Upcoming Events
Log In
Pricing
Free Trial

Rapid MRI for Acute Pediatric MSK Infection, Dr. Jonathan Samet (7-30-26)

HIDE
PrevNext

0:02

Hello, and welcome to Noon Conference hosted by Medallity.

0:06

Noon Conference connects the global radiology community through free live

0:10

educational webinars that are accessible for all, and it's an opportunity to learn

0:14

alongside top radiologists from around the world.

0:18

Today, we are honored to welcome Dr.

0:20

Jonathan Samet for a lecture entitled Rapid MRI for Acute

0:24

Pediatric MSK Infection. Dr. Samet is an associate professor of

0:29

radiology and orthopedic surgery at Northwestern University Feinberg School of

0:34

Medicine, primarily based at Ann and Robert H.

0:37

Lurie Children's Hospital of Chicago, serving as the division head of body

0:42

imaging and section head of MSK imaging.

0:45

He also works in the adult MSK radiology section at Northwestern Memorial

0:49

Hospital. He studies nerve imaging with MRI, neurography, and ultrasound,

0:54

rapid MSK imaging, and MRI techniques for bone marrow evaluation.

0:59

At the end of the lecture, please join Dr.

1:01

Samet in a Q&A session where he will address questions you may have on today's

1:04

topic. Please remember to use the Q&A feature to submit your questions so we can

1:09

get to as many as we can before our time's up.

1:11

With that, we are ready to begin today's lecture. Dr.

1:14

Samet, please take it from here.

1:17

Welcome to the Noon Talk. So today, I'll be speaking about

1:23

rapid MRI in pediatrics for musculoskeletal infection.

1:27

It's a

1:28

technique that we helped develop here at Lurie Children's in Chicago, and we've

1:32

been

1:33

really enjoying it and finding that it really has changed the way that we

1:40

work up musculoskeletal infection.

1:42

It really helps with throughput, and it's been really a game changer for us.

1:47

So I wanted to take you through the

1:50

story of how we did this, and answer some questions of why,

1:56

what, and how.

1:59

And so let's start with why should we do this?

2:02

I want to give a little background of it before we dive into it.

2:06

In pediatric radiology, a little bit different than an adult side,

2:11

we have to deal with problems with sedation and contrast

2:15

use.

2:18

We know that just by giving sedation or general

2:22

anesthesia for a lot of our exams, it can add a significant amount of time and

2:27

strain on the resource in the hospital.

2:29

We do know that there are some risks in pediatric patients of anesthesia.

2:34

And similar in pediatrics and adult, there are long-term risks of

2:38

gadolinium-based contrast use that are unknown.

2:41

But we do know that it deposits in tissues all over the body, so we really should

2:46

try to only use it if we absolutely need to.

2:49

And there are a growing number of indications for non-contrast MRI.

2:54

Every time I talk about this topic, there's one more thing that people are doing

2:58

without contrast. So in musculoskeletal, we have sacroiliitis.

3:04

We're doing an MRI for ovarian torsion now, MRI for

3:07

appendicitis. There's papers on inflammatory bowel disease.

3:12

So

3:14

it's a very exciting time for rapid

3:17

limited MRIs. So

3:20

we know that

3:22

this has been successful in other things.

3:25

So for example, in pediatrics for acute brain,

3:30

rapid brain MRIs for looking at the ventricles for hydrocephalus.

3:35

We wanted to really look at just the time burden.

3:40

We talk about all this extra time that these things take, but what is the

3:45

true

3:46

burden so we can understand the issue at hand?

3:49

And we did a little QA project at our group to look at this and really to

3:53

understand the difficulties. So what we did was we looked at all

3:58

of our MRIs, and we wanted to assess the burden of just adding anesthesia or

4:03

contrast at our institution. So we took a two-year period pre-COVID,

4:08

where we looked at just all of our MRIs, body, MSK, and neuro, that required

4:12

anesthesia or contrast. And

4:18

33% of our exams had contrast. So interesting, not terribly

4:23

surprising, a third of the exams. What we did is we said,

4:29

you went into the scanner, you had the pre-contrast, you injected,

4:34

that was the contrast administration, and then when the post-contrast phase was

4:37

over, that was the exam end. So we determined that the contrast burden, meaning the

4:41

extra time the exam takes for the contrast portion, was the time between

4:46

when you injected and the end of the exam divided by the total exam

4:51

time.

4:52

And then we also looked at what were the observed times.

4:55

So when someone says, "How long does your knee MRI?" And you say, "Oh, it's 10

4:58

minutes." But what was the actual true time, right?

5:01

So a lot of times when people tell you the times, they're really just saying, what

5:04

is the

5:06

time on the books? But we know that things take longer, patients move,

5:10

have to repeat things. So that was important for us to try to find out as well.

5:16

And then we looked at sedation burden.

5:19

So about a quarter of our exams required general anesthesia.

5:23

Some peds, it's a big deal. Sometimes conscious sedation.

5:27

And so for the sedation burden,

5:31

again, we had to find some start point.

5:33

So when they came into the preoperative area, and then there's all of that

5:38

dead time, they have to

5:41

prep the patient, talk to the patient, bring them in, set up their equipment.

5:45

So we then had an anesthesia induction when you actually put them to sleep, and

5:49

then the time out of the room. So similarly, we define the encounter

5:54

as the elapsed time between the preoperative anesthesia preparation and the time

5:59

out of the MRI room,

6:01

and expressed as proportion of the total exam time.

6:03

We wanted to have the non-scanning time.

6:05

As we know, coming in to get a knee MRI with anesthesia, it could be like going to

6:10

the airport. You got to get there early. You got to wait around.

6:13

So even though the scan itself is 20 minutes, it could have taken a really long

6:18

time. So let's get into some of those results.

6:20

So when we look at the

6:23

average contrast time burden, it was about 29 minutes.

6:27

So if you think 30 minutes

6:30

extra just to give a contrast exam.

6:33

So I'm sure all of you have read exams before and thought, "I didn't really need

6:37

the post-contrast phase. It didn't really add that much." Well, it was about 30

6:41

minutes extra. Now you say, "Well, how is that possible?

6:44

It's much faster than that." But there's a lot of dead time.

6:47

There's a lot of time that you wouldn't think.

6:48

Depends if you do

6:50

3D post-contrast imaging, whereas if you do spin echo three

6:55

planes, but it's a lot more than you would think.

6:58

So if you look at a pie chart, the contrast burden on average,

7:02

a third of the exam was attributed to contrast.

7:07

So we really should make sure we know why we're giving contrast and make sure we

7:11

really need it. When we look at sedation burden

7:14

About 56% of the exam time was attributed to the encounter.

7:19

So the whole encounter for a GA exam was about three hours and 43

7:24

minutes. So that means that the non-scanning time is a

7:29

huge chunk of it. It's about two hours and 26 minutes.

7:33

I know that we as radiologists don't often choose if someone gets

7:38

sedated or not. We are protocoling exams, and we may say knee without and not

7:42

realize that it's under full anesthesia.

7:44

But if you look into it at your place, you might

7:48

realize that sometimes there's just age cutoffs, and when they schedule it, they'll

7:51

just put them into a GA slot. And we really want to be just careful

7:56

of

7:58

our patients, do they really need GA and things like that.

8:01

So you can see this is a very big issue.

8:04

And the preliminary or the results of our just QA project show that the

8:09

post-contrast portion of the exam is about a third of the exam, adding about 29

8:13

minutes on average.

8:15

All of our contrast in extent had the observed time that were significantly longer

8:19

than expected. So if the book said it was 20 minutes, it would actually be 31

8:23

minutes of overtime. And for whatever reason, it's just difficult to

8:28

get the exam done.

8:32

84% of our sedated exams required

8:36

GA, and GA added about two hours and 26 minutes just by putting them

8:41

on a GA

8:43

exam. About 16% of our sedated exams required conscious sedation,

8:48

so conscious sedation added about 54 minutes on average.

8:53

So we really want to optimize the use of contrast and sedation to improve radiology

8:57

workflow. This is what was really kind of getting us, and we thought we got to do

9:01

something. We know that if we could possibly decrease the sedation and contrast, we

9:05

could offer cost savings for the healthcare system and get patients through the

9:09

scanners faster. So enter rapid MRI for MSK infection.

9:15

I was very fortunate to partner with someone in orthopedic surgery, Dr.

9:20

Ghibli, and we kind of really put our heads together and thought, okay, well, what

9:24

do you really need to make your decisions in orthopedics?

9:28

And then he said, "Well, what do you need as a radiologist to make these

9:31

diagnoses?"

9:33

We know that acute MSK infection, we know we're talking about

9:38

osteomyelitis, septic joint, pyomyositis, abscesses.

9:42

We know that MRI often requires sedation and contrast for these exams.

9:47

Traditionally, you have pre and post-contrast MR for osteo.

9:53

But in pediatrics, you could be waiting

9:56

for hours a day to get the anesthesia.

10:01

Patients are NPO for very long.

10:04

So we really need to develop protocols to optimize

10:08

sequence acquisition and to circumvent the need for contrast and

10:12

anesthesia.

10:15

These diagnoses frequently require hospital admission, long-term antibiotics,

10:19

operative debridement.

10:22

We know that they can cause AVN, growth arrest, deformity, joint destruction.

10:26

So it's a big deal, right?

10:29

For the surgeons, their goal is to get the diagnosis quickly so they can start the

10:32

treatment. What are they waiting for?

10:35

MRI is the gold standard. So when

10:39

someone says, "Well, how can you prove that this was osteomyelitis?

10:43

Show me the lab, show me this." Orthopedic surgeon will say, "Well, the MRI said it

10:46

was positive." So we really are kind of the gold standard.

10:49

We tell them if it's osteo.

10:52

But

10:55

we usually are traditionally doing contrast and anesthesia.

11:00

And what we used to see was that patients wouldn't get a treatment decision until

11:05

12 to 24 hours after their arrival.

11:07

They are being kept NPO for a long time.

11:11

And so

11:13

you start asking, well, why? Why are we really waiting for this?

11:17

So when you kind of simplify it, and this is where it's very helpful to think like

11:21

an orthopedic surgeon. So you have a patient who's concerned for MSK infection.

11:25

There's sort of three possibilities. MRI is negative, there's no infection.

11:29

Infection, but no fluid, and infection with abscess.

11:34

So if there's no infection, do nothing, send them home, right?

11:38

Super helpful if you can do a non-sedated, non-con exam.

11:41

They get a negative, they can just go home straight from the ER.

11:44

If there's osteo, but there's no collection, give antibiotics.

11:48

And if there's osteo and an abscess, might need IR or

11:52

operative debridement.

11:55

But how much detail do you really need to make this decision, right?

11:59

We as radiologists are always afraid we're going to miss an abscess, we're going to

12:01

miss something if you take away contrast.

12:04

But what was so helpful to me was hearing from the orthopedic surgeons, they're not

12:08

really operating on these tiny little abscesses.

12:10

If it's a one-millimeter abscess, they're not going to be operating anyway.

12:13

So we really want to keep that in mind that what we could possibly

12:18

miss is something that's not probably clinically relevant anyways.

12:22

So when you look here, you see a rapid MR.

12:26

This is one of our earlier iterations.

12:28

A regular SIR and a regular post-contrast phase sagittal knee MRI.

12:33

And everyone here can see there's a big effusion.

12:35

There's a lot of soft tissue edema around the joint. Did contrast really help us?

12:41

We see synovial hyperenhancement, but that doesn't necessarily tell you that it's

12:44

an infection, right? It just tells you that the synovial is inflamed.

12:46

So how much detail do you really need to decide what treatment this needs?

12:50

We all know that this is a pathologic joint effusion, and it's concerning for

12:54

septic joint, right? It could be just reactive.

12:56

It may be something else, but it's clearly something that is a pathologic joint

13:01

effusion. So we don't need the full-on MR to diagnosis.

13:06

So if we can get there fast enough, can we skip IV access and anesthesia?

13:10

Will this unlock more of the MRI schedule?

13:12

So we had a goal to basically develop and deploy a rapid MRI protocol

13:16

for acute musculoskeletal infection to improve speed and access.

13:21

And we know that there's many studies showing that contrast is important, right?

13:24

But let's look at some of them. So

13:27

this is a study where they looked at pre and post-contrast MRI, and they had people

13:32

kind of look at it and decide if they could still make these diagnoses.

13:35

And they kind of determined that only like a 3% of them were facilitated by

13:38

contrast. There was very few additional abscesses that were

13:42

identified with contrast, but it was very small, and they concluded that

13:46

it probably wasn't clinically significant, the ones that were missed.

13:51

We do know that if you're very young,

13:56

the unossified bone

14:00

is hard, right? It's very hard to diagnose osteo in these cases.

14:03

And so studies have shown that contrast is helpful in those

14:07

because you will see actually a counterintuitive, you will see a

14:12

dark spot, a hypo-enhancing area of the epiphyseal cartilage.

14:15

But even this, we're working on a new project

14:20

showing that you may not actually need it in those cases either.

14:24

But-

14:28

There are other papers as well showing that contrast doesn't really change the

14:33

sense of your specificity when you are taking away

14:37

contrast. We do know that it kind of helps confidence, right?

14:40

Radiologists like contrast. It makes you feel more confident about your diagnoses,

14:44

but

14:46

it might not be enough to justify all of these negatives that I just described.

14:51

So we know it can work in rapid MRI for acute head trauma.

14:54

We know we can do it for appendicitis. We do that here.

14:58

And we know in other areas for now, breast imaging as well.

15:02

So what is a rapid MSK MRI? If you ask 10 radiologists, you might get 10 different

15:07

answers. For us, we're very strict when we define it.

15:10

It's no contrast, no sedation.

15:13

Yes, some patients may slip through the ER and get a little intranasal versed, but

15:17

basically, we're not having anesthesia come down, and we are not giving contrast.

15:24

So

15:26

before we get into the pictures of what it looks like,

15:28

I think of a traditional MRI as the Mona Lisa.

15:31

So a very sharp image, beautiful image, sedation and contrast, patient's not

15:36

moving. They're under GA. Great image.

15:40

You do a rapid, more like a Seurat, okay?

15:43

Little dots, but you can all see what's still going on in the picture.

15:47

Might not be as good of a scan. Now, as we've developed the protocol and the scan

15:52

has gotten better, the rapid sequences are actually pretty good.

15:56

I don't think that they're any that much worse than a pre-contrast MRI, but that's

16:00

the general concept. So for us, it's no sedation, no contrast,

16:05

and we're using fewer sequences, and we're optimizing them for speed and fluid

16:09

detection.

16:11

So we start off, and I'll use the lower extremities as the main example when I give

16:14

these

16:16

examples, and that's the most common that we see.

16:19

We start off with a large field of view, a coronal STIR HASTE.

16:23

We're looking bilateral thighs, bilateral lower legs.

16:27

So if it's non-localizing symptoms, we're just going to do uppers and lowers

16:31

to cast a wide net. Before we started doing this, I had a case once where we

16:36

did a full-on GA pre/post-contrast knee MRI in a young five-year-old.

16:41

It was negative. They went back up to the floor, came back a day later.

16:44

Osteo was in the hip. So it just seems silly to not catch

16:49

that in the beginning. So you've got to have a large field of view.

16:51

And then, once you detect the area of pathology,

16:56

you'll do the small field of view images.

16:58

So we do coronal, sagittal, axial, T2 fat suppressed.

17:02

A lot of times the ED will know that it's the knee, so the large field of

17:06

views are just extra. In the case of a non-localizing, a truly non-localizing

17:11

patient, the tech might call us and say, "Hey, where do you think we should put the

17:13

field of view?" And then we'll tell them the area.

17:16

Usually, it's often obvious, and they can

17:19

choose the area themselves. We'll always throw in a T1 for the marrow.

17:23

And then instead of using contrast, we found it's very helpful to use

17:26

diffusion-weighted imaging in order to detect abscesses.

17:31

So here's an example. We do our large field of view, and it's

17:36

non-localizing symptoms. We're not sure what it is.

17:38

We just have a patient coming in with a limp.

17:39

And everyone can see here there's something going on in the right proximal tibia, a

17:42

lot of edema there. So, that will point us to focus exam on that

17:47

area.

17:49

Different patient showing you what some of the rapid sequences will look like.

17:52

So here we can see a large pathologic joint effusion.

17:57

We don't need post-contrast to tell us that this could be a septic joint, right?

17:59

This is a big effusion, and that should be enough.

18:03

This is actually two things. So we have, this is a different patient.

18:07

We have an intramuscular fluid collection.

18:11

This was an abscess. And then you can actually see a subperiosteal abscess of the

18:16

fibula. So it's a very small subperiosteal.

18:18

So people get concerned they're going to miss subperiosteals, but I actually like

18:21

it on T2 weight imaging better than a post, and I'll show further examples.

18:25

You're basically seeing the black line of the periosteum lifted up underneath white

18:29

fluid. So it's actually the contrast of that is very helpful to see that there's

18:34

something that shouldn't be there.

18:36

Another example of osteomyelitis of the fifth metatarsal.

18:39

You can see the marrow edema, and then you can see this

18:43

circumferential fluid collection representing the subperiosteal abscess

18:47

around the bone. Another example.

18:52

This patient happened to come in with unknown

18:58

subacute to chronic osteo. We usually reserve this for acute osteo, but as you can

19:02

see here,

19:03

there's a lot of sclerosis, and this little tiny linear thing is actually a bony

19:07

sequestrum in this patient with more subacute to chronic osteo.

19:11

On the sagittal T2 fat-suppressed image over on the left, you can see this

19:17

lobulated subperiosteal abscess. So don't contrast to see that.

19:21

You can see the T2 bright thing. The periosteum is lifted off the bone.

19:26

Here's how we use diffusion-weighted imaging.

19:27

This is not a stroke of the muscle as we know.

19:30

This is an intraosseous abscess because abscesses, as you

19:34

know, in other parts of the body will have restricted diffusion.

19:37

So we found this very helpful. This is a bright signal on the DWI,

19:42

dark on the ADC, and this was how we used diffusion-weighted imaging instead of

19:46

contrast.

19:49

So what are the diagnoses that we want to make sure we can make with the

19:54

rapid? So we're looking for septic joint, osteomyelitis, subperiosteal

19:58

abscess, intraosseous abscess, and soft tissue abscess.

20:03

So you're going to ask yourself, but can we still make these diagnoses without

20:06

contrast? Radiologists are very, very attached to their contrast, and

20:11

they get concerned if you take it away, "I'm going to be missing things." So let's

20:14

go through some more cases. So here is another

20:19

rapid MRI. So we have axial and coronal T2 fat sat, and you can all see here that

20:23

there is a small to moderate size right hip joint effusion with

20:28

periarticular soft tissue edema. Also, a little bit of marrow edema in the femur.

20:32

So this is a pathologic joint effusion. This was a septic joint.

20:36

We don't need contrast. Contrast will not tell us if this is infected.

20:39

It might show synovial hyperenhancement, but that doesn't really tell you

20:44

anything more. So, the non-con is good enough to tell you this is a pathologic

20:48

joint effusion, and this was a septic joint.

20:52

Here's an example of osteomyelitis.

20:54

You can see a coronal T1 and a coronal T2 fat sat.

20:57

Everyone can see low on T1, high on T2, edema around the bone.

21:02

This is osteomyelitis. If I gave contrast, I might see enhancement in the areas

21:05

that were T2 bright. Doesn't necessarily tell me anything more.

21:11

Here is another case of a subperiosteal abscess

21:16

of the distal femur. What you can see here is the periosteum lifted off the bone

21:21

surface, this black line here. It tapers and sort of

21:25

terminates at the level of the perichondrium. This is a fixed point.

21:29

And you're seeing this nice

21:32

lentiform collection. So, you know this is a subperiosteal collection, and

21:36

in the setting of osteo, this is a subperiosteal abscess.

21:43

We got lucky and we also were able to see this on the diffusion-weighted imaging.

21:47

So we see bright on DWI, dark on ADC.

21:50

And so this is a restricted diffusion in the subperiosteal abscess, helping you be

21:53

more confident that you're dealing with an abscess.

21:58

How about a small subperiosteal? People say, "Well, no, you certainly can't see a

22:01

small one." Here is a small one. So you have osteo of the distal tibia,

22:06

and a little thin sliver of bone here. They might not even go in and drain this.

22:10

It's so small, but you can see here the black periosteum is lifted up, and you see

22:14

small fluid underneath it.

22:20

Here's a small intraosseous abscess.

22:22

So T2-weighted imaging,

22:24

a focal rounded area of T2 bright signal, a little bit more than just osteo, so we

22:29

call this a small intraosseous abscess, Brodie's abscess.

22:34

Diffusion-weighted imaging is trickier in bone, but this just happened to be a nice

22:37

example showing

22:39

that that abscess actually was bright on DWI and dark on ADC, so it had restricted

22:43

diffusion to help you be more confident that you're dealing with an abscess.

22:50

Here's an example that I think where the diffusion can really help you.

22:53

So you're looking at the T2 and you're thinking, is this just complement edema,

22:57

myositis? I don't really see a rounded collection. Is this an abscess?

23:01

But then when you put the diffusion-weighted imaging, watch what happens.

23:05

Beautiful, bright, restricted diffusion.

23:09

Everything else is dark, and you can clearly see that this is a pus pocket.

23:14

And so, I really like the restricted diffusion for an alternative to

23:19

contrast. This patient happened to get a post-contrast MRI

23:24

at a later point, a different setting.

23:27

And this, of course, is what it would look like on the post-contrast, the rim

23:31

enhancement.

23:35

Here are two different patients. So here's the subperiosteal abscess of the femur

23:40

that I showed you before. This is a different patient showing the subperiosteal

23:43

abscess of the tibia. And I like to try to make a convincing argument, hopefully,

23:47

that

23:48

subperiosteals are actually easier to see on T2.

23:52

If you're a little bit less experienced and you don't read MSK as much,

23:55

you're looking for, it's like a bright thing on a dark thing, whereas on a

23:59

post-contrast, it's dark and dark.

24:02

So it's hard to see this if you're not really looking for it.

24:06

So, I actually think on T2, subperiosteals are really nice to see.

24:12

So I spent this time convincing you by imaging.

24:15

We did do a study on it, and we put an article in Journal of Bone and

24:19

Joint, JBJS Orthopedic Journal, just talking about our experience.

24:25

So we have been doing this for five years.

24:26

So this data was from our first couple of years.

24:30

So we implemented the protocol and we

24:35

compared a traditional cohort, so pre-COVID era patients who got regular

24:39

MRI contrast and sedation, and we had our so-called rapid cohort where we had all

24:44

patients with no sedation, no contrast.

24:47

We used patients who were coming in for acute infection.

24:50

We focused on ED patients and inpatients. We used this for extremity and pelvis.

24:55

We excluded outpatients, spine, chronic osteo,

25:00

other non-infectious causes.

25:03

We're not using this for oncology or rheumatology.

25:07

And what we did is that we compared before and after the implementation.

25:11

We looked at some MRI-related outcomes.

25:13

We wanted to see, are they getting into MRI faster?

25:16

Are they

25:18

getting shorter MRIs, hopefully? Are they getting their reads faster?

25:22

And so we also looked at some failures of the MRI.

25:25

So how often did we have to basically abort because the patient was

25:29

moving? Were there false negatives? Are we missing stuff?

25:32

That's what everyone's afraid of. How often were we repeating the scan?

25:36

So if someone says, "You're just going to repeat it anyway, you're wasting your

25:38

time." So let's look at that. And then we looked at some final diagnoses.

25:42

We looked at how many operative interventions, IR interventions, hospital length of

25:46

stay.

25:48

And we had a pretty nice mix of cases.

25:51

So we had similar amounts of both cohorts.

25:54

We had similar amounts of each diagnosis, osteo, septic joint.

25:59

Similar amounts of interventions, surgeries.

26:03

Obviously, the ones that had the traditional MRI were going to have much more

26:07

amounts of contrast and sedation on board, whereas the rapid didn't really have any

26:11

contrast or sedation.

26:14

So what we found, rapid being in orange and traditional in blue, you can see

26:19

here that it was just a lot faster. So when you order the MRI

26:24

to time to get the MRI was faster. Because you're not waiting for anesthesia,

26:27

you're not waiting for contrast. Just bring them upstairs, put them in the scanner.

26:31

We got our final reads faster.

26:34

The scan was shorter, obviously, because we are doing faster sequences and

26:39

not doing contrast. So the imaging outcomes were really nice.

26:44

We actually had patients with a less length of stay

26:49

in the rapid MRI. So they were in the hospital quicker.

26:52

Part of that is because if you're negative, if you

26:57

have no infection, you can send them home.

26:59

If you

27:01

really have a problem, if you had a lot of positive cases, a positive scan, you're

27:06

going to be in the hospital. But a lot of the power of this is the

27:11

negative as well. Just send them home if they're negative.

27:14

We found that hospital charges were less

27:18

because we were using less resources.

27:21

And so,

27:23

our study highlights were that our sedation rates plummeted,

27:27

contrast rates went to zero,

27:29

MRI rates were

27:31

much shorter. We got the reads in faster.

27:35

And for orthopedic surgeons as well, it's

27:38

really nice for them if they get a scan in the evening, they have a result at 1:00,

27:42

2:00 in the morning, they can plan for the OR first case.

27:46

So that's actually very helpful in talking to orthopedic surgeons.

27:50

Hospital length of stay reduced almost two days.

27:54

We reduced the hospital charges. So, what about failures?

27:58

So sometimes the exam was canceled or ended early due to patient motion, about 15%

28:03

of the cases, but

28:05

still the vast majority were able to complete the exam.

28:08

So you're not really giving contrast, you're not giving sedation, so you're not

28:11

losing that much.

28:13

If most of the patients far and away succeed, I still think that's actually worth

28:16

it to do it.

28:20

Of the ones that basically ended due to motion,

28:24

it turned out they watched the patient.

28:26

Only 9% of those required a repeat with sedated MRI because the remainder of them

28:31

clinically improved and they were like, "Yeah, we don't need the MRI anymore." So,

28:36

it was even less when you look at it that way.

28:39

We had two missed cases, and I put them in quotations because I don't really truly

28:44

think they're misses, they're just hard cases.

28:47

So the first one was a patient who had a rapid.

28:50

There's a very subtle soft tissue edema along the left

28:55

pelvic rim

28:57

We called it negative, but I don't know if anyone would even call that.

29:00

It's just a hard case. Came back two days later, and there was a tiny little

29:05

subperiosteal there, but I think it's just a hard case.

29:08

Another one,

29:10

it was a six-month-old, so a few things were wrong with this.

29:12

We typically don't do rapids

29:15

this young.

29:17

When we started the implementation, we would say, "Okay, if you're above 18 months

29:21

old, we'll do it. But if you're below 18 months, forget it."

29:25

So they did this on a six-month-old. Also, they just did the large field of images.

29:29

They didn't do the small field of view.

29:31

So this tiny little area, I don't even think there's anything there.

29:36

But on the follow-up, it was a little more prominent.

29:38

So that was one of the other ones. But there weren't any big abscesses that we sent

29:42

home or obvious osteos. No one really came back with osteo that we

29:47

had. So when you talk about deployment,

29:52

we're very focused as radiologists on

29:56

the scanner, the technique. Everyone wants to know what's the secret

30:01

sequences they use. But think of it more, this is not just a protocol, this takes a

30:05

team. You're really going to need to collaborate with, first of all, your own

30:10

department. So your radiologist, to convince your own radiologist that this is safe

30:14

to do, that you can do this. But you're also going to talk to your ED, talk to

30:19

orthopedic surgery.

30:21

Figure out what do they truly need to make a diagnosis.

30:24

Are they comfortable with this? Talk to ID as well.

30:28

It is a practice change. So it takes

30:32

a whole group of people to discuss this.

30:35

And one thing to stress is appropriate use.

30:38

So

30:40

we named the protocol Rapid MSK Infection.

30:43

So that's what it says when we log into PACS.

30:46

And that just reminds everyone that we're not using this for meniscal tear, we're

30:50

not using this for tumor. And so that's how you can really make sure that

30:54

the exam is used appropriately.

30:58

In the beginning, we monitored it. We would monitor every case.

31:02

It's nice to have your own radiology champion that can

31:07

do this. So, I made myself available to my colleagues who were on call.

31:11

We have people who aren't MSK trained who are taking call.

31:15

Of course, they're going to be reading rule out osteos and so they might text me,

31:20

"Hey, what do you think about this case?" And I'll look at it and just give them a

31:25

little bit of reassurance. And then after a few weeks, people were okay with

31:29

it.

31:33

And so, we created a little flowchart

31:37

that we shared with the ED to help them decide when to do

31:42

this. So,

31:44

in general, we're just trying to

31:47

help with

31:49

appropriate use. So if there's a concern for acute infection,

31:53

just reminding people that this is not for tumor, trauma, or arthritis.

31:57

If yes to that, then we say, "Is it a chronic process?" If it was chronic, we'd

32:01

say, "Okay, just do a traditional MRI. This could be complicated.

32:05

Have you had an X-ray?"

32:08

And then in the beginning, we would tell them, "We're trying this new thing out.

32:12

There's a possibility that we might

32:15

need to repeat this and do a full MRI anyways, but at least we can try to do

32:20

this unsedated and without contrast." And most people would say yes.

32:24

In the beginning, as I was saying, we didn't offer it

32:28

for under 18 months. And that is because of the concern for

32:33

this unossified cartilage and whatnot.

32:35

So we kept it for patients who were over 18 months.

32:38

So in conclusion,

32:40

our rapid MSK MRI effectively eliminated sedation and contrast, decreased the time

32:45

to MRI, the scan length, and the time to the final read.

32:49

We decreased the hospital length of stay and the charges.

32:52

We had a less than 10% rescan rate.

32:56

And remember that deployment requires a multidisciplinary collaboration, and it's

33:00

really been a game changer for us here at Lurie Children's, and

33:05

we are able to just give the clinical teams what they need to

33:09

make their decisions and, as you know, it often hinges on the MRI.

33:13

So, it's been great and some of you may already be doing this, but for those

33:18

who haven't, hopefully this talk

33:21

gave you some more confidence and some more information on how to get it started.

33:25

So, I am going to show my references here and then say thank you to you

33:29

all and open it up for questions for the group,

33:34

and

33:36

see if we have any comments or questions. So thank you very much.

33:44

Awesome. Thank you so much for that lecture, Dr. Salmon.

33:49

At this time, we will open the floor for Q&A.

33:53

You may submit your questions through the Q&A feature.

34:00

And I see

34:03

one question so far here.

34:06

Do you use any sort of anxiolytic or minimal sedation with your rapid protocol?

34:11

So we let the ED do

34:15

what they feel appropriate. So sometimes they'll give patients a little anxiolytic

34:20

or something before they come up. We don't say, "No, you cannot do that." So

34:25

it's really the main

34:29

time problem was having a whole anesthesia team and keeping them NPO.

34:34

So if they want to give them something, that's fine.

34:37

We don't monitor what they give them.

34:41

I am going to try to find the Q&A. I'm in the chat one now.

34:45

Oh, here it is. I see it. Okay.

34:49

Yeah, it's a little bit hidden.

34:51

Got it. Okay. I'm going to read off these questions and try to answer them one by

34:55

one. What's the conscious sedation?

34:59

Conscious sedation is just

35:02

basically Versed or something like that, just something that's not a full general

35:07

anesthesia. So they're just a

35:10

little bit out of it, but it's not general anesthesia.

35:15

Next question.

35:17

What is the charge of fast MRI versus conventional?

35:20

So this will depend on your hospital and where you practice.

35:25

When we set it up, we felt strongly that

35:30

if we're making clinical diagnoses that will affect actual management, then it

35:34

should be a full charge even though it's

35:38

slightly faster. So we still just charge it as a full charge.

35:42

It is a non-contrast exam, so it's going to have

35:47

less of a charge than with or without, of course.

35:52

But yeah, so we do a full charge.

35:56

Next

35:58

question. When do you follow the MRI with radionuclide infection scanning?

36:04

So we in Chicago at Lurie Children's,

36:08

we

36:10

Don't typically do nuclear medicine for

36:14

infection anymore because the MRI is good enough to diagnose

36:19

most of the

36:21

cases that we have.

36:24

Next question, what are the medications used for conscious sedation?

36:28

Most of these are crying and lots of movements.

36:32

So, I'm not the one giving the medicine.

36:36

It would be the ED, and I think often they give a little intranasal Versed, but

36:41

as far as movement,

36:44

you'd be surprised, a patient who's

36:47

sick,

36:49

not feeling well,

36:51

they actually hold still a lot more than you would think.

36:54

And so

36:56

we have found that it's actually very successful.

37:00

Next question, how many of these fast exams are you doing per day or per

37:05

week?

37:08

So,

37:09

hard to know exactly. I think we probably get

37:15

maybe one a day or something.

37:19

It really depends. It goes in waves.

37:23

That'll come to the ER.

37:26

Okay. Next question. What resources, books, et cetera, do you recommend for

37:31

follow-up imaging? It is difficult to find out what kind of changes is normal

37:35

in particular timeframe or not.

37:38

So, as far as rapid MSK infection,

37:42

I actually put out an article in, not to pat myself on the back, but we did

37:47

give our experience and we put a review article in pediatric radiology.

37:51

If you look for my name, Samet, and you can see we put out an article on what

37:56

the

37:58

infection looks like on our rapid MRI protocol. So go look that up.

38:03

Next question. Have you used AI to speed up exams?

38:07

So,

38:10

we are using a number of different

38:15

techniques

38:16

with our new scanners.

38:19

The DeBoost, Resolve, all this stuff, parallel imaging,

38:25

just trying to go to the scanner and try to-- When we first

38:29

developed it, it was with the technologists and the physicists and trying to just

38:35

limit quality a little bit, but you said that you could still see, and there's

38:39

certain things that they can do as well to tweak it.

38:41

But the newer scanners are so fast anyways, with parallel imaging and

38:46

all this kind of stuff. So, yeah. But AI,

38:50

yeah, could definitely help in the future for sure.

38:54

Can we use STIR instead of T2 fat sat?

38:56

Yes, you can definitely use STIR instead of T2 fat sat, whatever you're comfortable

38:59

with. I have found

39:04

even just doing a

39:08

non-contrast study,

39:14

the time savings is...

39:20

Removing the anesthetic component, the NPO, that's the huge bulk of

39:25

the time. So if you change one minute here, two minutes

39:30

there,

39:32

it's a drop in the bucket compared

39:36

to

39:39

actually removing the general anesthesia and the contrast.

39:44

So,

39:46

there's that.

39:48

Let's see.

39:53

Any other questions about

39:58

how to-- Any barriers people are worried about facing?

40:01

We usually get questions on

40:04

barriers we might face.

40:06

It is hard to people who are used to reading pre- and post-contrast

40:11

MR.

40:12

It can be tough at first, because you're taking away things that

40:17

might help you interpret these studies.

40:20

But

40:22

hopefully, this talk gave everyone a little more confidence that you can still make

40:25

these diagnoses well. And as long as everyone is on board

40:30

that this is a new process and

40:33

you're talking to your orthopedic surgeons, talking to ED, it's a huge game

40:35

changer. Will get your patients in and out much faster and less

40:40

invasively.

40:43

Okay.

40:46

Let's see here.

40:50

Any other questions on the rapid MSK infection

40:55

protocol?

40:58

Let's see here. Oh, I see another one here.

41:00

Do you have childlike specialists or other support to help the child through the

41:05

MRI? Good question. So, we do. They're very hard to come by,

41:10

as you know.

41:11

They're amazing. We do have them during the daytime

41:16

to sort of float around the MRI department, help us out.

41:20

In the evening and off hours, it's much harder to have someone.

41:23

But we're at a pediatric center, so our technologists are very

41:28

comfortable interacting with children and really helping them, and they've learned

41:31

a lot over the years of how to get them through these exams.

41:35

But yes, child life, if you can get it, they're awesome.

41:40

Let's see here.

41:44

Looks like you knocked out all those questions pretty quick.

41:46

I got them all. Okay, wonderful.

41:47

Awesome job.

41:50

Yeah.

41:50

Dr. Samet, thank you for that lecture today,

41:55

and thank you to everyone for participating in our noon conference and asking such

41:59

great questions.

42:01

You can access the recording of today's conference and all our previous noon

42:04

conferences by creating a free account.

42:07

We'll also email out a link to the replay later today.

42:10

Be sure to join us next week on Thursday, August 6th at 12:00 pm Eastern,

42:16

where Dr. Justin Stole will deliver a lecture entitled

42:20

Imaging Considerations after Gender-Affirming Therapies.

42:24

You can register for that at modality.com and follow us on social media for updates

42:28

on future noon conferences. Thanks again and have a great day.

Report

Faculty

Jonathan Samet, MD

Division Head, Body Imaging Section Head, Musculoskeletal Imaging Department of Medical Imaging Ann & Robert H. Lurie Children's Hospital of Chicago Associate Professor of Radiology Northwestern University Feinberg School of Medici

Ann & Robert H. Lurie Children's Hospital of Chicago, Northwestern University Feinberg School of Medicine

Tags

Musculoskeletal (MSK)