoring, we find Edi useful in different
ventilation modes, so we can see the adequacy
of ventilation support.
Patient ventilator asynchronies are evolution
of neuromuscular deficits and respiratory
We can even monitor the Edi of spontaneously
breathing patients when they come off of the
In ventilation therapy we find Edi useful
in applying the mode Neurally Adjusted Ventilator
Assist or NAVA.
Dr Muttini and his team will now demonstrate
the application of Edi on an ICU patient.
They follow a simple standard procedure for
insertion of the Edi catheter.
Select Edi catheter size for the patient:
You need to know the patient’s height. A
table provides more details.
Make a NEX measurement:
From the bridge of the Nose,
To the Earlobe,
And then to the Xiphoid process.
This is the NEX measurement.
Calculation of insertion distance for the
Edi catheter depends on the catheter size
and if nasal or oral insertion.
Next dip the Edi catheter in water for a few
seconds and insert up to the calculated insertion
It is placed in the esophagus just like a
normal feeding tube.
If you intend to also use the Edi catheter
as a gastric feeding tube, remember to follow
your hospital routines for gastric tube placement.
Verify the Edi catheter position by analyzing
the ECG waveforms.
Ideally, P and QRS waves are present in the
top ECG curves.
While the P waves gradually decrease and disappear
in the lower ECG curves, where QRS amplitude
If the bottom leads are highlighted, insert
the Edi catheter further until the pink highlight
appears in the center.
If the top leads are highlighted, retract
the Edi catheter slightly until the pink highlight
appears in the center.
The pink indication can only be used if there
is a stable Edi signal.
If the Edi signal is very low or absent, there
will be no pink highlights.
An absent or low Edi signal may be due to
hyperventilation, sedation, muscle relaxants
or neural disorders.
If the respiratory drive becomes inhibited,
the Edi catheter can stay in place until the
Edi signal is re-established.
Our routine is to perform the NEX measurement
first and then we look carefully at the esophageal
ECG wave form.
Finally, we look at the Edi curve and perform
an expiratory pause to note if the Edi signal
curve coincides with the pressure curve deflection
during airway occlusion.
Now we are ready to illustrate the value of
the electrical activity of the diaphragm for
This patient is a patient admitted more than
two months ago in the intensive care unit
because of septic shock and a secondary acute
respiratory distress syndrome.
Now the patient has become better from septic
shock but she remains with respiratory failure.
She’s now on assisted ventilation and Pressure
Support and we can see the parameters of pressure
Now the patient is on 5 cmH2O of PEEP, a low
FiO2 and Pressure Support level of 14 cmH2O
And if you look at the parameters like respiratory
rate and tidal volume, you can see that the
patient she’s receiving an adequate amount
of ventilation because the respiratory rate
is below 20 and the tidal volume is 6/6,5
ml per kilo predicted body weight, so I can
say that the patient looks adequately ventilated
with these parameters of assisted ventilation.
Even the blood gas analysis is ok with this
But the true problem is what we can say about
diaphragm activity on this ventilatory setting
I can talk about the adequacy of ventilation
because of respiratory rate and tidal volume
and because I can’t see any ineffective
efforts or any major asynchrony on the flow
But I think I need something more and the
something more is the Edi signal.
So let’s see what’s happening when we
display an Edi signal on the ventilator through
an Edi catheter.
As you can see, we don’t have any Edi activity
displayed on the ventilator.
But we have some degree of flow asynchrony.
The flow asynchrony means lack of proportionality
between patient needs and ventilator assistance.
You don’t have almost anything performed
by the diaphragm, but you have a planned activity
So let’s try to decrease the Pressure Support
and see if we can see something more coming
from the diaphragm.
We’ll go down in 3 cmH2O steps and we will
look at the Edi signal.
As you can see there is an ineffective effort.
Something more happens.
I can see something coming from the diaphragm;
it’s still very low in amplitude.
So I think I can go down further.
3 cmH2O step.
Let’s wait a few seconds.
It’s very important not to resize the Edi
scale because you can’t perceive the variation.
As you can see, the diaphragm is starting
We don’t have any ineffective efforts on
the flow curve but even on the diaphragm Edi
curve the diaphragm activity is still low
so maybe I can go down a little bit with Pressure
Support, remember we were starting from 15
and we have lost 7 cmH2O Pressure Support
above PEEP, let’s go down to 5 and let’s
see what happens.
We can see that the Edi peak is increasing,
we lack some tidal volume but we don’t have
high respiratory rate now, we have to wait
that the central respiratory drive of the
patient can understand that there is some
support from ventilation and generate a more
intense signal for the diaphragm.
Let’s see if it will happen in the next
Ok, at the moment we have good diaphragmatic
activity, we have still low tidal volume that
maybe is not enough but we will see on the
blood gas analysis.
We don’t have high respiratory rate.
If the patient will not increase the Edi peak
too much, I guess this set is a good set of
ventilation for this patient.
We will check the blood gas analysis in a
We will check the Edi activity in a few minutes.
But if the Edi peak is stable, I can say that
the patient is properly working with the diaphragm
and the patient has lost almost 10 cmH2O of
That means that the patient was really over
assisted before even if her respiratory rate
and tidal volume tell us that the level of
assistance was adequate.
Let’s see what happens if you want to trust
the usual vital sign that we use in assisted
ventilation like respiratory rate and tidal
As you can see with this Pressure Support
setting we have a good and pretty evident
diaphragmatic activity but we have a little
bit high respiratory rate and a tidal volume
that is quite low.
So usually what we have to do, if we don’t
have Edi signal, we want to increase Pressure
Support because we want to lower respiratory
rate and to have a higher tidal volume.
So we go up in 3 cmH2O steps; remember that
Edi is more or less 10 microvolts now, and
let’s see what happens.
The central respiratory driver response is
very fast usually, so let’s see in a few
seconds what’s happening.
Nothing special, the respiratory rate is still
Let’s go up again.
In 3 H2O steps
Let’s give the time to the patient to know
that we are assisting her.
As you can see the diaphragm activity is less
evident, we have 6 microvolt of Edi peak.
Not lower and not so evident is the present
respiratory rate and not a big augmentation
of tidal volume.
So I want to have a respiratory rate below
3013, so 3 cmH2O increase.
An ineffective effort.
A sigh and then you can see an almost complete
diaphragm activity suppression.
The Edi peak goes down, you have major asynchrony,
this one is a timing asynchrony.
This one is timing asynchrony that you can
see there on the flow curve.
This one is timing asynchrony again.
So ok, this is the situation where we have
a normal respiratory rate, a pretty normal
tidal volume, about 6 ml/kg, but no diaphragm
activity, not at all.
So, this is a clear example of over assistance
during Pressure Support ventilation, If you
are targeting just respiratory rate and tidal
volume and maybe SpO2.
Ok now we are back to 5 cmH2O of Pressure
Support and we can see that we have a good
diaphragmatic activity and we have a good
tidal volume, slightly higher respiratory
rate but this is the true respiratory rate
of the patient the neural respiratory rate
of the patient
It’s not an example of over assistance but
is an example of adequate assistance in Pressure
Support Edi guided.
Ok, let’s talk about asynchrony.
We can have two different asynchrony types.
One is a timing asynchrony and one is flow
The timing asynchrony happens when the ventilator
overcomes the neuroventilatory time of the
And the flow asynchrony is something like
a lack of proportionality in the adequacy
Now we can try to record if with high Pressure
Support level we can see some asynchrony with
Let’s wait a second and then let’s look
at the Edi signal. and let’s see if we can
see something on the Edi curve.
Of course asynchrony is much more evident
with a higher level of Pressure Support.
And asynchrony means morbidity increase, longer
stay on ventilator, longer stay on ICU, a
huge risk of ventilator associated pneumonia,
so we have to avoid asynchrony in the ventilatory
assistance of our patient.
Example of flow assist asynchrony: low Edi
but higher Pressure Support level .
And an example of timing asynchrony: the starting
of the ventilator is not the start of neural
inspiration by the patient and the cycling
of ventilator is good, but the triggering
is not good enough.
And then an example of autotriggering: no
trigger from the patient, but trigger from
And this is another one: no trigger from the
patient trigger from the ventilator, autotriggering.
This one is an example of a small diaphragm
activity but timing asynchrony again.
Flow assist asynchrony is small, not a huge
delay in opening the valve, but a small effort
from the diaphragm and triggering the ventilator.
And the same there, no Edi activity but trigger
from ventilator.That is called autotriggering,
usually is due to the high trigger sensitivity,
but here you can see that the trigger sensitivity
is pretty normal and the flow trigger is not
too low so the auto triggering is really caused
by the high Pressure Support level and the
That means that is a true asynchrony: both
flow and time asynchrony are evident with
high Pressure Support levels.
Let’s go down and let’s see: again the
diaphragm activity will increase and it is
5, 7, 12, Ok, and here we don’t have any
timing asynchrony, even if we are in Pressure
Support, but more importantly because maybe
in this case the ventilator triggered anticipated
the inspiratory neural time of the patient,
but most important we don’t have any flow
That means that lowering Pressure Support
following Edi values give us the chance to
give adequate Pressure Support level assistance.
In their earliest experience of Edi-guided
Pressure Support ventilation, Dr Muttini and
colleagues studied 10 Acute respiratory failure
patients ventilated over 96 hours with inspiratory
Pressure Support titrated to obtain tidal
volumes of 6-8 ml per kg predicted body weight
and active inspiration.
They compared the clinical Pressure Support
setting with Edi guided Pressure Support.
Pressure Support clinical levels were set
according to the study by Thille and collegues
in 2008. At the positioning, Edi peak value
was always below 10 micro volts, with mean
5 micro volts.
Respiratory Rate was below 25 breaths per
minute. Mean clinical Pressure Support was
12 cm of water.
Decreasing the Pressure Support level with
the aim to obtain a “good and evident”
Edi signal, they discovered that Pressure
Support levels were decreased by a mean value
of 5 cm of water with minor changes in Tidal
Volume and Respiratory Rate.
Our need to improve assisted ventilation,
avoiding over assistance and major asynchrony,
and the need of weaning parameters became
clear to us after experiencing Edi guided
During the past seven years we have routinely
treated over 400 patients in our ICU with
Edi guided Pressure Support and NAVA, in acute
respiratory failure patient as well as COPD
patient in invasive ventilation and also noninvasive
We also use Edi to monitor the patients when
they are disconnected from the ventilator.
We feel that Edi is easy to implement and
simple to use.
We think that over assistance and asynchrony
with pressure support ventilation are very
frequent in the global intensive care community
and leads to a greater morbidity.
Edi is a true vital sign coming from our patient
and give us very important information about
patient ventilator interaction.
If you understand Edi, the ventilation mode
NAVA is easy to use.
Just as ECG became a vital sign of cardiology,
let’s try to use Edi as a respiratory vital
sign for monitoring the adequacy of ventilatory
To find out more about ventilator induced
diaphragmatic dysfunction, or Edi monitoring
and NAVA, please refer to the selected list
of publications, or go to the Critical Care