contains animated respiratory function graphics. If you cannot see the
animated ventilation waves below, click
The "art" of ventilation – evaluating clinically
what the baby is doing and making changes in ventilation settings – is now being
replaced by the "science" of ventilation – looking at data collected by
respiratory function monitors and manipulating the ventilator accordingly.
This has only been possible with the
sophisticated electronics now available on most ventilators. Respiratory
function monitoring involves the integration of information such as airway
pressures, flow, and volume to evaluate changes in pulmonary mechanics. Data
provided by these systems are ignored at your own peril. Theoretically, being
able to tailor the individual baby’s ventilation according to these data should
reduce the incidence of complications from ventilation, such as air leak and
chronic lung disease. However, no controlled studies have ever been performed to
demonstrate the benefit of monitoring.
Each system is slightly different in terms of
what is measured. No system directly measures volume but, by the measurement of
flow (either directly, as with the Babylog, or indirectly, as with the
VIP-Bird), volume can be calculated if as flow (ml/sec) versus time (seconds).
This has opened a whole new world to neonatologists who were previously obsessed
with pressure (barotrauma). There is animal evidence to suggest that although
pressure may lead to lung damage, it is volume leading to overdistension that is
far more injurious.
From the volume delivered, compliance and
resistance can be calculated. Minute volumes (which truly reflect ventilation)
can be also calculated, and it is easier to know how a baby has responded to a
change in ventilation or management.
There are three basic parts to respiratory
function monitoring. There are "waves",
and indirect measurements (that is, volumes, compliance, and resistance which
are calculated by changes in the measurements in pressure and flow).
The waves are plotted against time. On the
horizontal axis is a time scale. On the vertical axis, the parameter studied.
The usual parameters are pressure, flow and volume.
Pressure (in blue)
is the least useful. It
really only tells us what pressure the ventilator is delivering at the
point of measurement, and tells us nothing about flow or more
importantly volume. However, it is usually permanently on display as
these are the settings that have been prescribed. To show how unhelpful
it is, place your finger over the end of the ETT flow sensor connector.
There will be no flow or volume but the pressure trace will continue
Flow (in red) is shown next. Above the time
scale is inspiratory flow, below is expiratory flow. From the area under
the wave, volume can be calculated. This is useful to detect
obstruction, leak (if it is large enough), and to set the inspiratory
time appropriately in SIMV or SIPPV.
Volume (in green) is probably the key element –
how much air is entering and leaving the lungs (through the ET tube) per
breath. This is calculated via flow versus time. It is useful from a
quantitative perspective to see the tidal volume (aim for 4-8ml/kg per
breath). It is particularly useful to detect leaks, where the volume in
expiration will not return to zero as some air is lost around the ET
There are essentially two loops which are useful
– pressure-volume loops and flow-volume loops.
loops essentially are looking at compliance (change in volume for a
given change in pressure). However, they also can give information about
over-inflation, prolonged inspiration, and leaks.
The image to the right shows a normal
Pressure-Volume loop. The line moving up from left to right is the
inspiratory cycle - as the delivered pressure increases during
inspiration, volume increases.
The line moving down from right to left
indicates deflation of the lung in expiration with a loss of volume with
The green line connecting the points of
minimum and maximum inflation is dynamic compliance. Compliance is
the change in volume for a given change in pressure.
Flow-volume loops can give some information
about leaks (the volume will not return to zero) as well as resistance. No
images are available yet for flow-volume loops.
Below are some scenarios explaining what the
various loops and waves show.
Time Too Short
In the following images,
the inspiratory time has been set too short.
The general form of the pressure wave
is unaltered, other than the plateau phase of the peak pressure is
The flow wave however shows that expiration
starts before a full inspiration has been delivered, and there is a rapid
down-slope on the flow wave.
The volume wave is still on an upslope at
the end of the inspiration, and has not yet started to plateau.
The inspiratory time should be
increased until the inspiratory flow reaches zero and the volume at this
time should plateau out also.
In the following images,
the inspiratory time has been set too long.
The general form of the
pressure wave is unaltered, other than that the plateau phase
of the peak inspiratory pressure is lengthened.
flow wave however shows that expiration starts well after
inspiratory flow has ceased and there is a horizontal line along the
time axis before the downward expiratory phase begins.
volume wave is flattened over the final 0.1seconds of the
inspiration, with no extra volume delivered for the longer breath.
The inspiratory time should be
decreased until the inspiratory flow reaches zero and the horizontal
component along the time axis is abolished.
A leak is present when the
amount of gas inspired does not equal the amount of gas expired.
As per the previous images,
the pressure wave
form is unaltered.
flow wave shows a normal inspiratory flow wave above the line
(with severe leaks, the inspiratory wave may not return to zero flow).
The expiratory flow also looks normal or near-normal but the area of the
expiratory part of the flow is less than the area of the inspiratory
The key wave however for
detecting leaks is the volume
wave. With a significant leak, less gas will come out past
the sensor than is going in, so the volume wave does not return to zero
at the end of expiration but instead resets at the start of the next
Some of the volume and flow
wave forms can take on bizarre shapes if the leak is huge.
The solution to this, if
ventilation is affected, is to change the tube to a larger size.
This may not be possible (for example, you would not put a size 3.0 ETT
in a 650g infant even if the leak was this large), in which case other
strategies may have to be developed. Another option is to extubate to
In huge leaks however,
there may be positive flow into the baby despite the ventilator cycling
through expiration and the flow wave may rise above the zero axis.
This, in some ventilators, can predispose to a situation of "auto-cycling"
or "auto-triggering" where the ventilator thinks that the baby is
breathing but it is in fact detecting flow from the leak. This
results in the ventilator delivering a few breaths in close succession,
then pausing. This can be manually compensated for in some ventilators.
Note that autocycling from
leaks does not occur in the Babylog as it compensates for the leak in
deciding when to trigger. But it will autocycle particularly if
there is water in the ventilator tubing.
In the pressure-volume
loops, overdistension can be detected by flattening of the last part of
inspiration - that is, as the lungs become full, further application of
pressure does not result in any appreciable increase in tidal volume.
This is recognised as "beaking" on the loop.
An alternative way of
objectively measuring this is via the C20/Compliance ratio -
that is, the compliance in the last 20% of inspiration (C20)
should be greater than 80% of the value of the dynamic compliance of the
entire breath. A C20/Cd ratio <0.8 suggests
Overdistension may be a
too high peak
too long an inspiratory
too high peak end
expiratory pressure (PEEP)
The appropriate strategy to
combat this will depend on the underlying reason.
In this example, the
ventilator essentially interprets positive flow in expiration (from
continued flow of gas into the baby because of a leak or pressure/flow
fluctuations in the ventilator tubing) as a breath and triggers
inappropriately. This may take the form of several breaths
together, followed by a pause while the ventilator compensates for the
fast respiratory rate. The way to fix this is to adjust the assist
sensitivity up to above the residual flow in expiration.
Although the Babylog
compensates for a leak, auto-triggering can occur when there is water in
the ventilation tubing. In the Babylog, in PSV or SIPPV mode, it
results in a very high respiratory rate and the
will sound. One option in this situation is to increase the
trigger sensitivity so that the baby has to do more work for a breath to
be recognised as such.
There are some key
principles to using respiratory function monitoring:
It helps to have
a basic understanding of respiratory physiology.
Look at what the
baby is telling you about their lung function. What they tell you now may be
quite different in 5 minutes.
Think about the
effect of any change you make to ventilation because of the monitoring. What
effect did the change in inspiratory time have on volume delivered? Did
turning the peak inspiratory pressure up make any impact on volume or did
the compliance decrease?
default settings – if you are turning the inspiratory time down, have a
limit where you are uncomfortable (e.g. you might set the lower limit of
inspiratory time to 0.25 seconds instead of the default 0.35).
brain to think about what the baby is telling you!
Don't use the
graphics as your sole tool in decision making - it should be used in
conjunction with looking at the baby, the blood gases, and all of the other
information at your disposal.