0:00

So just shifting gears briefly. So I'm def I'm, um,

0:03

I have a interest in access to breast imaging and, um,

0:09

you know, part of that is the utilization of breast imaging. So our group did a,

0:13

a study recently looking at the utilization of tomosynthesis, um,

0:18

at our own institution. So we started, uh,

0:20

doing breast tomosynthesis in 2013. And we, um,

0:26

we found that although insurance, most insurances do cover it,

0:29

not all of them do. And we saw that there were, we,

0:32

we were hypothesizing that there were disparities in who was actually having

0:35

their screening mammograms performed with tomosynthesis.

0:38

And this is what we found in that study. So, when starting in April, 2013,

0:43

when we first started doing screening mammograms of 2013 with, uh,

0:47

tomosynthesis, we saw a pretty dramatic increase in utilizations from, um,

0:52

19% in the first year up to about 90%, um, currently.

0:57

But at all the time periods, we saw differences based on, uh,

1:00

patient's race with a, a disparity between, um, white patients having, uh,

1:04

being more likely to have their, um,

1:07

screening mammogram performed with tomosynthesis compared to black and Asian

1:10

patients. So, and this just highlights that there is, uh,

1:13

definitely work to be done to make sure that this modality is available for all

1:17

patients. I'm going to, um, move towards,

1:22

uh, tomosynthesis guided biopsy now for the second part of the lecture.

1:31

Um, so before we can talk about tomosynthesis guided biopsy, we need to, um,

1:35

make sure everyone's on the same page with stereotactic biopsy. So,

1:39

historically, uh,

1:40

mammographic findings that did not have a sonographic correlate have always been

1:43

biopsied with stereotactic guidance. Um,

1:46

so with stereotactic biopsies, we take a paired image.

1:50

These are two images that are taken at a, um,

1:52

plus and minus 15 degrees from the, the scout image,

1:56

and that that's used to determine the position of the finding, um,

2:00

specifically the depth of the finding.

2:05

But now we've been doing all of these mammograms with, um,

2:08

tomosynthesis and there are many findings that are only seen

2:13

or best seen with tomosynthesis. And so what do we do about those?

2:17

So before we had, um, the ability to do tomosynthesis guided biopsies,

2:22

that was a, it was really a problem to try to figure out what we should do.

2:25

So these patients were either, um, recommended to have,

2:29

go straight to surgery and we would use, um,

2:31

tomosynthesis to localize the finding. Um,

2:34

and then they would do a surgical excision. Um, we'd also, uh,

2:38

sometimes we would evaluate these findings further with M R I, um,

2:41

but if the M R I is negative, you're still left with this, um,

2:44

this finding that hasn't been, um, that has not been fully resolved.

2:49

So luckily tomo sinus got added. Biopsy was, um,

2:54

was started to be used. And this allows us to biopsy findings that are either,

2:58

uh, best seen

2:58

Or only seen on tomosynthesis imaging. Uh, so our group started,

3:03

started doing this and we had a lot of success with it.

3:06

There were some early studies that with high technical success,

3:08

they showed shorter procedure times. Um,

3:11

but after we had done this for some time,

3:13

we thought there was still limited data in the literature and it was reasonable

3:17

for us to, uh, look at our own experience to try to help, um,

3:20

to help show whether this has been a successful, successful or not. So, uh,

3:25

most of the cases of tomosynthesis guided biopsy were for architectural

3:29

distortions. So that was the first project that we did. We, um,

3:32

looked at all of our cases of architectural distortions that did not have a

3:35

sonographic correlate and were biopsied with tomosynthesis guidance. Um,

3:40

this was, uh, published in a j r a few years ago. Uh,

3:44

the study included 151 unique architectural distortions without

3:50

a sonographic correlate. We found that, um,

3:53

19% of that cases were malignant, 32% were high risk,

3:57

and 49% were benign. Uh,

4:00

most of these malignant lesions were invasive carcinomas, and interestingly,

4:05

most of the invasive carcinomas were lobular, which is, um, overall less,

4:09

much less common than, um, ductal, which is the,

4:11

the most common type of invasive breast cancer. The high risk lesions, uh,

4:16

most often represented radial scars or complex rosing lesions.

4:20

And I think this just emphasizes that we really wanna make sure that these

4:24

distortions, even if they don't have a sonographic correlate,

4:26

are bios seeded since 19% of them represented malignancies and another

4:30

32% represented high high risk lesions.

4:35

Um, we did a follow-up, very similar study looking at developing asymmetries.

4:39

These are not distortions, but um, are, are still suspicious findings. Um,

4:43

again, that did not have sonographic correlates. Um, this was, um,

4:48

published a couple of years ago in radiology,

4:50

and we included 85 developed as asymmetries,

4:53

did not have a sonographic correlate, and we found a very similar, um,

4:58

malignancy rate of 20%.

5:01

And almost all of those malignancies were invasive carcinoma. So again,

5:04

this highlights the importance of biopsying these findings,

5:07

which are are suspicious on mammography even if they don't have a sonographic

5:11

correlate.