Transcript
- Welcome to the Sonosite Behind the Scan webinar,
titled Lung POCUS in Critical Care,
with our guest speaker, Dr. Cameron Baston,
from the University of Pennsylvania.
My name is Chris Pennell,
and I'll be hosting today's webinar.
Cameron Baston is a clinical advisor for Penn Health-Tech
and an assistant professor of clinical medicine
in the Department of Medicine at the University
of Pennsylvania's Perelman School of Medicine.
He serves as an associate program director
for the Pulmonary And Critical Care Medicine Fellowship
and as a director of Clinician-Performed Ultrasound
for the Department of Medicine.
He has an interest
in helping create low-cost medical devices
in the resource-limited critical care setting,
and works with several organizations
on POCUS and critical care education.
A mechanical engineer, epidemiologist,
critical care physician, and medical educator,
he spends about two-thirds of his time caring
for critically ill patients,
and the remainder working on the education,
innovation, and health technology.
And with that,
I'll turn it over to Dr. Baston to get started.
Just one moment.
- [Cameron] Hi, my name's Cameron Baston.
I'm a pulmonary critical care doctor
at the University of Pennsylvania.
And I'm here to talk to you about one of my favorite topics,
which is lung ultrasound in critical care.
I'm going to try to organize this talk
with four main objectives.
We're going to briefly review lung ultrasound,
we'll talk about some of the lung ultrasound applications,
and then I'll dive into some
of the critical-care-specific lung ultrasound applications.
We'll end with a little bit of what the future holds
for lung ultrasound as we get to the end of the talk.
So quickly, a brief review of lung ultrasound.
I like to start with this version of "Harrison's,"
which was little over 20 years ago.
And it said that ultrasound
of the lungs was not a useful endeavor,
because it scatters the sound,
and it doesn't provide any meaningful information.
And then one of the grandfathers of this science,
to quote him, "In the quiet of an obscure hospital,
the radiology department was not defended,"
and he stole an ultrasound machine
and surreptitiously scanned patients
in the intensive care unit
and found patterns of sonographic images
that aligned so tightly with the clinical treatments
that he was doing, that he created this art.
And you can look at this,
and if you look at this image of the lungs,
you can see the ribs, the pleural line,
and the A-line, all highlighted by arrows.
And he described this as looking like a bat, the bat sign.
Why this matters to me and why I got so excited about it
is because ultrasound is a portable, inexpensive,
battery-powered diagnostic tool,
which means that you can use it anywhere in the world.
The places where you can't get a chest X-ray
or a CT scan, you can get an ultrasound.
And lung pathology is incredibly prevalent in those regions.
I quickly want to go through the physics of lung ultrasound,
because I think it's so fascinating,
and because I'm a little bit of a nerd.
And I feel like this helps us understand what we're seeing.
So typically, what happens
is that sound waves leave the transducer,
hit the reflector of sound, come back up,
and the transducer hears those and says,
"Oh, there must be a bright white line at this depth."
The pleura, however, is such a good reflector of sound
that some of those sound waves bounce off of the inside
of the skin, go down, hit the pleura again,
and come back up, and the transducer says,
"Oh, those sound waves have traveled twice as far.
There must be a second pleura exactly twice as deep."
This continues, creating a third
and fourth reverberation artifact.
In contrast, if you think about the patients
that have fluid in their alveolar or interstitial space,
the sound waves get trapped inside that little space,
and they create vertical artifacts,
made up of tiny little horizontal lines very close together,
but they look like hyperechoic vertical lines
going all the way to the bottom of the screen.
And those are referred to as B-lines.
Solid structures look like liver,
as you can see here with this densely consolidated lung,
just overlying the diaphragm.
And if there is a pleural effusion,
you can see fluid really easily with ultrasound.
You can see the posterior acoustic enhancement
and the atelectatic lung dancing
inside this pleural effusion.
When it comes to ultrasound, one of the complaints
that many people have is that it's very operator-dependent.
And while that is true for many forms of ultrasound,
particularly lung ultrasound is an achievable skill.
And we refer to the counting of B-lines
as the kindergarten of echocardiography.
This was a study comparing medical students
with an hour of training
to echo sonographers who had been doing this for decades.
And you can see that there is a very tight correlation
in the counting of B-lines acquired.
And we can think about this
as more than just A-lines and B-lines.
We've moved past that,
and we can think about the lung ultrasound alphabet.
So we have A-lines, those horizontal hyperechoic artifacts
that are present in normal lungs, COPD, asthma, PE,
anything where there is normal alveoli and interstitium.
We have B-lines,
where we have the hyperechoic vertical line,
seen here moving back and forth,
present in pulmonary edema, pneumonia,
diffuse alveolar hemorrhage,
anything where there is filling
of the alveolar interstitial space,
or scarring of the interstitial space.
Consolidations are present in pneumonia
or dense atelectasis,
once you've totally deaerated the lung,
making it look like a solid organ.
D stands for diaphragmatic ultrasound.
And this is a growing field
where we can understand the main muscle of respiration,
and we can say whether or not that's a part of the cause
of somebody's respiratory failure.
E is for pleural effusions.
And you can see that free fluid here.
F is for feeding tubes, with this idea that we get a chance
to see whether or not the feeding tube
is traveling down the trachea or through the esophagus.
We can also see the vocal cords here,
if you're thinking about evaluating that,
but that's a more advanced skillset.
G is for gastric emptying.
In the era of GLP-1 inhibitors,
it's really important to understand
whether our patients still have a full stomach
prior to intubation,
or if that full stomach
might be impairing their respiratory excursion.
And then just like my youngest daughter,
we'll skip a few letters of the alphabet.
We'll jump down to S, where we'll get to lung sliding,
which allows us to rule out pneumothorax.
T for a thickened pleura,
which is present more in inflammatory, infectious states,
and allows to differentiate them from the alveolar filling
of cardiogenic volume overload.
And Z for the Zed lines,
throwing back to the French originator
of the lung ultrasound alphabet,
which are present in healthy lungs,
not necessarily a sign of pathology,
but do rule out pneumothorax.
Great.
Now, let's go into some of the lung ultrasound applications
that are present for any specialty,
not critical-care-specific.
When it comes to lung ultrasound for hypoxemia,
the original trial that Lichtenstein published
was the BLUE Protocol.
And what this does is, he basically said, "Hey,
I'm going to use no clinical information,
other than the lung ultrasound,
for a bunch of hypoxemic patients,
and tell you what their cause of hypoxemia is."
I do want to put a little caveat here. Please don't do that.
Please always use the clinical information
that you have available to you,
physical exam and history, labs,
when (laughs) you're trying to figure out the cause
of respiratory failure in your patient.
But I'll highlight the way that this was used.
The first question he asked is,
"Hey, is lung sliding present?"
And you can look here at this clip,
where I highlight the glistening,
shimmering of lung ultrasound.
And you can say, "Yes, lung ultrasound is present."
We go down this channel.
As a reminder,
this is what non-present lung sliding would look like
in a patient who had a pneumothorax.
So since we know lung ultrasound is present,
the next thing we have to ask is,
is there an A profile diffusely or a B profile diffusely?
And if you look at the same clip
that we were evaluating before,
you can see that there are B-lines,
in comparison to the A-lines highlighted,
the blue arrow pointing to the pleura,
the black arrow is pointing to the A-lines,
both here, and the B-lines in this other clip.
So based on the clip that I just showed you,
with a present lung sliding and a diffuse B-line profile,
we would say
that this patient most likely has pulmonary edema.
The thing that I do want to throw out there is,
if you look at that algorithm,
you can see that there's a lot of different ways
to get to pneumonia.
And it is useful to think about pneumonia
because it's such a common cause of respiratory failure.
We have a lot of different findings in lung ultrasound
that are present in pneumonia,
air bronchograms, which are the most specific finding,
a focal area of B-lines, a para-consolidation effusion,
subpleural consolidations, or thickened pleura,
and consolidations, frankly, being a sign of pneumonia.
So if we think about that BLUE algorithm,
we can see that there are multiple different pathways
that end with pneumonia as the final diagnosis,
whether that's an A profile,
with a focal area of B-lines at the base,
or if it's just a mixture of A-lines,
B-lines, and consolidations,
or if it's a absent lung sliding,
with B-lines just about everywhere.
So we want to think, "Is this a useful test?"
And when we look at the results
from Lichtenstein's original paper,
they may seem unbelievable.
Look how high that accuracy is between those sensitivity
and specificity figures.
I'll offer that this is a select cohort
that didn't include patients with pulmonary fibrosis,
post-transplant, anything like that.
That said, when we ask ourselves specifically
about some of these diagnostic buckets,
such as cardiogenic pulmonary edema,
meta-analyses of thousands of patients continue
to show extraordinarily high sensitivity or specificity,
especially when compared to the diagnostic accuracy
of chest radiograph or physical exam.
Similarly, pneumonia, which had a sensitivity
and specificity in Lichtenstein study of 89 and 94%,
has followed up in meta-analysis
to have similar diagnostic performance,
much better than what you'll get
out of a chest radiograph or auscultation.
Knowing that, knowing that this
is a useful diagnostic modality,
let's think about the ways that it's already been used.
And this is a study where we said,
"Hey, if we have long ultrasound and a pulse oximeter,
do we really need the chest radiograph
and ABG that we use to diagnose ARDS?"
There's a more general set of how to do this
for the cardiologist in order to help them understand it.
And we can get beyond just ARDS and volume overload,
to start looking at how to apply this widely,
to the point that we can think about,
"Is this a portable CT scanner,"
instead of thinking about this as a portable chest X-ray.
We can apply this in our patients
with inflammatory lung disease,
and think about a rapid way in the clinic
to detect the efficacy of immunosuppressive therapies.
And this started to grow.
And we could see, year after year,
the number of publications looking at applications
for lung ultrasound was growing, until 2019.
And then came the COVID pandemic.
POCUS in general, not long ultrasound specifically,
but all over, thrives in a resource-limited setting,
whether that limited resource is time,
radiology equipment, space, or providers.
And from the very beginning, we were getting descriptions
of how lung ultrasound was presenting in patients
with coronavirus and coronavirus-induced ARDS.
We talked about how you could integrate this.
And Dr. Resa Lewiss
and others created an article just describing
that POCUS was having its moment in 2020.
This is becoming standard of care in ICUs.
And the pattern of lung ultrasound was being used
to help us differentiate some of the questions
that we had before.
Is this ARDS?
Is this volume overload? Is this both?
And so this was an explosion of research.
And I put that in quotations, because while some
of these are really useful clinical case reports,
and some of these
are very informed lung ultrasound anecdotes,
the number of research articles doubled in a single year.
And we really want to think about this
as something we should be excited by,
but also something that we should study
with the same level of rigor
that we would any other diagnostic modality,
because if we think about the patterns of lung ultrasound
that were described in COVID ARDS,
they're the same that are described
in just about any form of ARDS,
whether it's viral or bacterial.
Outside of ARDS,
lung ultrasound is used in patients
with end-stage renal disease,
because we know that volume overload
is such a critical part of their care.
And we were able to do a study,
showing that the number of zones you scan
will affect your accuracy,
and that the number of B-lines per zone can predict
whether that patient is going to end up in the hospital
or dead at the end of 30 days.
We even started thinking about,
can we use lung ultrasound to just forecast
whether or not a patient is going to do well
if they have end-stage renal disease
and signs of volume overload?
This collection of research got to the point
where the American College of Physicians did a review
of all the evidence,
and put appropriate use of POCUS
for patients with acute respiratory failure
in their recommended guidelines.
And you can see that they were able to say,
"POCUS augments our clinical ability,
adds to physical exam and history."
So that's more generally how we use lung ultrasound.
But I want to talk about my favorite area, which is the ICU.
So how do we use this in patients who are critically ill?
And one of the things we can do is, we can just say,
"Hey, our patients are sick and complex.
And oftentimes, a chest X-ray
does not give us sufficient information."
And so a direct comparison
of lung ultrasound to chest X-ray,
with CT scan as our gold standard,
shows us not just that ultrasound is more accurate
than chest X-ray, with CT as gold standard,
but also where the places that lung ultrasound
is going to fall down will be.
And specifically I'm talking about if you have areas
of peripheral bullous lung,
that's going to block your ability to see anything
that's deeper and more anterior.
Or if you have a mass that has surrounded
by normal lung tissue,
ultrasound won't see it, and CT scan will.
Or if you have a pneumonia
that's hidden behind a gastric bubble,
as is the case in this last slice of a CT,
then you'll have a false negative on your lung ultrasound.
This sort of refinement of the specific diagnostic accuracy
of our patients is incredibly helpful
in knowing how to use lung ultrasound in the ICU.
Beyond that, one of the hardest questions
that I sometimes am faced with is,
is this opacity on a chest radiograph infectious
or simply a matter of a patient being supine for too long
and having atelectasis?
And this study said, "Hey, can we use a couple
of different findings to differentiate this?"
And you can see here an example of a patient
with dynamic error bronchogram,
which they found to be highly specific for pneumonia
in comparison to atelectasis,
with a specificity of 99% in their analysis.
They also added color Doppler of these
to see whether you could have signs of hyperemia
in your infected patient
versus normal blood flow in your atelectatic patient.
The other thing that comes up frequently in the ICU is,
does this patient
have a new ventilator-associated pneumonia?
And while we can do this with their existing standards,
using chest radiograph showing new opacity,
increase or change in character of sputum,
and other lab values,
using lung ultrasound can help us make this more accurate.
And this was a nice protocol, where they said,
"Hey, we want to suspect ventilator-associated pneumonia
if they have one of these findings,
then we're going to do ultrasound,
and we're going to points to our ultrasound to make a score,
and we're going to take the combination of clinical indicators
and ultrasonic graphic indicators."
And they demonstrated that this is highly accurate
for the clinical diagnosis
of ventilator-associated pneumonia.
Another question that arises is,
is this a simple volume-overload-driven effusion,
or is this an effusion that's becoming an infected space
that we need to think about drainage of?
And so this is where we looked at,
can the ultrasound findings differentiate transudative
from exudative pleural fusions?
And what we found is that anechoic fluid
does not help change our post-test probability
of this being transudative or exudative.
But a highly echoic fluid, as seen here,
or a loculated space, as seen here,
are super helpful for helping us understand
whether or not this space is infected.
Beyond just diagnosing whether or not we think
that a fluid space around the lung is infected or not,
we need to be using ultrasound-guided arthrocentesis.
And whether that's just marking a space
and making sure that there's no blood vessels, as seen here,
or in this new protocol proposed in the "POCUS Journal,"
are we thinking about a standardized approach
for visualizing the blood vessels above and below the space
in which we're thinking about putting our needle,
in order to understand
if our patient has an unusual vascular anatomy
that would put them at higher risk,
as seen with this artery
that's somehow right in the middle of the rib space?
We can also think about using POCUS
to meet our diagnostic criteria for ARDS.
This was originally described in what's known
as the Kigali modification of the Berlin criteria for ARDS.
But in the last year, both the American
and European groups have created new definitions
for the acute respiratory distress syndrome.
And you can see specifically that they call out ARDS
in a resource-limited setting,
where you can use POCUS
instead of chest radiograph or chest CT
to meet your necessary imaging diagnostic criteria
of bilateral abnormalities.
Beyond just making the diagnosis of ARDS, though,
we can use scoring systems that have been proposed
for understanding whether or not people are getting better.
And so this is where you take somebody who has ARDS,
and you look at each zone,
and you say, "Does that zone have an A-line pattern,
a pattern with discrete B-lines,
a pattern with confluent B-lines, or consolidations?"
And you give them points,
increasing as they move from one to the other.
Based on these number of points,
and some, you can start to understand a little bit about,
not just the severity of the imaging findings of their ARDS,
but also, are things are getting better or worse?
So for example,
one of the first ways that they used this was to say,
"Hey, can we use this to guide recruitment with PEEP?"
And they did this by taking a patient down to zero PEEP,
or ZEEP, as it is colloquially known,
and then adding PEEP
and seeing what happened to the lung ultrasound findings
to identify recruitable zones of lung.
The protocol here is very clever.
And they were able to show that this indicated which parts
of the lung were going to be recruitable
and which parts weren't.
When they did that, they had very tight correlation
with their exhaled tidal volumes.
The other primary therapy for ARDS that we know works
is prone positioning for moderately severe ARDS.
But sometimes we will wonder,
in our patients who might be at higher risk for proning,
whether that's for hemodynamic instability
or for skin vulnerability,
we'll wonder whether they're likely
to benefit from an oxygenation standpoint.
And so this study was able to show
that if you look at those lung zones
and score them in the same way that we did before,
that you can predict a pattern, more posterior,
higher-scored zones being the patients that are more likely
to have a response to prone positioning.
And that response is seen both in imaging
and potentially in oxygenation.
I do want to put the caveat here
that I teach all of my trainees,
which is that the benefits of prone positioning
in terms of mortality in ARDS are not necessarily dependent
on the improvement in oxygenation,
but this can be helpful when we're trying
to come up with a plan for the day.
As I said before, one of the main questions we have
in our critically ill patients is,
are the imaging abnormalities I'm seeing
from volume overload,
or are they from progressive ARDS,
cardiogenic or noncardiogenic pulmonary edema?
Copetti et al. tried to answer this
by looking at groups of patients that had known ARDS
and groups of patients that had known acute pulmonary edema.
And when they did that,
what they found was that B-lines were present
for all of these patients.
But if you looked closely at the pleura,
you could see a pattern,
where the cardiogenic pulmonary edema patients
had smooth B-lines,
and the ARDS patients tended
to have more commonly pleural changes.
Similarly, you could look at reduced or absent lung sliding,
and the presence of consolidations or effusions
as being useful for differentiating the two groups.
This has been validated in further follow-up studies,
helping us understand the utility of this technology.
If you have identified
that your patient is volume-overloaded,
then lung ultrasounds has a new utility,
which is, hey, how much congestion remains?
Because you can quantify it in a way with B-lines
that you really can't with chest X-ray.
And so when you look at this,
they looked at patients who had more than 15 B-lines,
and they said, "Wow,
this really predicts event-free survival
from heart failure patients when they're discharged."
Those are the ways
in which lung ultrasound is currently being used.
I want to talk now a little bit
about where I see the future of lung ultrasound going.
And one of the first places is expanding the indications.
We're learning more and more
about how lung ultrasound presents
in different pathologic etiologies.
And so, for example, PE,
which is typically described as just having clear lungs,
can actually have small micro-consolidations.
And you can look at the blood flow inside of them
to say, "Hey,
is this because of an infectious inflammatory etiology,
or is this because a clot is preventing blood
to getting to that portion of the lung?"
And then there's this idea
that you could use ultrasound contrast
to actually see the blood flow inside of the lung
and understand where in the lung
is not getting blood flow.
Beyond this, I like the idea
that lung ultrasound is easy to do,
and so it should not be limited
to just the intensive care unit.
Obstetrics and gynecology residents are learning
how to do this to help look for pericardium
and postpartum cardiomyopathy.
They're looking for amniotic fluid embolism.
The Family Medicine group has said
in their ACGME guidelines that this should be a part
of every primary care doc's residency training.
And when it comes to intraoperative lung ultrasound,
there are a number of ways that you can apply this
while the patient is on the table in the OR.
And then afterwards,
you can look at their lung ultrasound findings
to see whether you can predict who's going
to have this more complex postoperative course.
Not just limiting this
to the area of the intensive care unit,
we can think about not just limiting it
to physician scanners.
We can see that APPs are starting to do this.
And respiratory therapists globally have started
to be involved in research on the applications
of lung ultrasound performed by RTs.
Nurses have been using lung ultrasound
in order to make diagnoses of pneumonia in dyspneic patients
and to help with triage in the ER.
And then if I'm trying to think
about where the best financial return on investment is
with the prevalence of heart failure,
it's got to be understanding if I'm making progress
and getting my patient to their ideal volume status
as quickly as possible.
And this is demonstrated in studies where we're trying
to teach internal medicine teams to use lung ultrasound
to drive their diuretic goal,
to decide when a patient's ready for discharge,
or to decide
whether or not a patient's going to be readmitted.
As I said before, lung ultrasound is a study of artifacts.
And so we're learning how to refine our technique
in order to make sure
that those artifacts are as visible as possible.
And that might mean changing the depth, the gain,
which presets happen in the back,
and where we set our zone of focus.
We also have the idea
that the traditional zones of lung ultrasound
are not necessarily the only zones of interest,
and maybe we should be just sweeping the chest,
understanding from top to bottom,
and covering a lot more surface area,
in order to get a better big picture.
When it comes to the even more advanced cases,
if you already have a transesophageal transducer inside
of a patient in order to look at their heart,
you can also look at their lungs.
And this is described
in a couple of case reports in "CHEST,"
where they were able to show that by doing so,
they could identify a PE
or a patient with refractory atelectasis
while in the TEE suite.
And then we have been stuck
with this handheld probe for a long time, and I love it,
but there are more and more scientists working on the idea
that we're going to just put ultrasound wherever we need it
to be, and can leave it there,
in order to understand, over time,
what's happening below that ultrasound patch.
Outside of the hospital, we were able to show
that you can do lung ultrasound on home visits
and answer important clinical questions in a patient
that can't be moved to a chest radiograph,
or who can't have blood drawn easily.
And you can use this to adjust therapies
in the hospital-at-home patient.
And if we're saying we don't need to be in the hospital,
and we're saying we don't need doctors to do it,
then the next step is, can we have our patients
doing lung ultrasound on themselves?
And the answer is probably yes,
as an increasing number of case series have demonstrated.
So to wrap this up, just to think about what I want you
to take away from today's talk,
we briefly reviewed lung ultrasound.
And I said before, and I'll say it again,
this is one of the easier applications for POCUS.
We have an alphabet that helps to guide us,
and it works for all sorts of different lung pathologies.
When it comes to how we're using it,
we're starting to understand that for just about everything,
it's better than a chest radiograph,
and you can see a lot of different pathologies with it,
not just our traditional volume overload versus COPD.
And it's becoming a part of the general guidelines
for specialty and national societies.
In critical-care-specifically,
we can use it to make diagnoses,
and we can use it to predict response to the therapies
that we use for our patients.
And in the future,
we're going to have this being applied outside of the ICU,
outside of the hospital,
being used by non-physician providers
and by patients themselves,
breaking the boundaries
on how this technology can be applied.
Thank you so much.
As we conclude, lung ultrasound is growing in popularity.
The evidence base is exploding.
In critical care, our ability
to detect ventilator-associated pneumonia, pneumothorax,
and modify our ARDS management is incredibly exciting.
And the future of lung ultrasound
is going to incorporate a lot of new techniques.
Here's some citations.
That was a bit of a whirlwind of evidence review.
Thank you so much.
I'm excited to talk to you about questions.
- All right, thanks so much, Cameron. Get you unmuted here.
All right, that was our presentation.
So we have time for Q&A now.
So if you have any questions,
you can enter them in the Q&A box
at the bottom or the side of your screen.
Cameron, thanks so much for being here with us.
Let's see, it looks like we do have a couple
of questions already.
Let's see, can you speak a little more to the training
and implementation of lung POCUS?
- Yeah, this is one of the things
that gets me really excited about lung ultrasound.
I mean, I mentioned that we got to compare lung ultrasound
by medical students, with an hour of training,
to ultrasound done by people who've been using ultrasound
for 10 years.
And their ability to count B-lines is basically the same.
The more exciting part is,
when can you get the kind of accuracy
that Lichtenstein offered in his papers,
or where you can say you're more accurate
than a chest X-ray for pneumonia?
And I'll tell you, the number is still less than 20 scans,
after several papers kind of looking at this.
And that's just an achievable number for anybody.
We can integrate that into medical school training,
we can integrate that into RT school training.
We can integrate that into residency.
And, you know,
my favorite place to target with this is residency,
because if you can get people
to change their clinical diagnostic algorithms
and integrate ultrasound,
as is recommended by the American College of Physicians,
as opposed to saying, "I'm just going to use ultrasound,
I'm just going to use something else,"
then we increase our diagnostic accuracy.
And that means we get the right answer faster.
You know, I like to joke that when a patient,
a typical patient that comes in,
who has a history of heart failure and COPD
and volume overload and airways disease,
and might have a pneumonia,
and you're trying to figure out what to do,
that we almost always end up giving them steroids
and diuretics and antibiotics.
And we kind of hope that it all comes out in the wash.
And usually, it does.
But that the side effect profile of all these things means
that we're not being as elegant as we could have been.
And this is where lung ultrasound can make a difference.
It can just let us nudge ourselves one way or another
to say, "Hey, you know what?
I see diffuse B-lines.
We're going to diurese before we do steroids."
That we'd be fighting each other otherwise,
or, "Man, we can finally see this focal pneumonia.
We can focus on our antibiotics.
We don't need to worry about diuresing this patient,
who might be becoming septic."
And so to me, the training question is really one of,
how do we best integrate this into the training paradigms
that exist for different professions,
so that we can get people to start using this
at a time when it matters, as opposed
to when they've already developed their own protocols,
and we're trying to convince 'em to change something.
I don't know if that's what you were hoping for, Chris.
- [Chris] (laughs) Yeah, that was great.
Yeah, we've got a couple more here.
Can we use this for the extubation of the trachea?
- Yeah, so I think this question is asking about,
can we use this to see tracheal edema?
And the answer to that is maybe.
I don't know of any studies that looked at this.
And the challenge is that the airway
is really tricky to visualize because there's air in it.
And the specific place where you'd be looking for edema,
you've got the balloon of the endotracheal tube.
And so that makes it a little tricky.
Now, if you're asking about extubation in general,
is this patient ready?
That's been studied a little bit more.
And so if your reason for intubation was,
for example, volume overload,
then because B-lines disappear in real time
as you diurese somebody, the answer is yes.
And the number of B-lines
can predict their likelihood of reintubation.
For me, I'll be honest, I don't count B-lines
and decide whether or not to extubate somebody,
but I do look and see,
"Hey, man, there's still a lot of B-lines.
I'm going to be extra aggressive with my diuresis
before extubation, and I'm going to think about,
'Is this the patient who I extubate to positive pressure,
to BiPAP or CPAP,
rather than extubating to high flow or nasal cannula?'"
And so, again, thinking about this not as one thing,
but just giving us information
about the overall alveolar interstitial space,
and kind of letting that drive our decisions.
Separate from parenchymal ultrasound,
there is a body of literature,
and I got to write a review on this,
about whether or not we can use diaphragmatic ultrasound
to decide who's going to do well after extubation.
And diaphragmatic excursion,
the answer's almost certainly no.
When somebody's on mechanical ventilation,
that's just driven by the vent.
And diaphragmatic thickening, the answer is maybe.
It does predict reintubation,
it does predict weakness post-extubation.
The challenge with this is it really is only doable
when somebody's on minimal vent settings.
They'd have to be on very low-pressure support
in order to have a meaningful diaphragmatic thickening.
And this is a higher-level skill.
It takes time to get those sort of millimeter measurements
to get really good at those,
as opposed to just counting B-lines,
looking for patterns of A-lines, B-lines, consolidations.
But if you do get good at diaphragmatic ultrasound,
especially if you're working in a transplant center
or a cardiac surgery center,
where you're seeing a lot of people who have potential
for damage of their phrenic nerves
or their intercostal nerves,
then this can be really useful.
You can say, "Oh, man,
this person has diaphragmatic paralysis.
We're going to think about early tracheostomy,
if everything's pointing that way,
as opposed to late tracheostomy,"
or, "Maybe we're going to think about extubating right
to positive pressure ventilation again,
because we're anticipating
that they're going to have a weaker diaphragm
and a longer time for recovery."
So yeah, so I do use it for extubation,
but I ask specific questions,
and I think about optimizing the patient
based on their lung ultrasound prior to extubation.
In the right patient, I think about,
"Is their diaphragm so weak that I want to think
about rescue maneuvers before pulling the tube?"
- [Chris] Awesome.
And next question we have,
do you feel that ultrasound is replacing an echocardiogram?
- Yeah, I mean, this is an interesting question.
I'm going to like parse this out a little bit.
I think the real question is,
is point-of-care ultrasound replacing cardiology
and cardio-sonographer-performed echocardiogram?
And then this is a place
where people's fiefdoms get challenged sometimes.
And I've heard and seen cardiology departments
that are very defensive of this, for good reasons.
They want to make sure
that the information that's being provided
of ultrasound of the heart provides good information
and not bad information,
'cause bad information is worse than no information.
And gosh, they'd rather you call 'em
at any time of day or night
than try to do it yourself and get bad information from it.
And so I like to differentiate the two.
In the same way that a radiology-performed CT scan
is better if you know the clinical question being asked,
but really is a, "Hey, tell me everything you see.
I want to look from top to bottom.
And I want to diffuse examination based on a protocolized time
and slice and thickness," and the sort of thing.
And a full echocardiogram, done by cardiology department,
similarly is a,
"I'm going to evaluate the heart from 20 different viewpoints
and make sure I am getting every single valve interrogated,
every single flow rate interrogated,
every single chamber size interrogated."
This is a bigger study.
Whereas a point-of-care ultrasound of the heart
is really saying,
"I have discrete clinical questions that I'd like to answer.
Is this patient in shock
because of decreased ejection fraction,
because of a pericardial effusion,
because of a failure of their right ventricle?"
And similarly, I think of lung ultrasound
as providing these discrete answers.
Now, if the question was,
is lung ultrasound replacing cardiac ultrasound,
then we get into my favorite thing,
which is none of these things should be done in isolation.
We should be doing all of these together.
Everybody who wants to ask a question
about respiratory failure or shock,
or whatever it might be, should be looking at the heart,
the lungs, the deep veins of the leg,
in order to put together as much knowledge as possible.
And with each of these examinations,
they're asking specific questions.
Because if your question is, "Hey,
does this person have cardiogenic pulmonary edema?"
It turns out that lung ultrasound
is superior to echocardiogram for answering that question.
If your question is, "What's somebody's wedge pressure?"
Lung ultrasound and cardiac ultrasound are like starting
to get close when you look at the performance of this.
Nothing as good as a right heart cath, obviously.
But the idea being that a much easier test
can tell you a lot in the right patient.
So when I do a lung ultrasound
of somebody who I know has heart failure,
and I'm worried about their volume status,
and I see a lot of B-lines,
I don't need to do very complex diastolic measurements
of the pressures in each chamber of their heart
in order to say, "Oh, this is volume overload."
Now, if I have the time to do so,
because it's not clinical emergency, gosh,
that additional information
can increase my diagnostic confidence and allow me to avoid,
again, those unnecessary therapies
that have additional side effects.
I hope that got to the kind of meat
of the question you're asking about,
whether or not this is replacing echocardiogram.
Again, if I was to summarize that,
I'd say, POCUS of the heart doesn't replace echocardiogram,
it augments it in the right clinical setting,
allows us to decide who needs a full echo.
And point-of-care ultrasound of the lungs,
I don't think, is replacing cardiac ultrasound,
but gosh, does it answer better some of the questions
that we sometimes ask cardiac ultrasound,
thinking that it's going to answer for us.
- [Chris] Great. We do have a couple more questions here.
And just a reminder, everybody,
you can get your questions in with the Q&A box at the bottom
or the side of your screen there for Dr. Baston.
Let's see, next question is,
how do you integrate lung ultrasound with other POCUS exams?
- Oh, great.
This tails off nicely of what I was just talking about,
this idea that when we're doing a lung ultrasound
in critical care, we're really trying to decide
what someone's source of respiratory failure is.
And we talked a lot about the evidence
that allows us to do that,
but think how much more powerful it is if I can say,
"All right, there's a ton of B-lines."
I'm trying to figure it out
if they're cardiogenic or noncardiogenic.
And maybe I've got some of that pleural thickening,
maybe I've got some of the pattern recognition
that allows me to nudge one way or the other.
But then I look at their heart,
and they have small chamber size,
hyperdynamic left ventricle, small left atrium, small IVC,
and then I go,
"Oh, this is probably noncardiogenic pulmonary edema,"
I start thinking about diffuse alveolar hemorrhage,
I start thinking about ARDS
in the setting of sepsis or pancreatitis,
or anything else like that.
And then I get to do the same thing
when I think about the deep vein ultrasound.
So I can do an ultrasound of the all lungs,
and I'll be like, "Oh, straight A-lines,
this is going to be probably an airway pathology,"
or, you know, thinking about normal lungs, COPD, asthma.
And then I'm thinking about DVT/PE as another potential way
to get A-lines in a patient who's in respiratory failure.
And then that's where I add a deep vein scan.
And all of a sudden,
I've nudged my pretest probability a little bit.
Now, if I don't find a deep vein thrombosis,
that doesn't mean that PE's off the table,
but gosh, if I find one,
and I've got clear lungs and respiratory failure,
and tachycardia, then I'm starting to really think PE.
And if I have the opportunity then
to look at the right heart and see,
"Oh, man, the right heart's dilated.
That means something."
Again, a small right heart does not rule out small PE,
but ooh, I have started to put together a multi-window view
of what could be causing somebody's respiratory failure.
So for me, in the same way that we would be surprised
if somebody walked in with a stethoscope,
listened to a couple points on somebody's chest,
didn't listen to heart sounds,
didn't examine capillary refill,
didn't look for modeling on the legs,
and said they did a complete exam,
I think this needs to be integrated
with all the other places that we can put the probe
that are aligned with the diagnostic investigation
that we'll perform.
Pulling it together gives us more power.
And that's been shown, there's a study in "CHEST"
where they actually just compared echo alone
versus echo with lung ultrasound,
and showed that the diagnostic accuracy was better
when you had two windows in the body as opposed to just one,
which is not surprising in any way.
- [Chris] Great.
Next question is,
do you use POCUS to help you identify the right site
to put in a chest tube in pneumothorax?
For instance, how far from the sliding point area
do you choose for the chest tube?
- This is a great question.
So short answer, yes.
I do all of my chest tubes with ultrasound guidance,
in the absence of cardiac arrest,
and this is an emergency,
where I don't have the time to even wait for the machine
to come into the room.
But if I'm there, and I have the machine, then yes.
And so then it divides into,
"Do I have time to check for vessels, yes or no?"
And then I have time to see
if there's lung sliding in the place,
in the reasonable anatomic targets.
So first thing I'll do is I'll check and see
if I can see areas where lung sliding appears,
and then disappears,
the lung point, as it's described in the literature,
the place at the bottom of the pneumothorax.
That lets me know I need to be one rib space above that
when I do my chest tube.
The second question that was,
do I want to do a Seldinger technique chest tube,
or do I want to do a dissection,
or open technique chest tube?
And I do both of those in my practice.
And in general,
I think that a Seldinger technique chest tube
is a little more elegant.
It's a smaller hole initially, with the kits we have,
involve a smaller tube site by default.
So I think of it as less discomfort for the patient,
a smaller hole to heal afterwards.
But if I can't find that full rib space above
where lung sliding appears and disappears,
then I probably won't feel safe putting a needle in
in order to do a Seldinger technique,
just for fear of having hit the lungs.
I think that in that case,
discretion being a better part of valor,
I'd rather do a dissected technique and use my finger,
and be able to feel that I'm in the pleural space,
before putting the tube in.
So the short answer to your question is yes,
I use ultrasound.
The longer answer is,
but if I can't find that safe space,
one rib space above where there's any change
between lung sliding and not long sliding,
but I know that this is pneumothorax,
then I still will put in a chest tube.
I'll do a dissected technique
instead of Seldinger technique.
And then if I'm doing that, and I have the 30 seconds,
then I'll quickly switch to a linear transducer
and put on the vascular preset
to make sure there's not that weird anatomy.
There's a lovely study where they looked at CT angiograms
of patients to see
what percentage have these really atypical vascular anatomy
in their intercostal spaces.
And this is specifically looking at,
instead of the above the rib space being clear,
they have vessels right in the target
of where you would go with a chest tube.
And there's certain risk factors that increase that, so age.
You know, if it's a much older patient,
I'm definitely going to look for vascular anatomy,
if I have a chance,
dialysis or other sources of fistulas,
'cause they're going to have these weird vascular,
tortuous paths for their veins
that can cause life-threatening bleeds.
So yes, when I have the chance, I use ultrasound.
If somebody is actively getting CPR,
and we think that massive hemothorax
or tension pneumothorax is a cause,
then I'm decompressing first with a needle,
and then going in with a dissected chest tube,
with or without ultrasound.
- [Chris] Awesome.
All right, next question is,
could you speak to some of the pitfalls of using lung POCUS?
- Oh, yeah, this is great.
Sam Brown really likes writing these articles
about pitfalls of POCUS.
And I've learned so much from reading his things.
And just as you do this,
you find the places where things are not as clean
as they were in Lichtenstein's studies
or in some of the other literature.
So part of it is that everyone thinks,
"Oh, man, this is super easy.
I'm going to put the probe on,
I'm going to see A-lines or B-lines."
But it turns out that the angle of intonation
for the new learner is a real source of error.
And what I mean by that is,
if you're perpendicular to the skin,
that doesn't necessarily mean
you're perpendicular to pleura.
And you need to be perpendicular to the pleura
to really hit the A-lines,
because of that bouncing back and forth
between the skin and the pleura.
And it makes the B-lines much more bright.
So being perpendicular to the pleura
might mean fanning a little bit inferiorly
when you're high up,
and it might mean fanning a little bit medially
when you're more lateral, looking at your different zones.
So the first thing is just understand
that there is a little bit of hand skill to lung ultrasound.
Again, that's why it takes 10 to 20 scans
before you're really getting these diagnoses.
Another pitfall is, you know,
we talked about this in Dave Tierney's paper,
where he compared chest X-ray with lung ultrasound,
with CT scan as the gold standard, which is,
if you have even a thin layer of pleurae or normal alveoli,
the sound waves see that.
So you could have a really early,
really severe bronchial pneumonia
that hasn't yet spread to the periphery of the lungs,
and lung ultrasound won't see it, until it does.
For me, the anecdote that really highlights this is,
I once took care of a patient who drowned.
And as I was doing the lung ultrasound,
it was straight A-lines.
They had come straight from the river to the hospital,
and there was A-lines everywhere, and I was very confused.
And then 45 minutes later, it was B-lines everywhere,
'cause that's how long it had taken for the fluid
to get all the way out to the periphery.
And I just scanned too early.
It had filled their respiratory tree,
but not yet filled all the distal alveoli.
And so I had this very distinct false positive,
sorry, false negative,
where I saw a normal lung,
when it was clearly not normal lung,
when you look at chest X-ray or normal CT scan.
So there's that.
And then, you know, the truth,
you know, there's Occam's razor saying,
"One unifying diagnosis is the most likely,"
and then there's Hickam's dictum, that says,
"A patient can have as many diagnoses as they please."
And I see this with lung ultrasound, where it's like,
yes, some people do have volume overload
and a COPD exacerbation, and a pneumonia.
And lung ultrasound's just going to show you the one
that's layered on top.
And so the place where lung ultrasound can cause us
to fall down a pitfall is if we think that it's better
than all of our other indicators,
because there will be other indicators
that there are multiple pathologies
happening simultaneously.
So holding onto the understanding
of which direction the error is going to be,
we're going to see B-lines,
that doesn't mean
that somebody doesn't have a COPD exacerbation,
it just means that they also have an alveolar
or interstitial process.
And thinking about what their lab show,
what their fever curve shows,
all these other things that we use clinically
to decide what somebody's source of respiratory failure is.
In the ICU especially,
when somebody initially gets intubated with one problem,
and then comes up with another,
oof, we can have real issues.
That's my issue.
Those are some of the pitfalls from parenchymal ultrasound.
I talked a little bit about the pitfalls
with diaphragmatic ultrasound,
with the idea being that if you're looking
when somebody's on the vent,
you may not see any thickening of the diaphragm.
And with effusions, the challenges I've seen is,
I've seen people mistake a gastric space
for pleural effusion when it's full of liquid.
And I've seen a lot of people think they're looking
at the pleural space, when they're actually looking
at ascites, subdiaphragmatic fluid,
because that curve of the liver
is mistaken for the diaphragm,
even though the diaphragm's kind of flapping around
in the fluid.
To me, the cure for all of these latter ones
is just a really good understanding
of the clinical scenario
and a really good understanding of the anatomy
that you're actually looking at,
'cause ultrasound lets you see stuff,
but if you don't know what's supposed to be there,
then it's really hard to know how to interpret
what you're seeing.
So yeah, the first few thoughts on that.
- [Chris] That's great.
I did notice one of the attendees, Judy P.,
they had their hand raised,
but if you could put your question in the Q&A box,
I'll make sure to get that to Cameron.
And while we're waiting for that,
you talked a little bit about kind of scanning techniques
and stuff in that last answer.
What are some of the strengths and weaknesses
of the sweep technique that you outlined?
- Yeah, this is really interesting to me.
So this was proposed several years ago, actually,
as a way to allow people with very little experience
with ultrasound to do a full lung ultrasound.
And the technique is, you take the transducer,
and you just, without stopping,
record a video as you move the transducer
across the thorax in several different lines,
as opposed to doing kind of the more discrete points
recommended by Lichtenstein or Volpicelli
in the international consensus statement,
or Tironi in his nine-zone scan.
And so there's an advantage there,
you get to see more surface area of the lung,
which is super interesting.
The disadvantage is that you don't see
a full respiratory cycle of each of those points.
And for those of you who have done lung ultrasound,
you know that sometimes you place the probe down,
and it's like, "Oh, A-line pattern,"
then poof, over the course of the respiratory cycle,
the B-lines appear and disappear.
So I wonder if sweeping's going to going to allow us
to miss those, or is it going to,
through, you know, integrating it into a single picture,
allow us to very quickly see the entire lung?
The real challenge at the moment is,
and, you know, Volpicelli proposed and demonstrated,
using this during COVID,
the real challenge is that most of the evidence we have
for how to do lung ultrasound,
and more importantly, how to interpret lung ultrasound,
that scoring system by Bouhemad and others,
was done with discrete points.
And so I don't know what to make of a sweep image
without the context of all the studies
that have been done before.
So it's not that this shouldn't be done,
but gosh, we need a little bit of sweep
and point driven to understand what information is different
between those two techniques,
so we can talk about how interpretation should be different
between those two techniques.
But I have these dreams get me,
"Gosh, thinking how cool it would be where like, you know,
the respiratory therapist comes by in the morning,
checks the plateau pressures,
and does a quick sweep of the lungs
and four zones they have,
they do it in 60 seconds or less,
and that's saved to the machine.
And you can pull up the visual cue
of what their lungs look like in the same way
that you'll be pulling up a daily chest X-ray in the air
when we did daily chest X-rays, but without the radiation,
without the unnecessary equipment expense,
because this is a machine that's there, it's by the bedside,
and it doesn't cause any harm to the patient to do it."
- [Chris] Well, I don't see any other questions coming in.
And it looks like we're getting toward the end on time.
So big thank you to Dr. Cameron Baston
for taking the time today
to put together this excellent presentation
and for hanging around to answer questions for us.
Thanks so much for being here
and sharing your expertise with everybody.
- Thanks so much for having me, Chris.
This has been really fun.
- [Chris] Awesome.
Quick reminder that you can watch previous webinars
and sign up for upcoming webinars
on sonosite.com/behind-the-scan-webinar.
And you can go ahead and scan that QR code right there.
That should take you right to the page,
where you can see previous and coming up webinars.
Thanks so much for everybody for joining us today,
and we'll see you at the next webinar.
The assessment of lung disease using POCUS has exploded since the pandemic, and continues to expand with the advancement of ultrasound technology. Join Cameron Baston MD to learn about the implementation of these growing applications in Lung POCUS to help care for critically ill patients.
What You'll Learn
- Apply and recognize the clinical guidelines for Lung POCUS in general applications such as hypoxia, shock, pneumonia, and pulmonary edema.
- Apply and recognize the clinical guidelines for Lung POCUS in Critical Care applications such as pneumonia, empyema, ARDS, and volume overload.
- Refine Lung POCUS techniques to differentiate the changing disease states in critically ill patients. Appreciate the adoption and utility of the growing applications of Lung POCUS.
Cameron Baston, MD, MSCE, is a clinician advisor for Penn Health-Tech and an Assistant Professor of clinical medicine in the Department of Medicine at the University of Pennsylvania's Perelman School of Medicine. He serves as associate program director for the Pulmonary and Critical Care Medicine fellowship, and as director of clinician-performed ultrasound for the Department of Medicine. He has an interest in helping create low-cost medical devices in the resource-limited Critical Care setting and works with several organizations on POCUS and Critical Care education. A mechanical engineer, epidemiologist, Critical Care physician, and medical educator, he spends about 2/3 of his time caring for critically ill patients, and the remainder working on education innovation and health technology.
This educational webinar is intended for healthcare professionals and not for patients or consumers. The material is provided for general educational purposes, as a reference and a supplement to professional experience, education and training, and should not be considered the exclusive source for this type of information. This educational webinar is not intended to recommend any device for a particular indication or to provide indications for use for any device. At all times, it is the professional responsibility of the practitioner to exercise independent clinical judgment in each particular situation. Fujifilm assumes no responsibility or liability for any misuse of the information imparted in this webinar. This educational webinar does not supplement, replace, or supersede device labeling, including instructions for use, which accompanies any FUJIFILM Sonosite product.