Ultrasound Imaging of the Breast - Lesion Characterization - HD

Introduction to Breast Ultrasound and BIRADS

Hi, I'm Richard Barr from South Woods Imaging in Youngstown, Ohio.

Today I'd like to give you a talk on an overview of breast ultrasound highlighting the birads classification.

Today I'd like to give you a general talk on ultrasound imaging of the breast talking about lesion characterization.

I'd like to go over some of the things that I do when I'm doing an examination, as well as review the birads categories and characterization of lesions.

Types of Breast Ultrasound Examinations

The first thing that I think you need to do when you're doing a breast ultrasound is decide if you're doing a detection exam, meaning that you're doing a diagnostic workup for a clinical finding or other imaging finding from mammography or MRI or if you're doing a screening examination.

If you're doing a detection examination, that is that the patient has a symptom or a problem that you're trying to further work up, then you also need to decide if you're gonna do a targeted exam to only include that area that's abnormal, or you're gonna do a whole breast ultrasound.

There's no real guidance as that one is better than the other.

I think that your institution should make a decision so that you have uniformity throughout your division if you're gonna do a targeted exam.

One thing you have to remember is the breast can roll when a mammogram is being performed, and therefore, I prefer to scan over 120 degrees.

That is 20 minutes on the clock face from where I think the lesion is.

So if I believe the lesion's at two o'clock on the mammogram, I always scan from 12 o'clock to four o'clock.

I always take a picture of the farthest clockwise and the farthest counterclockwise image of where I scan for documentation.

This way you can go back and you always remember at what locations you've scanned in the breast.

Scanning Planes and Coverage

I always perform scans in two planes, and I prefer the radial and anti radial.

But there are you can do sagittal and transverse if you prefer.

Another thing that I think a lot of people don't realize is if you're going to be scanning the entire breast looking for lesions, you do need to scan superior up to the clavicle, inferior to the inframammary fold, medial to the midline and lateral to the mid axillary line.

And we also include the axilla as part of a whole breast examination.

I always document at least one image in each quadrant, even if no lesions are present.

Again, that's just for documentation that that area of the breast was scanned.

If there are greater than four cysts or well circumscribed solid lesions such as fibroadenomas that we see in patients with fibrocystic disease, I usually take only one image to document the size and position for further follow-up.

And really just characterize one lesion in each quadrant, which is either the most suspicious or the largest.

BIRADS Classification Overview

When we find a lesion on ultrasound, we want to use the breast imaging and reporting data system birads to characterize that mass.

The Birads classification includes this discussion on masses, calcifications, special cases that are unique to ultrasound vascularity, and an assessment category.

To be able to characterize a lesion, you have to have excellent BMO images.

You need to evaluate the entire lesion.

It's not sufficient if you see a lesion that you think is a cyst to just take a transverse or a radial and an anti radial position and not interrogate the entire cyst wall, you do need to look for subtle findings as for irregularity of the masses or small speculations this will help to put the lesion in the appropriate birads category that we'll discuss later on for the patient to have appropriate treatment and follow up.

Mass Shape

In terms of masses, one thing that we want to do is shape.

This can be either oval, which is elliptical or egg shaped, and may include two or three undulations, round spherical, ball shaped circular or globular irregular, neither round nor oval in shape.

And here are some examples of these.

Three oval round and irregular.

Oval is usually characteristic of a benign lesion.

Round is indetermined, excuse me, intermediate and irregular is more suspicious for a malignant lesion.

Mass Orientation

We also want to describe the orientation of the lesion.

Is it parallel the long axis of the lesion parallels the skin line otherwise stated as the lesion is wider than taller or horizontal, or it's not parallel.

The long axis is not oriented along the skin line, meaning that it's taller than wider or vertical.

And here are examples of parallel where the long axis of this lesion parallels the skin and not parallel where the lesion is taller than wider not parallel is more suspicious for malignancy.

Parallel is more suggestive of a benign lesion margins.

Mass Margins

These can either be circumscribed or not circumscribed.

By circumscribed we mean the margin that is well-defined or sharp with an abrupt transition between the lesion and surrounding tissue.

Not circumscribed.

The mass has one or more of the following features.

It can be indistinct, no clear demarcation between a mass and its surrounding tissue.

Angular.

Some or all of the margins have a sharp corners, often are forming acute angles.

It can be micro modulated where there's short cycle undulations imparting a scalloped appearance to the margin of the mass, or it can be speculated where a mass is formed or characterized by sharp lines projecting from the mass.

And here we have examples of all of these.

Circumscribed, again, clearly defined margins.

These are often benign indistinct, where here we have a hypoechoic lesion to go into a hyper coic rim, and it's very hard to determine exactly where the transition is between those.

So the borderline is indistinct angular.

You can see we've got several irregularities here in the mass with angular margins.

Again, more suspicious for malignancy.

Micro lobulated, we've got several well circumscribed undulations and the spiculated mass where we've got multiple speculations coming outta the mass more suggestive of a malignancy.

Lesion Boundary

We wanna look at the lesion boundary.

It can either have an abrupt interface, meaning that there is a sharp demarcation between the lesion and surrounding tissue, which can be imperceptible or a distinct, well-defined echogenic rim of any thickness.

We can also have an echogenic halo where there's no sharp demarcation between the mass and surrounding tissue, which is bridged by an echogenic transition zone.

Here are some examples of an abrupt interface in the epidermoid cyst.

And you can clearly define a very well-defined border of this lesion.

And you can pinpoint in one location or one pixel where the lesion begins and ends, and the image on the right, we have a echogenic kalo.

Again, we have a hypoechoic lesion that kind of fades into the hyper coic rim and then fades into the adjacent tissue.

So it's very hard to determine where the boundaries of this lesion are.

Echo Pattern

We want to describe the echo pattern, anti coic without internal echoes.

Hyper coic having increased echogenicity relative to fat, or equal to fibro glandular tissue.

And again, in breast imaging, we always want to describe the echogenicity as compared to fat.

The lesion can be complex as a mass containing both anti coic and echogenic components.

It can be hypoechoic defined relative to fat.

Masses are characterized by low level echoes throughout that is the appearance of a complicated cystic fibroadenoma, or it can be iso coic having the same echogenicity as fat.

A complicated cyst or fibroadenoma may be hypoechoic or iso coic.

And here are some examples koic in this simple cyst hypoechoic where the lesion is uniformly increase in ity to the adjacent fat.

If you see such a lesion as this, it's very important that you scan through the entire lesion and make sure that the entire lesion is hyper coic.

Those lesions are most often benign.

However, you need to be aware if there is a hypoechoic central area, you may be looking at a malignancy with a large hyper coic rim of surrounding tissue.

So again, in these hyper coic lesions, make sure you interrogate the entire lesion and that they're uniformly hyper coic throughout complex, having both anti coic and echogenic components.

Iso coic in this lipoma and hypoechoic again compared to fat up here and most malignancies that we're gonna be looked at are usually hypoechoic, are very hypoechoic.

Posterior Features

We also want to look for posterior al features.

There's no posterior cial features meaning there's no posterior shadowing or enhancement.

We can have enhancement where there's increased posterior echoes shadowing where there's decreased posterior echoes and edge shadowings or excluded.

We can have a combined pattern where there's more than one pattern or posterior attenuation, both shadowing and enhancement.

And here are some examples.

We have two simple cysts here.

One, we have in the same patient, very good through enhancement, and the other one has no features.

Here's an area of shadowing, and this lesion has both root transmission and areas of shadowing.

Surrounding Tissues

We also want to look at the surrounding tissues.

We wanna look for duct changes, so we wanna look for an abnormal caliber and or aberration of the ducts.

We wanna look at cooper's ligament changes.

Are they straightened or thickened edema, which has increased echogenicity of the surrounding tissue, A reticulated pattern of angular hypo coic lines architectural distortion where there's disruption of the normal anatomical planes.

Skin thickening where there's focal or diffuse skin thickening.

Normal skin is approximately two millimeters or less in thickness except in the peri areolar and the lower portions of the breast.

We also wanna look for skin retraction or irregularity.

And this is where the skin surface is concave or ill-defined with appearance that's pulled in.

And here are some examples.

You can see that these ducks are enlarged and they have areas of focal narrowing and contain some internal debris.

Here.

We have edema where we see this anti coic or very hypoechoic fluid tracking along all the fascial planes.

And here we see the Cooper's ligaments are destroyed by this mass.

So whenever we see Cooper ligaments, we really have to be concerned about malignancy.

I think that when we see destruction of the Cooper's ligaments, that's telling us that that mass is very aggressive and is most likely going to be a malignancy.

When we see edema, I think in our practice we see edema more from patients that have congestive heart failure or hepatic or renal disease.

As opposed to patients that have a malignancy architectural distortion.

Here you can see that the normal anatomy the breast is distorted skin retraction.

Here you can see that the skin is concave.

And what I like to do is I like to use a lot of gel to fill in that concave portion and apply very little pressure so that we actually get an image that documents how much the skin is retracted.

So we can compare that in the future.

If you wanted to get a little bit better imaging inferiorly, you may want to push then, but you'll lose that appearance of the retraction.

For skin thickening.

What I like to do is do a dual display and you can do the normal and the abnormal side in the same location.

So you can compare the skin thickness from one breast to the other.

And I find that very helpful for calcifications, where you look at macro calcifications, and these are greater the Ann are equal to 0.5 millimeters in size, and these calcifications usually shadow, and we can have microcalcifications out of a mass, or we can have the microcalcifications within the mass.

And these small microcalcifications outside of a hypo coic mass are very difficult to identify because they do not shadow.

However, if these microcalcifications are in a very hypo coic mass, they're often seen as pun date foci within the lesion.

And here are some examples of a macro calcification where we see shadowing.

And here we have microcalcifications that we can see within a hypo coic mass.

And in this case you can see there's several white dots, and here's a macro calcification with shadowing, but there are multiple other small white dots in the normal breast parenchyma, which is very hard for us to determine.

If these are really calcifications ultrasound is very unique.

It allows us to actually characterize lesions that with specific features and can classify them as benign or malignant.

Special Cases in Ultrasound

So clustered microcysts are a cluster of tiny anti coic foci each smaller than two to three millimeters in diameter with thin less than 0.5 millimeter intervening sep day with no discrete solid components.

Complicated cysts mostly characterized by homogeneous low level internal echos.

Complicated cysts may also have a fluid filled or fluid debris level that will shift with the change in position.

It's very important to distinguish complicated cysts where there is no solid component that's fixed to the wall versus complex cystic masses, which have a solid component and do not move with repositioning of the patient.

We can look for a mass in or on the skin.

These are usually clinic clinically apparent and may include sebaceous or epidermoid occlusion, cysts, keloids, moles, and neurofibromas.

And here's an example of a clustered microcyst with these very fine septations.

Here's an example of an epidermoid cyst as a mass in or adjacent to the skin line.

And here's a complicated cyst that has some debris inside a cyst.

Additional special cases include foreign bodies, and these may include marker clips, coils, wires, catheters, sleeves, silicone and metal or glass related to trauma.

We can also see lymph nodes.

They can be either intary or in the axilla.

And these are normal findings within the breast.

And lymph nodes resemble small kidneys with an echogenic hilas and a hypoechoic surrounding cortex.

And again, they're found normally in the breast, including the axilla.

And just because you find them does not mean that they are abnormal.

So here's an example of an inter mammary lymph node.

One of the things you want to do to make sure it's a normal lymph node is that the hypoechoic cortex is relatively uniform throughout its appearance.

Again, you're gonna have an echogenic fatty hilum.

Here's an ular lymph node, which has the same appearance.

And here I've turned on color.

And that's to show you that in a normal lymph node, the artery and vein enter through the hilum and have a pattern just similar to a kidney.

This normal aberration, if you ever see blood flow coming through the cortex from out instead of through the hilum, those lymph nodes are abnormal and then need to be biopsied.

And here's an example of a foreign body a needle taking a biopsy.

Vascularity Assessment

We also want to assess vascularity, and there may be no vascularity present or it's not assessed that the vascularity can be present within the lesion.

It can be present immediately adjacent to the lesion, or we can have diffused increased vascularity in surrounding tissue.

So here's an example.

We've got the color box turned on and there's no blood flow.

You do wanna make sure that you have your settings appropriate and that you get blood flow.

And usually we try to get at least a dotted blood flow, not in a lesion, but in the surrounding tissue to make sure that our settings are correct.

In this case, we've got blood flow surrounding the lesion and not within the lesion.

If these vessels tend drape around the lesion, that's more likely a benign lesion.

Here we have blood flow within a lesion, and this does not necessarily have to be a malignant lesion.

In general, benign lesions have blood flow that surrounds the periphery of the lesion before entering.

In malignancies, we often have vessels that are perpendicular to the lesion giving more appearance of a malignancy.

And we can have diffuse increased vascularity in this case with mastitis.

BIRADS Assessment Categories

So after you've found your lesion and you've characterized it using the words we've just described, the next portion of the birads is the assessment category.

And there are seven assessment categories.

These include category zero, incomplete, category one, negative category two.

There is a finding, but it is definitely benign.

Category three, there is a finding and is probably benign and must have a less than 2% probability of a malignancy.

Category four is a suspicious abnormality, and these are subclassified into four A, four B, and four C with the percentages of malignancies listed here.

Category five is highly suggestive of malignancy greater than 95%.

And category six is a known cancer.

Reporting Guidelines

Your report should give a concise summary of the clinical history comparison to previous ultrasound examinations.

You should tell the scope of the exam if it was a targeted exam or a survey exam, and you want analysis of significant lesions or findings.

You wanna see the type of brush tissue lesion size, given at least two dimensions and the location.

I prefer to give the clock face and distance from the nipple.

There are other methods you can use to describe where the lesion is.

You want to give us succinct characterization of abnormal findings.

And then you want to do correlation of the ultrasound with mammography or MRI if available.

And then you want to give your overall final assessment using the birads classification that we just discussed.

And from that BIRADS assessment you want to give your management recommendations and those flow based on the BIRADS category.

So if you have a BIRADS three lesion, you'll probably recommend a six month follow up.

If you have a BIRADS five lesion or BIRADS four lesion a biopsy is usually recommended.

Important Points and Conclusion

Important points to remember is that the most diagnostic information is from the B mode image.

So you really want to have good technique and good equipment is a must.

You want to documentation is important for comparison to other modalities and follow-up examinations as well as medical legal issues.

And remember, when you're doing comparisons to other modalities, always confirm the position of the lesion, the background tissue.

And by that I mean is if you have a mammogram and the lesion is in fatty tissue.

You should also find the lesion on ultrasound in fatty tissue.

And again, correlate the size and shape of the lesion.

Should be the same on both modalities.

And if they're not, you have to be concerned that the lesion that you have does not correspond to the lesion on the other modality.

Probably the hardest part of this is the a birads assessment.

Some are pretty easy if you have a simple cyst, it's a birads two, if you have a speculated mass with blood flow and shadowing, it's a birads five.

However, the majority of these lesions are in the middle.

And experience is very helpful in determining the appropriate assessment category.

You should always follow up your assessments to see how you've done.

And I gave you the percentage percentages of lesions that should be malignant for each of these categories.

I think it's very important that you do your follow up and realize if you're over calling, under calling cases.

So for example if you're calling lesions BIRADS three, and you notice that 10% of those are malignant, you need to readjust your scale because we should not be calling greater than 2% of your lesions.

Excuse me, no greater than 2% of the lesions you call birad.

Three should be a malignancy.

And actually, in most studies, that should probably be less than 1%.

So in conclusion, BIRADS is resulted in a uniform reporting system that we all understand.

Lesion characterization requires attention to detail and looking for subtle findings.

If you go do a good job, your surgeons will want you to do the workup.

I think that when we started, our surgeons were a little reluctant to have us do biopsies.

And as we've done a very good job also evaluating lymph nodes, also looking for other lesions and we can present them with a very high quality examination and provide them the information they need.

I think this eliminates a lot of problems with the disagreement between who should be doing what.

Lesion characterization and recommendations is not enough.

You need to look at the whole picture.

So, for example, whenever there's a cancer, we really search for abnormal lymph nodes and biopsy those if needed.

And again, if we also look for a second cancer to make sure we do a complete study.

Thank you for your attention.