0:02

Hello, and welcome to Noon Conference hosted by Modality.

0:05

Noon Conference connects the global radiology community through free live

0:09

educational webinars that are accessible for all and is an opportunity to learn

0:13

alongside top radiologists from around the world.

0:17

Today, we are honored to welcome Dr.

0:18

Alejandra Bedoya for a lecture entitled "Normal and Abnormal MRI

0:22

Appearance of the Pediatric Bone Marrow." Dr.

0:25

Bedoya completed her radiology residency and MSK subspecialty training

0:29

at the University of Pennsylvania.

0:31

She did her pediatric radiology and fetal subspecialty training

0:35

at Boston Children's Hospital, Harvard Medical School, where she stayed as faculty

0:39

for four years. Currently, she is a pediatric radiologist at SickKids

0:43

University of Toronto. At the end of the lecture, please join her in a

0:47

Q&A session where she will address questions you may have on today's topic.

0:51

Please remember to use that Q&A feature to submit your questions so we can get to

0:55

as many as we can before our time is up.

0:57

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

1:00

Bedoya, please take it from here.

1:02

Thank you so much, Ashley, and thank you so much for the invitation.

1:05

It's a great pleasure to be here today talking about normal and

1:09

abnormal bone marrow assessment in children.

1:13

I don't have anything to disclose.

1:16

So we're going to start with a case.

1:18

This was a case that I had during fellowship, and I think it describes

1:22

why this topic is so important. So we have a

1:26

four-year-old male that was transferred from another country with a

1:29

diagnosis of severe juvenile idiopathic arthritis,

1:33

anemia, and subdural hematomas. So the history is kind of

1:37

bizarre.

1:38

We were consulted by rheumatology to review the prior

1:42

studies and the history that we got. This was an international transfer.

1:46

The history that we got is that he had one year of fever, generalized

1:50

pain. He had a hip MRI. After that hip MRI, he was

1:54

started on prednisone and methotrexate. He developed papilledema.

1:58

He got a CT scan, and six months later, he had a forehead nodule,

2:03

and neurosurgery was consulted. So the history is atypical.

2:07

So these were the first radiographs from 2019, and

2:11

they were relatively normal, no focal abnormality seen.

2:15

And this was the concurrent MRI that was performed at that time.

2:19

So as I said, this was one year before he presented to us.

2:23

So we have a coronal T1, we have a coronal T2 FAT sat in two

2:27

different sections. And I want you to look at the images and

2:30

think about what is the principal diagnosis in this

2:34

MRI. So is it bilateral joint effusion?

2:37

Is it septic arthritis of the right hip, left-sided

2:41

peritendinar myositis, or bone marrow infiltration?

2:47

So I want you all at the end of this presentation to be confident

2:50

100% that this is bone marrow infiltration, diffuse bone

2:54

marrow infiltration. So let's talk a little bit about

2:59

what happened to this patient, unfortunate case.

3:02

So he had an ultrasound of the bilateral

3:06

knees, and unfortunately, this was also described as knee effusion.

3:10

As I said, we don't know. As this is an international patient,

3:14

unfortunately, based on the first MRI, he was started on steroids and

3:18

prednisone.

3:19

This was the ultrasound of the knees, and something was measured here,

3:23

and there was reported bilateral knee effusion.

3:27

But we need to remember that when we do ultrasounds in pediatric, it's

3:30

different than adults. Pediatric patients have a lot

3:34

of epiphyseal cartilage, which is like this is an MRI of

3:38

a similar age patient, and you can see that the epiphyseal

3:41

cartilage is relatively big. And on ultrasound, the epiphyseal

3:45

cartilage looks completely anechoic.

3:48

How can we differentiate it from joint effusion?

3:51

The good thing is that ultrasound is a dynamic study, so we can do compression to

3:55

see if the fluid moves. We can actually move the joint or the knee to

3:59

see if this is actually fluid or if it's just an anechoic epiphyseal

4:03

cartilage. But in this case, you can see the normal femoral

4:07

epiphysis. So this is like a longitudinal view of the knee.

4:10

You can see the tibial epiphyseal cartilage, the physis, the

4:13

epiphysis, and the metaphysis in this area.

4:17

So unfortunately, in this case, the important findings were

4:21

missed. This is non-effusion.

4:23

And unfortunately, as I said, this patient was started on prednisone.

4:28

So we have CTs from September 2019 and June 2020.

4:32

And on the subsequent, on the last CT

4:36

from 2020, unfortunately, this was reported as subdural

4:39

hematomas. But the truth is that the most important finding was missed,

4:44

and it was the model appearance of the bones.

4:47

It looks really regular. It looks really atypical.

4:49

And we can see these forehead nodules. So these are extracts.

4:53

So these are external. So this could not be

4:56

subdural hematomas. These are actually extruose

5:00

of tissue masses. So when he presented to us on admission, we can see

5:04

that the radiographs now are abnormal.

5:06

There is a diffuse,

5:08

patchy marrow infiltration process.

5:11

We can actually see on the CT that there is mass

5:15

in the left pelvic, the retroperitoneal region at the Zuckerman

5:19

organ with metastatic lesions in the bilateral

5:22

supraclavicular nodes. And unfortunately, the whole bone

5:28

appearance is abnormal, and he developed AVMs of the

5:32

proximal

5:34

humeral heads as well as the femoral heads.

5:37

It's unclear if this is secondary to the primary disease or the high

5:41

doses of steroids. So this case is

5:44

actually a case of neuroblastoma.

5:46

So he underwent an MIBG in our institution with

5:50

the highest score we have seen, 29 over 30.

5:53

So this MIBG scan looks like a bone scan, but

5:57

all these are just metastatic osseous disease.

6:01

So the reason why I'm showing this case is because I think we can make the

6:05

diagnosis with the first MRI, and throughout this

6:09

talk, I'm going to give you tips on how to make the diagnosis confidently.

6:13

So neuroblastoma, in this case, is responsible for 15% of

6:17

cancer-related children deaths.

6:20

So he was a four-year-old. So every time we see a bone marrow infiltration

6:23

process, we have to think about lymphoproliferative disorders such as

6:27

leukemia or lymphoma, and also neuroblastoma, which can have

6:31

diffuse bone marrow infiltration, and it's a pretty common

6:34

finding in children of diffuse metastatic

6:38

disease. So let's just start with the basics.

6:41

What are the different types of bone marrow, and why do we care

6:45

about this? We care about this because children, you're going to

6:48

see the normal process of conversion from red marrow to

6:52

yellow marrow throughout the skeleton.

6:55

And we need to remember that histologically, the red marrow is

6:59

going to be the hypercellular, the one that is in charge of

7:03

hematopoiesis. But we need to see, and I'm

7:06

highlighting this with a star, is that it has 40% fat.

7:11

So this is extremely important because the

7:15

only role of MRI is differentiating between

7:18

fat and water. So if we know that red marrow has

7:23

40% fat, we will be able to differentiate normal

7:26

red marrow from muscle or from a bone marrow infiltrative

7:30

process that is going to get rid of that fat.

7:33

Of course, the yellow marrow or the fatty marrow is going to have 80% fat

7:37

and is going to have a more nutritional support,

7:41

a goal in the bone marrow with limited

7:45

vascularity. So we need to remember there are two types of red

7:49

marrow. And something important that we need to remember is that there

7:53

is a normal bone marrow

7:56

conversion, which is the transformation from

7:59

red marrow to yellow marrow. So this bone marrow

8:03

conversion happens in a predicted way,

8:07

and you have to know it because an abnormality of this normal

8:11

process is pathologic. So the first important

8:15

rule is that the red marrow converts from

8:18

distal to proximal, and that's why in the axial

8:22

skeleton, for example, in the spine, you can still see red

8:26

marrow, particularly in women, because there is a higher need of

8:30

hematopoiesis. And it's a normal finding

8:34

to see a little bit of red marrow in the proximal femurs or in the pelvis or in the

8:38

spine. So this is from distal to proximal.

8:41

But also there is a predictable bone marrow conversion

8:45

within the long bone. So within the long bone, and I

8:49

don't know if you can see my video, but I always say that the bone marrow

8:52

conversion is like epiphysis, diaphysis, metaphysis.

8:56

Epiphysis, diaphysis, metaphysis.

8:57

So first it's going to be the proximal

9:00

epiphysis, then the distal epiphysis, then

9:04

the diaphysis, and finally the metaphysis with a little bit of

9:08

residual red marrow in the proximal metaphysis.

9:12

So why do we care about this? Because in a child,

9:16

you're going to have, for example, first the development of your

9:20

distal aspects. So for example, a neonate or a newborn is going

9:24

to have your distal secondary ossification center in the femur, but not the

9:27

femoral head. So the first

9:31

rule is predicted

9:35

conversion, distal to proximal. The second rule is epiphysis,

9:39

diaphysis, metaphysis. And the third rule that is really important is that

9:43

at the beginning of the appearance of the secondary

9:47

ossification center in radiographs, you're going to have red marrow.

9:52

But after six months of the radiographic appearance, that

9:56

red marrow is going to turn into yellow marrow.

9:59

So why do I say this? It's because each bone is going to

10:02

have a different time where the secondary ossification

10:06

center is going to appear.

10:09

So for example, in the femoral head, as we know, we do

10:13

ultrasound assessment of the hips in patients younger than six months.

10:18

And usually after six months, we do hip

10:21

assessment for hip dysplasia with radiographs because we

10:24

have a secondary ossification center in the femoral head.

10:28

So at the beginning, it's going to be red marrow, and then six months

10:32

later, it should be

10:35

yellow marrow. So the best sequence

10:39

to evaluate the marrow

10:42

is a T1 without a fat sat. So the

10:46

routine T1. So if you are doing a pediatric

10:50

MSK protocol, it should always have a T1 sequence.

10:54

So why is the T1 so important? The T1 is important

10:58

because the normal red marrow, as you can see here, for example, in the

11:02

proximal femur in a five-year-old, is going to have higher

11:06

signal intensity than the adjacent muscle.

11:10

And this is the fifth rule. Because you're going to have

11:14

40% fat. So that's how you know you have a normal red

11:18

marrow. So one is distribution, and the other one is signal intensity.

11:22

So here we can see with different age ranges how the

11:26

normal bone marrow conversion happens.

11:28

So it first happens in the knees compared to the hips.

11:33

And you can see in a one-week-old how we pediatric

11:36

radiologists, we struggle to evaluate the marrow in

11:40

neonates just because when they are born, they don't follow the

11:43

rules, and they have so much red marrow, and they need

11:47

so much hematopoiesis that even your normal red marrow is going to be

11:51

darker or in lower signal intensity than the adjacent muscle.

11:55

And that's why it's so hard to identify osteomyelitis in these babies.

12:00

But after that, it's really easy to differentiate what is red

12:04

marrow and what is yellow marrow. So you can see that at the beginning, you don't

12:07

have a secondary ossification center, but then at one year of age, you

12:11

should have your normal femoral head with

12:15

yellow marrow.

12:18

This is an MRI of a fetus, like a T1 5, and you can see

12:22

how much

12:24

red marrow is in that diaphysis. This is the distal

12:29

femoral epiphysis that is not ossified.

12:32

Actually, it's like one little secondary ossification center is forming,

12:36

and your femoral head is not ossified.

12:38

But all this is just really dense

12:41

hematopoietic red marrow. So the signal intensity in the

12:45

bone marrow is important, as I just described.

12:48

And for me, the T1 sequence is the most important.

12:51

So the T2 sequence is going to help you to--

12:55

Your eyes are going to go to the abnormality, like high signal on

12:59

T2, but the T1 is a sequence that is going to

13:03

tell you if it's aggressive, non-aggressive,

13:07

or normal red marrow. So remember, T1 red

13:10

marrow, when we compare to the muscle, it should be equal or higher

13:14

signal, just because it has 40% fat.

13:19

The T2 is going to be bright and the red marrow usually enhances.

13:24

The yellow marrow, of course, is going to have high signal on T1, low signal on

13:27

T2, just because it has a higher fat component.

13:30

But as you can see, the red marrow compared to pathologic

13:34

infiltration and edema, the T1 is going to give you the

13:38

clue if it's normal or abnormal, because edema and pathologic

13:42

infiltration is going to have lower T1

13:45

signal when compared to the adjacent muscle.

13:48

So a normal muscle should have 0% fat, so you

13:52

should compare with a normal muscle.

13:54

And if you don't have, for example, in a spinal MRI where you don't have your

13:58

muscle to be able to use as a comparison tool, you're going to

14:02

use the intervertebral disc as your

14:07

factor to decide if it's higher signal or lower signal, but it should

14:11

be higher signal.

14:14

So chemical shift imaging is another tool that we can use to

14:17

evaluate marrow signal,

14:20

and fat presence within the red marrow.

14:23

So normal red marrow and yellow marrow should have drop of

14:27

signal in the out-of-phase, just because of the microscopic

14:31

fat. But the truth is that in tiny babies, this is not going to be

14:35

useful, and I didn't use it in

14:38

neonates or younger kids because there is so much red marrow that

14:42

it's just not useful. So this is, for example, a neonate, where the

14:45

in-phase signal actually looks that it has lower signal

14:49

intensity than the out-of-phase. That doesn't mean that is abnormal.

14:52

We just don't use this in pediatrics as much.

14:56

So if we go back to the case one, the case that started all this

14:59

conversation, and if we apply all the rules, we

15:03

can see how abnormal this first

15:07

image is. And you only need the T1 to make your diagnosis in this case.

15:11

So, for example, as we said, the femoral head should

15:15

appear at six months of age. So that means that at one

15:19

year of age, you should have yellow marrow.

15:22

But we can see in this T1 that at four years of age,

15:26

this is not a normal fatty marrow

15:29

within the humeral head. So there is an abnormal

15:32

distribution. Number two, it's so dark.

15:36

In some areas, it's lower signal intensity than the adjacent muscle,

15:40

so we have an abnormality in the distribution and an abnormality in the

15:44

signal. And the problem is that in the T2, the bone

15:48

marrow infiltration is so diffuse that sometimes it's

15:52

easy to miss the finding. Or sometimes it's easy for

15:56

your eyes to go to more focal abnormalities.

15:59

So, for example, here in the ischium,

16:03

you can see that there is higher signal.

16:05

There may be some periarticular muscle

16:09

edema, but your eyes, that's not the finding.

16:13

And sometimes when there is so much bone marrow infiltration, you see

16:16

areas of necrosis or AVM, but that's not the

16:20

diagnosis. The diagnosis is not AVM.

16:22

The diagnosis is just diffuse bone marrow infiltration.

16:25

So the T2 sometimes can be a little misleading.

16:28

So go to your T1, apply those rules, and you won't

16:32

miss any bone marrow infiltration process.

16:37

So let's look at this case. So this is 17-year-old with shoulder

16:40

instability. So if we apply the rules, there

16:44

is a little bit of abnormal signal in the

16:47

subchondral region that is a little bit bright on the

16:51

PD, a little bit low signal on the T1.

16:55

But if you see it mirrors the

16:58

normal red marrow in the metaphysis, is higher signal intensity

17:02

than the adjacent muscle, but it has a relatively abnormal

17:06

distribution. It's in the epiphysis.

17:07

You shouldn't have, at this age,

17:10

anything like red marrow that looks in the epiphysis.

17:14

So the question is, is it osteomyelitis? Is it bone marrow infiltration?

17:18

Is it a normal variant or subchondral edema secondary to cartilage loss?

17:22

And with all the rules, there are always exceptions.

17:27

So this is actually a normal variant that we all need to identify, so

17:31

we don't call it abnormal. So it's really important when we evaluate red marrow in

17:34

children,

17:36

how there are some normal variants, and some of

17:40

them can be, for example, flame shape.

17:42

You can see these

17:44

metaphyseal flame-shaped red marrow signal

17:48

abnormalities. But if you follow the rules, it is

17:52

higher signal than the adjacent muscle.

17:54

It has normal distribution, which is in the metaphysis.

17:58

The epiphysis look normal. It's ill-defined.

18:01

So that's a normal red marrow.

18:03

You can also have marrow appearance, particularly in the feet

18:08

and the ankles. We don't know if this is just

18:11

clusters of red

18:14

marrow, particularly after you have a fracture, you are

18:18

immobilized, you cannot weight-bear, or if it's just

18:22

microfractures within

18:25

the bones. It's unclear. We believe that it's just a normal

18:29

variant, and you will see in our report that we sometimes didn't even mention it

18:33

because it's so normal in kids.

18:37

This is an example in the ankle where you can see all this mottle

18:40

appearance of patchy red marrow, and this is completely normal.

18:45

Another normal variant is the epiphyseal halo, and this is the

18:49

example I gave in case two. We need to remember that it exists, so we don't

18:52

call it abnormal. So, the rules of this one is that it's

18:56

parallel to the

18:59

articular surface. We usually see it in the hip or the

19:04

humerus or the shoulder. And if you see, it follows the rules that

19:08

is higher signal than the muscle, is ill-defined, and it has to

19:12

be parallel to the subchondral bone.

19:17

Another variant is the fascial anchoring,

19:21

which is these little lines that we see that is not true bone

19:25

marrow edema. We don't know if it is just red marrow or if

19:29

it's just fibrous anchoring of the fascia.

19:32

But this is not a FOPE, which is the focal

19:36

edema, perifascial edema. This is just a normal anchoring of the

19:40

fascia.

19:42

So let's go to the case three. So this is a 16-year-old with

19:45

contralateral right femoral osteosarcoma on

19:49

chemotherapy. So all these two are T1 weighted images.

19:53

This was in 2018, and this one was in 2019.

19:57

And you can see that there has been a change in the bone marrow appearance.

20:01

So on 2018, at time of diagnosis, we can see that there is

20:05

normal red marrow in the proximal femoral metaphysis, which is normal

20:09

in appearance. But in 2019, it started to be too dark,

20:13

too lower signal compared than the muscle, and there is this

20:16

abnormality in the epiphysis and the diaphysis.

20:19

This is abnormal. So the question is, what can this be?

20:24

Is this osteomyelitis? Is this metastasis?

20:27

We know that the patient has osteosarcoma.

20:29

Is this bone marrow reconversion or pseudostoma or stress

20:33

changes?

20:36

So in this case, this is bone marrow reconversion.

20:40

And we need to be familiar of what is bone marrow reconversion.

20:43

So bone marrow reconversion is the opposite of what we discussed. Right?

20:47

The good news is that it happens in a reversible,

20:50

predictable pattern. So it's the completely the

20:54

opposite of the conversion. So that means that if your body

20:58

needs more hematopoiesis, so for example, you have

21:02

anemia or you are on

21:05

granulocyte stimulator factors,

21:08

as part of your treatment,

21:12

the bone marrow is going to convert to red marrow.

21:16

So that's what we call a reconversion, and it's going to happen in a

21:20

reverse, predictable process. So that means that first it's going to be your

21:23

metaphysis, then it's going to be your diaphysis, and finally your

21:27

epiphysis with the axial skeleton happening first, and it's going to be

21:31

from

21:33

proximal to distal. So it's a reconversion.

21:36

So how do we know if it's normal or abnormal?

21:39

So the clinical history is going to be really important.

21:42

Don't use this diagnosis if it's a

21:45

completely healthy patient with pain.

21:48

There has to be a reason. You have to have severe anemia.

21:51

You have to be on

21:53

a granulocyte stimulator factor treatment to be able to call

21:57

it. But you can see in this example how in 2018 it was

22:01

normal. 2019, he had post-treatment response from bone marrow

22:05

reconversion. He was on granulocyte stimulator factor.

22:08

And then in 2020,

22:10

as the treatment stops, it goes back to normal bone

22:14

marrow. So this is a normal process that you don't have to

22:17

confuse with metastasis.

22:21

So as we have discussed throughout this talk, the bone marrow

22:24

abnormalities can be either in distribution, so that's why

22:28

you need to know normal distribution, or signal abnormalities.

22:31

So the T1 is too dark or the T2 is too bright.

22:36

We can also divide the bone marrow abnormalities in focal,

22:40

multifocal, diffuse, or treatment effects.

22:44

We don't have time to go through all of them individually, but my

22:47

goal is to give you some

22:51

tips to be able to decide if you should

22:55

raise concerns or if you should actually be

22:59

more

23:02

conservative in your recommendations.

23:04

So focal lesions, we have, of course, infection.

23:08

Focal lesions can be infections, tumors, any type of

23:12

primary bone tumor. Multifocal, we always think

23:15

about metastasis,

23:18

Langerhans histiocytosis, CRMO, and

23:22

in children, particularly in small children, we think about infection, which in

23:26

adults we don't think about infection just because in children,

23:30

the osteomyelitis is going to be,

23:34

it has a hematopoiesis,

23:36

it's spread by bacteremia

23:40

instead of by direct contact, which in adults is like you

23:44

have a diabetic foot, you have an ulcer, and usually it develops

23:47

osteomyelitis. But in children, they usually have just

23:51

bacteremia, so it can seed in multiple bones.

23:55

We have a diffuse bone marrow infiltration, so when we think about

23:59

storage disease, leukemia, anorexia,

24:03

anemia, but we have to think, as I said, about

24:06

neuroblastoma and sometimes rhabdo in babies.

24:10

They kind of spread diffusely within the bone marrow.

24:14

And of course, we need to talk about treatment effect, which is what we just talked

24:17

about, bone marrow reconversion.

24:19

So let's talk a little bit about focal bone marrow

24:22

lesions. And my rule for focal bone

24:26

marrow lesions in the differentiation between is this

24:30

benign or malignant, should I recommend a biopsy or this is a

24:34

non-aggressive pathology,

24:38

I use the T1. So as you will hear me throughout this talk, for me, the

24:42

most important sequence is the T1 for bone marrow.

24:46

Why? Because Benign lesions or inflammatory

24:49

lesions are going to have ill-defined margins on

24:53

T1. Because the bone marrow infiltration is

24:57

going to be more diffuse. It's going to, the

25:00

pus or the edema is going to start going

25:04

through the medullary canal, but it's not going to be replacing the

25:08

fat, if that makes sense, as abrupt.

25:11

So in this case, so this is a 12-year-old with osteomyelitis.

25:15

We can see here on the T1 how it's really

25:18

difficult to create a true margin

25:21

of the abnormality on T1. So if you cannot follow

25:26

the abnormality on T1 with a pencil,

25:29

it's most likely non-aggressive. So it's most likely a benign lesion.

25:33

And if you really think about it, it's different to radiographs.

25:37

So in radiographs, if you can define your

25:40

margins that are sclerotic and you know exactly where they are, you think that it

25:45

makes you feel more comfortable. But in MRI, it's difficult, and it's the T1.

25:49

If you cannot follow it, it's a good

25:52

sign. So inflammatory lesions, we have, of course,

25:56

chondroblastomas in the epiphysis, which you are going to have a lesion with a lot

26:00

of inflammation. We have LCH, osteochondroma,

26:04

osteomyelitis, and as I said, the bone marrow edema, you're going to see it

26:07

better on the T2,

26:09

but the T1 is going to tell you how to characterize it.

26:13

So in this case, we can see that there is a more focal lesion here in

26:17

the metaphysis of the distal femur.

26:20

So that's a focus of osteomyelitis with transphyseal extension into the

26:24

epiphysis with a subperiosteal

26:28

abscess, which is pretty common in

26:31

kids. And this is a focal non-aggressive lesion consistent

26:35

with osteomyelitis.

26:38

So when we talk about malignant tumors,

26:41

the rules are the opposite on T1. So if you

26:45

can take a pen and define the

26:48

margin on T1 with a pencil,

26:52

that's abnormal. Because that means that it's a bone marrow infiltration

26:56

process, it's moving the fat

26:59

from the bone marrow, and usually we see that in osteosarcoma,

27:03

Ewing sarcoma, or lymphoma. So as I said, the T1 is your margin.

27:07

And when you are doing preoperative assessment or initial assessment of these

27:11

lesions, I always recommend my trainees to measure it on

27:15

T1 because some of the lesions could have edema on T2,

27:19

and it's really hard to know exactly what is the true margin of the lesion or

27:23

if it's just reactive edema, and I think the T1 is the most

27:27

useful. In this case, there is no doubt that it's an osteosarcoma

27:31

case. We can see the abnormality on the radiographs

27:35

with some burst periosteal bone formation, the

27:38

model appearance, the ill-defined margins.

27:41

But as I said, the T1 is going to give you the clue with defined margins.

27:45

So we have here two different cases in the distal femur.

27:48

So the first case, you are already familiar with, which is the osteomyelitis case,

27:52

ill-defined margins. But this is another case of osteosarcoma.

27:56

So if you can grab a pencil and you can draw the line, that's an

27:59

aggressive lesion, even if you don't have the radiographs, but hopefully you

28:03

have the radiographs in these cases.

28:05

But as I said, the T2 is really hard to define where

28:09

is the margin of this lesion. The T1 is going to help you to define

28:13

exactly where is the lesion. The T1 is super useful to

28:16

evaluate focal bone marrow anomalies.

28:20

So let's go through some cases. So this is

28:24

actually, I think it's ChatGPT, and I

28:27

asked for interesting pediatric MSK cases, like images just

28:31

to have it here. And actually, it's really interesting.

28:35

They don't make any sense, but they are indeed interesting if you

28:39

really start looking at them individually. So let's see.

28:42

So this is a 16-year-old with

28:46

pain in the distal tibia.

28:50

So just starting with the radiographs,

28:53

this is a concerning lesion. So you have a lytic

28:56

lesion in the distal tibia. The

29:00

margins are not well-defined.

29:03

There is a narrow zone of transition.

29:06

Even in this radiograph, it's a little bit hard to see, but there

29:10

could be an aggressive-looking periosteal bone

29:13

formation with a comet's tail here.

29:16

So it doesn't look great. So when we did the MRI, we

29:20

confirm that it's an aggressive lesion.

29:22

And I'm showing these two because sometimes, unfortunately, we don't have the

29:26

radiographs when we are looking at the MR, and the MR can be misleading.

29:29

So in this case, you can follow the margins with a

29:33

pencil on T1 that's aggressive. We can see on a

29:37

stir that there is not that much bone marrow edema.

29:39

There is an exophytosis of tissue component that enhances, and this

29:43

was a case of osteosarcoma, even though we didn't

29:47

see any osteoid forming matrix in

29:51

the radiograph. So sometimes it's really hard to differentiate between

29:54

osteosarcoma and Ewing sarcoma, but the T1 helps

29:58

you to say that this case needs a biopsy.

30:04

So this is another really interesting case.

30:08

It's an UNMME case. So if you haven't seen one of these,

30:12

I hope after you see this case,

30:16

next time you see it, if you have this case, you're going to be able to just give

30:20

the diagnosis because it's relatively rare, it's relatively uncommon,

30:23

but the imaging findings are pretty characteristic.

30:26

So we have a 15-year-old with atraumatic recurrently

30:30

pain and effusions for two months.

30:33

So the radiographs are pretty much normal.

30:36

And then when we do the MRI, we can see that there are

30:40

abnormalities in the bone marrow, that there are multifocal.

30:43

So here it is involving the distal femoral epiphysis, the

30:47

medial femoral condyle, and it involved with the proximal

30:50

tibia from the epiphysis to the metaphysis with transphyseal

30:54

extension. That's a little atypical.

30:57

But more importantly on the T1 is that you have sharp,

31:01

well-defined margins. You can draw with a pencil.

31:05

So if you really think about

31:08

multifocal focal lesions

31:11

that are normal on radiographs, your differential diagnosis gets

31:14

narrower. So this is, even though it's multifocal, it looks

31:18

aggressive.

31:20

On the STIR sequence, you can see

31:23

T2 abnormalities, and in post-contrast sequences, more

31:26

importantly, is that it has these

31:30

serpentine areas of

31:33

post-contrast that look like infarcts.

31:37

So as I said, if you haven't looked at this before,

31:40

I invite you to remember this because this is an UNME

31:44

for primary osseous lymphoma. And Dr.

31:48

Duffy and Dr. Eklund from Boston Children's, they published this in

31:52

2023, a series of these cases, and they realized they all look the same.

31:56

They are rare. Of course, primary osseous lymphoma is rare.

31:59

They usually start in the lymph nodes or the

32:03

spleen or the thymus, but this is rare. This is a primary of the bone.

32:08

And as I said, the key features to make this diagnosis is

32:12

multifocality, epiphyseal involvement, sharp

32:15

T1 margins, and an enhancement pattern that look like infarcts.

32:20

And most importantly, post-therapy,

32:23

they can have persistent or residual osseous abnormalities, even if

32:27

it's a good treatment response.

32:31

So this is another case. This is a 12-year-old with back

32:35

pain. And this was a case that

32:39

was, I think it was done overnight, and the fellow was in

32:43

charge of preliming this. And what caught the eye of the fellow was

32:47

this T1, T10 abnormality. It

32:51

looks really heterogeneous on T2.

32:54

It doesn't have contrast enhancement.

32:56

It doesn't enhance it

32:58

normally. It has some paraspinals of

33:02

tissue prominence. So he called, this could

33:06

be a metastatic lesion. This patient has a

33:09

really old history of neuroblastoma when he was a two-year-old.

33:13

So he's like, "Oh, could this be something focal in

33:17

T10?" But what it was missed, and this is why the

33:21

diffuse bone marrow abnormalities are so hard to

33:25

identify, is that it's not only T10.

33:28

It's the whole spine is abnormal. So as I said, go to your

33:32

T1, and in the spine, you're going to use the

33:36

intervertebral disc as your gold standard to evaluate

33:40

if it's higher or lower signal. And you can see that all the bone

33:43

marrow is lower signal than your

33:47

intervertebral disc, so this is completely abnormal.

33:51

The T2 is misleading just because it's so

33:55

diffusely abnormal that it's hard to see.

33:58

And then in this case, it has a more

34:00

ischemic

34:03

process of the bone marrow infiltration in T10, and that's where your eyes go.

34:07

But this is a diffuse bone marrow infiltration

34:11

in

34:12

unfortunate neuroblastoma recurrence 10 years later.

34:15

But lymphoma leukemia can look like this. Correct?

34:19

So any bone marrow infiltration process can look like this, and in this

34:23

case, it was neuroblastoma. So this

34:26

is a 10-year-old that present-- No, sorry, a five-year-old that

34:30

presented with

34:33

thigh pain. We did the radiographs.

34:36

We can see that there is non-aggressive but

34:40

abnormal periosteal new bone formation in the

34:45

femoral diaphysis. So we recommended an MRI.

34:49

And when we do the MRI, we can see that there is a more focal

34:53

intramedullary lesion

34:56

that is heterogeneous on T2,

34:59

low signal on T1. But importantly, there is

35:03

all these surrounding bone marrow abnormalities.

35:05

So as I said, where do you measure? Do you measure the whole femur?

35:09

And on T1, this bone marrow signal abnormality, it

35:12

is non-aggressive. It goes all the way to the distal

35:16

femoral metadiaphysis, but it's non-aggressive, and you see the pattern of

35:20

enhancement is non-aggressive. You cannot draw it with a pencil.

35:23

So this means that is an inflammatory mass.

35:28

There is a lot of inflammation surrounding this

35:31

lesion. And even though

35:35

LCH can look like anything, every time in a child that you

35:39

have an intramedullary lesion with extensive

35:43

inflammatory changes, I think your differential diagnosis

35:47

should be LCH number one, two, and three.

35:51

And of course, this patient will need a biopsy,

35:54

but this is pretty characteristic for LCH.

35:59

So we're going to go to our last case, which is really

36:03

abnormal. I actually have seen this

36:06

only in

36:10

conferences until last year with this case.

36:13

So this is a 15-year-old with a horrible history of

36:17

fulminant sepsis. You can see that he had cardiac

36:21

arrest outside of the hospital, a multi-organ

36:25

failure,

36:26

multiple infections, opportunistic infections,

36:30

including cutaneous mucormycosis.

36:32

So he was a pretty

36:34

sick kid. After he was

36:38

getting better and he was more responsive, he was complaining of hip pain.

36:42

So he went for an MRI of the hip,

36:45

and these were the initial images.

36:48

Because they were concerned about osteomyelitis, so we did a FAST protocol, which

36:51

is a T1 and a STIR sequence. And the

36:55

tech sent me these images, and I was looking at them, and the tech was like, "Oh,

36:59

did you want contrast?" What do we want? And the images look really abnormal.

37:03

I want you to think about which one is the STIR

37:07

sequence and which one is the T1 sequence.

37:11

And I'm going to show it now.

37:14

And the clue is actually in the urinary bladder,

37:18

because it's so abnormal, but you won't be able

37:22

to know just from the fat, the muscle, and the bone marrow.

37:25

So this was labeled STIR sequence,

37:29

but this doesn't look like a STIR sequence.

37:31

The STIR should have fat suppression, and you see the

37:35

subcutaneous high signal, which is abnormal.

37:38

The bone marrow is too hypointense, so it's really

37:42

abnormal. And this is what the

37:45

technologies label T1, which is,

37:49

again, abnormal. Where is the fat in the subcutaneous tissue?

37:52

Where is the fat within the bone marrow? This is a 15-year-old.

37:55

You should have more yellow fat, particularly in your

37:59

femoral head. So it looks really, really abnormal.

38:02

But if you see the bladder, which is fluid, it

38:06

maintains the same signal. So when you see

38:10

this,

38:11

and I want you to think that this is, "Oh, it looks like the images are flipped."

38:16

This looks like a T1, and this looks like a STIR sequence.

38:19

So the first thing that you have to make sure is that there is not a

38:23

problem with the labeling. So I called the tech right away, "No problem with

38:26

labeling.

38:27

Are you sure that you activated the fat suppression

38:32

in the STIR and you didn't activate fat suppression on T1?" "Correct."

38:36

"Are you sure that the gradients are working in the scanner?

38:38

Did we scan another patient today that didn't have any problem?" And she

38:42

said, "Yes. Everything, yes." And the way to prove that this is a true

38:46

flip-flop phenomenon is that you can either do an spectroscopy

38:50

and check how much fat you have in the subcutaneous fat and the bone marrow,

38:55

or

38:56

easier, which is the thing that I did, I requested a Dixon

39:00

sequence. A Dixon with water in-phase and

39:04

fat reconstruction. And you can see that this is a completely different

39:08

technique, and it shows the same finding.

39:10

So in the water, you can see that there is

39:13

water signal in the subcutaneous fat.

39:18

In the fat, there is no fat signal. There is no bright

39:22

signal on the fat, so this is like the findings were true.

39:25

And interestingly, this patient underwent an MRI the day before

39:29

of the elbow, and this was read as,

39:32

"Many technical issues we are unable to assess.

39:35

Patient will be rescheduled because of technical issues with the scanner." And

39:39

if you see, it's the same findings. That is, the STIR sequence looks like a T1.

39:44

The T1 doesn't look like a T1. There is so much bright

39:47

signal on the subcutaneous fat, and this was a true

39:51

case of a flip-flop phenomenon. So a flip-flop

39:55

phenomenon is also called serous atrophy

39:58

of the bone marrow or gelatinous

40:02

transformation of the bone marrow. That looks really scary.

40:04

But what it means is that there is a catabolic

40:08

state,

40:09

which this patient had multi-organ failure.

40:13

There is progressive depletion of the fat and protein stores, and

40:17

then all that fat within the subcutaneous tissue as well as the bone

40:21

marrow is going to transform into hyaluronic

40:24

acid. And that's the reason why you have

40:28

fluid-like signal in the bone marrow as

40:32

well as the fat. This is extremely rare, and you can

40:36

see it in cachectic states, anorexia

40:39

nervosa, or HIV/AIDS.

40:43

It's really rare, but I'm sharing them with you because I learned a lot from this

40:47

case. So if you have a question, always use a different technique of fat

40:50

suppression. Use a Dixon that can give you the fat and the water, and you are

40:54

sure that this is not a technical problem.

40:57

So in conclusion, it's really important

41:00

to identify what normal looks like in the normal bone

41:04

marrow in children so you don't think that this is

41:07

an abnormal bone marrow infiltration.

41:10

You need to learn the rules of what normal bone marrow looks like.

41:14

So remember, first, it converts distal to proximal,

41:18

and it converts epiphysis, diaphysis, metaphysis. Epiphysis, diaphysis, metaphysis.

41:23

The red marrow should be higher signal than muscle on T1, and I

41:27

believe that the T1 is the best sequence to characterize the red marrow

41:31

and to increase or decrease your concern in

41:35

these patients. So thank you so much, and I'm open for

41:39

questions.

41:41

Thank you so much for that lecture, Dr. Bedoya. That was excellent.

41:45

We are opening the floor for questions, so if you have any, please go ahead and

41:49

put those in that Q&A feature.

41:52

And there's one in there right now, Dr. Bedoya, if you're able to pop that open.

41:58

I am, but I'm sorry, I don't see a question.

42:02

Try hovering over the top of your Zoom.

42:04

You might see the Q&A with the

42:07

bubble with the question mark in it.

42:09

Let's see. I'm sorry.

42:11

It's okay.

42:13

Oh, question mark. Yes. So

42:16

in vertebrae, how can you see bone reconversion?

42:19

So in vertebrae, it's hard because in vertebrae,

42:23

you can have normal red marrow in women who

42:27

have their periods. They need more hematopoiesis, so it's normal to have

42:31

red marrow. So if you only have lumbar spine,

42:34

it's really hard to know. But the normal red marrow

42:38

should always be higher signal than the intervertebral disc.

42:42

So that should be your normal control to know if it's abnormal or normal.

42:47

Thank you so much. Do you have any experience with whole-body MRI

42:51

for marrow evaluation? So that's a really good point.

42:54

So it depends on what you're looking for.

42:56

So for example,

42:59

and this was a big discussion in

43:01

the Society of Pediatric Radiology meeting this month in Boston,

43:05

because it's what you're looking for.

43:07

So in patients of CNO,

43:10

where you only want to see focal areas of osteitis or

43:14

abnormal high T2 signal, sometimes that's the only

43:18

thing that the rheumatologists need.

43:21

So a T2-

43:24

May be enough. However,

43:27

I always recommend to do a T1 and a T2 STIR

43:31

of the whole body MR, because sometimes there are some lesions,

43:35

particularly in cases of,

43:40

for example, in ED from many cases, where

43:43

any abnormality is really abnormal, but these kids, they

43:47

can run, they can hurt their knees, they can have a lot of contusions and

43:51

edema. And I feel that T1 helps me to raise my

43:54

concern or not. So if I see that there is a focal lesion and then the T1

43:58

is too dark, I say, "We need additional imaging. We need biopsy.

44:02

We need to do more for those whole body MR."

44:05

If I see that is an area of common contusions, like femoral condos, and

44:09

the T1 is non-aggressive, I may say, "Oh, let's just do a

44:12

radiograph in three months to make sure that nothing is there." So I think the T1

44:16

helps you with that in the whole body MRI.

44:18

Right now, in Europe, what they're doing for whole body MRI is that

44:22

they are doing a T2 STIR, and then they're doing...

44:26

No, sorry, not T2 STIR. They're doing a T2 Dixon

44:30

Corona, and they are just reconstructing the water,

44:34

which looks like a STIR, right? It's like a T2 fat

44:38

suppress. And then they reconstruct the fat, like the T2

44:42

fat Dixon, and that's going to tell you if there is fat

44:46

within those lesions. So,

44:49

use all your tools that you have. I know sometimes in body

44:53

imaging for kids, we don't have a true T1 sequence.

44:56

But usually, we have

44:59

an in and out of phase, so you can use those to see if it's abnormal,

45:03

abnormal, and if there is any question, always add your T1.

45:08

Someone asked, "How does myelofibrosis look?" So myelofibrosis,

45:12

it can be really hard to see on MRI. I feel like it will be

45:16

dark on T1 and dark on T2.

45:20

But sometimes I have seen cases where it looks like a--

45:24

The clue for me is the T2, particularly in myelofibrosis.

45:29

But I haven't had a case where I'm confidently,

45:33

like I know it looks abnormal and I recommend a bone marrow

45:37

aspirate to make sure

45:40

what it is.

45:42

But that's why I didn't, I don't have a true case.

45:44

But it should be a lower signal on T2.

45:48

There is another question. "Is the rule of chart margin and

45:52

TY apply on adult malignant tumor?" I

45:56

don't see why not. I

46:00

don't see why not. I feel it's just that

46:03

in children,

46:06

sometimes we are into like, "Is it normal red marrow or is it a pathology?"

46:10

Right? That's how we are usually finding ourselves, and that's

46:14

why we use that T1. But I don't see why not.

46:17

I have noticed that most of the primary metastasis,

46:21

some of the metastasis,

46:23

they all have chart T1 margins. But you have to be

46:27

careful with

46:29

post-treatment changes because in the post-treatment changes, they get more

46:33

fatty infiltration, even if that is still present on the T2, and the

46:37

T1 can be a little misleading. But I use the T1 a lot to

46:41

evaluate metastatic disease also. So I don't see why not.

46:45

"How do you differentiate osteoporotic spine

46:49

fractures from a malignant one?" So that's a really good question, and I think

46:53

this goes with adult

46:56

T1 signal. I train under Dr. Ben

47:00

Cardino, and I remember many times that she was saying to me

47:04

that a pathologic fracture, if there is

47:07

multiple myeloma or there is lymphoma or something there,

47:11

you should still see a little bit of T1 bright

47:15

signal when compared to the intervertebral disc.

47:18

Edema, the same, it should be ill-defined, and the T1 will

47:22

help you. But you should see a little bit of normal

47:27

red marrow, right? Which is what we discussed today.

47:30

If you see complete

47:32

effacement of the normal T1 signal, that means that it's way

47:36

darker than the intervertebral disc, you have to raise the concern for a

47:39

pathologic fracture. I think the T1 helps.

47:43

Also, use all your tools if you

47:46

have. Sometimes diffusion helps that too.

47:51

But it can be hard sometimes. But I do use the T1 as the rule.

47:55

I remember Dr. Ben Cardino saying that if it gets rid of the fat,

47:59

that is

48:00

malignant until proven otherwise.

48:02

And I guess that the MRI is really helpful to differentiate fat

48:06

from just fluid.

48:09

"How all patients..." Okay.

48:13

Sorry, another question. "In many old patients, diffuse dark T1

48:17

signal. If it still looks okay, how did you interpret?" So,

48:23

in all patients, diffuse T1 dark signal is abnormal, right?

48:26

Because there is no reason why.

48:29

It's not red marrow, right? So why did you have diffuse?

48:33

And that's where some cases of

48:38

diffuse bone marrow infiltration can get lost, right?

48:42

So if I only see really dark T1 signal, I am concerned about

48:45

bone marrow infiltration and I would recommend a bone marrow infiltrate if there is

48:49

no other reason to explain it, right?

48:51

If there is no severe anemia, if there is no

48:54

granulocyte stimulated factor. I feel the T1 is

48:58

so useful because every time if I see a

49:02

lesion or something on the T2, I open my T1 because if

49:05

it's bright on T1, you're like, "Okay, this is non-aggressive, right?

49:10

This is either a venous malformation or what we call an intrasystem

49:13

angioma, or this is for fat deposition." So the T1 is your

49:18

clue. And if it's dark on T1, I will

49:22

raise my concern and I will follow it.

49:26

Okay, another question. Can TB cause

49:30

diffuse vertebral body loss on T1 and T2?

49:35

So I guess that it depends on the bone marrow

49:38

infiltration. The answer is yes,

49:41

but it will look more ill-defined.

49:45

It will have some edema around, and I guess you

49:49

always think in TB about the multilevel involvement, the disc

49:53

preservation. You have to use other rules to be

49:56

sure.

49:58

But it does. The other question is, "Is T1 Dixon

50:02

helpful?"

50:04

I think it's helpful.

50:07

It depends on what you're looking for, but I think it's helpful if you have

50:10

in older kids, if you have in, out-of-phase that

50:14

you can see if there is drop of signal in out-of-phase.

50:17

Because you're always just trying to look if there is

50:21

microscopic fat. If you have microscopic fat, it's a good

50:25

sign.

50:27

So yes, I find it helpful.

50:29

But I like my routine T1

50:33

in all my pediatric protocol cases.

50:36

I

50:38

have to play more with the T1 Dixon.

50:40

We use T1 Dixon for the abdominal imaging in kids,

50:44

just because you can use those two sequences

50:47

without any additional sequences. It's helpful.

50:52

But if you have any question, just add your SAG or your coronal T1.

50:59

Okay, I think that those were all the questions.

51:02

I think you got them all. Yeah. Thank you so much.

51:06

Of course.

51:07

Yeah. Thank you so much for this presentation. That was fantastic.

51:09

Really appreciate you being here.

51:11

Of course. Happy to be here. Thank you for the invitation.

51:15

And thank you for everyone else for participating in this noon conference.

51:19

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

51:22

conferences by creating a free account.

51:23

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

51:27

Be sure to join us next week on Thursday, July 2nd at 12:00 PM Eastern, where Dr.

51:31

Rahseeh Malak will deliver a lecture entitled Vessel Wall Imaging.

51:35

You can register for that at medallity.com.

51:37

Follow us on social media for updates on future noon

51:40

conferences. Thanks again for learning with us and have a great day.