Breast US: Image Acquisition Pitfalls, and Bi-Rads - SD

Introduction

I am Dr. Ellen Mendelssohn. I'm professor of radiology at Northwestern in Chicago, where I am also chief of the section of Breast and Women's Imaging.

And I'm going to be talking about one of my favorite topics, which is breast ultrasound, how to do it, what some of the pitfalls are in performing it, and how to interpret it using BIRADS, the breast imaging reporting and data system, which has been developed by the American College of Radiology.

I'll be talking today about breast ultrasound image acquisition pitfalls and BIRADS.

Objectives

And the objectives of this presentation are to discuss acquisition of high resolution ultrasound images, to present criteria for assessing the quality of images, to exemplify some of the pitfalls and suggest ways to avoid them, to review the purpose and the goals of BIRADS, which stands for breast imaging reporting and data system, present feature categories and their descriptors within each of these feature categories, and to look at future possibilities for feature analysis and increased accuracy in interpreting breast sonograms.

Past Limitations of Handheld Ultrasound

In the past, there have been limitations of handheld ultrasound that held ultrasound back, and they consist of inadequate spatial resolution, a small field of view, five centimeter square at most, acoustic speckle, noisy image, poor inconsistent scanning technique, inexperience with ultrasound, depiction of breast anatomy and pathology, and lack of a standard for lesion analysis and reporting.

And all of this has been rolled up into the phrase operator dependence.

Current Equipment

Currently, we have wonderful equipment. We have high resolution broadband with linear transducers to use for breast imaging, and they range sometimes between 17 in the near field and five megahertz deeper into the image. 15 8, 12 5 or higher, or a center frequency that should be higher than eight to nine megahertz volume acquisition is now possible with matrix probes and resolution in the scan planes ranges from 0.1 to 0.5 millimeters for those of you who do mammography also, that range in size is the diameter of a micro calcification.

We can't, with ultrasound assess the morphology of microcalcifications, but we can see them better than we had in the past. With ultrasound, the penetration of these broadband with probes goes down to five to six centimeters, and with good positioning, you usually can see from the skin all the way down to the pectoral muscle.

Electronic focusing has been a real boon, and the footprint of these transducers ranging from 38 to 50 millimeters allows for efficient survey scanning in patients for whom you are doing that type of study.

There is also a wide field of view in a single sweep that's available on many high-end systems, which enables you to see the full scope of breast tissue at that site.

Spatial compounding and coded harmonics have also been helpful in speckle reduction and in cleaning up the image, the transducers for breast are illustrated here, where you can see in the image a fibroadenoma, its margins clearly outlined, and all of the different tissue components are visible from the skin at the top, and it's normal at two millimeters of thickness.

Deep to that, the fat lobules outlined in Cooper's ligaments and connective tissue arcs deep to that, the fibro glandular tissue with tiny normal sized ducts, and then going down deeper, the band of pectoral muscle and then into the thorax.

Scanning Technique

The technique that we recommend is to use for the skin or lesions within the skin and offset, no matter how high resolution your transducer is, you still need to focus with an offset to allow you to see that layer of tissue optimize the gray scale.

One way to do it is with a high dynamic range so that you can differentiate among the different breast tissue types without losing contrast resolution.

Set the gain and the TGC for each patient, gain settings that are too low, and contrast settings that are too high will lead to misinterpretation of solid masses as cysts, select the appropriate field of view, and the default settings of most of these transducers is at three centimeters.

If you have too shallow a field of view, your image will be very noisy and you'll have a pseudo magnification, and the lesions characteristics will be degraded. Too deep. There'll be no information in the lower portion of your image, always, as with any imaging technique and modality, do at least two perpendicular views of the area of interest.

Here, as you can see on the left skin, deep to skin is a small layer of fatty tissue and a large band of muscle within the muscle. And shadowing posteriorly is something that you might, if you didn't know the breast anatomy you might feel is a lesion that needs biopsy. It doesn't.

You turn your transducer orthogonally, and what you see there is the elongated rib. And as you scan up and down the outer part of the breast or the far medial aspect, you'll be able to see in sequence these ribs.

Knowledge of the anatomy helps a lot. Knowing that you are in the pectoral muscle will prevent you from making a mistake of a breast lesion that sits within the muscle, high contrast black and white.

I hope all of you will look at this image and think about what the diagnosis might be. It is not a cyst, and here in another patient is a similar image, and it's the same diagnosis.

These masses are poorly differentiated, invasive ductal carcinoma, and if your contrast settings are too high and all you see is black and white without any shades of gray, you will make a mistake. This would be a very serious one.

When you have shades of gray, you're able to see actually the marginal irregularity and the different types of tissue components. Both of these masses are living within fibro glandular tissue, and it's very difficult to make out. In the image.

On the left dynamic range is the ratio of highest to lowest amplitudes that can be displayed. It is a logarithmic expression, and its units are decibels.

A wide range permits the best differentiation of subtle differences in echo intensity, and it's preferred. You really do wanna see all of the different shades of gray.

The trade off, however, is that if your dynamic range is too high, you won't have enough contrast to be able to make out the tissue differentiations and the lesions characteristics, your image will be grayed out, and that also is not an optimal type of image.

The newer systems have increased dynamic ranges up to, or even greater than a hundred decibels, but it's different for each manufacturer.

And hard spatial compounding is angulation of the ultrasound beams into tissue anywhere from three to nine beams, and they intersect in the center of the image and effectively reduce speckle and noise so that you can get a much clearer marginal definition of a mass seen.

Centrally in this image is a spot compression view of a cancer with its irregular shape and margins on the left. With conventional imaging where the acoustic beam descends directly vertically into tissue and ascends back to the transducer, you see a lot of noise.

It's a very speckly image, difficult to see the cancer really at all, and its margins are partly obscured by the speckle in the image on your right. The area is much better defined.

Here you do see this vaguely triangular mass, a small one. You can make out all of the different tissue components of the image fat, and then where the mass is seen in the fibro glandular tissue with the muscle deep to it.

But it's much easier to see in an image without the noise. The marginal definitions and margins and shape are two of the most important features that enable you to assess a lesion in terms of its likelihood of malignancy.

Pitfalls in Shadowing

Shadowing can be artifactual or real, and one of the features of masses that are malignant can be shadowing from the mass. About 60% of cancers will shadow, and here on the left is an image which shows a tiny little area, perhaps four millimeters in diameter, and there's some shadowing that you can see within this posterior to this mass.

How can you resolve that issue of artifact versus real versus a real finding? Change the patient's position slightly vary the amount of compression, either increase it or decrease it slightly, and change the probe angle to reduce the artifact.

If you can, don't change the parameters too much or you'll be in a different area of the breast, but just slight angulation or change in compression can help you.

You can see here in the image on the right that it's in the same position. And now what you see is that the shadowing we saw on the left is attributed to the intersection of curved surfaces.

It's refraction shadowing at that intersection, and here we have it resolved. So we can say that this area is normal.

Wide Sweep Imaging

The wide sweep is illustrated here. This is panoramic or wide field of view imaging. This patient had had a surgical procedure, and there's a large seroma in her breast, but you get the sense of the entire organ from that wide sweep.

Technical and Clinical Pitfalls

What are some of the pitfalls in scanning technical pitfalls or improper gain settings, fields of view that are smaller than or larger than they should be over or under compression and use of too little coupling gel, which can lead to artifact.

You'll have poor contact and some shadowing that is artifactual.

Clinical pitfalls. Real-time imaging and targeted physical exams should be done when one is using handheld breast ultrasound for evaluation of lesions, failure to palpate the mass or the or a thickening to direct the ultrasound image can lead to a misinterpretation and also to missing the lesion.

No correlation with mammography or MRI when that is what had prompted the ultrasound exam and how positioning there can change measurements and clock face notations.

Then the MR exams are done prone, and there the breast probably has its longest anterior, posterior dimension, and ultrasound is supine there. The breast anterior posterior dimension is truncated or shortened.

So very difficult to compare. Those two errors can occur when there's poor correlation with the mammographic or MRI findings that have prompted the ultrasound and the positioning can alter measurements enormously.

What happens is the anterior posterior dimension changes, it is elongated in MRI where the patient is positioned prone, and the breast is dependent in ultrasound. The anterior posterior dimension is truncated, shortened, and for mammography with the patient upright, that anterior posterior measurement is probably closer to MR than to ultrasound.

Another possible way to go astray is to use the wrong transducer, using one that's too of too low frequency. A five megahertz transducer, for example, shouldn't be used. It may make recognition of a mass more difficult and may make its assessment impossible.

Mistaking normal anatomy for a mass. An example would be fat lobules and ribs, and I've shown you the ribs and the possibility for error there.

Interpretation, cyst, solid differentiation, mistaking cancer for cyst. Optimize the gray scale there and make certain that you do it for each patient and for each lesion.

New Developments

New developments, matrix and volume acquisitions are now available on many of the probes and high-end systems. 3D four D at high resolution perfusion imaging for monitoring response to neoadjuvant chemotherapy is under study elastography.

Will that have a role to play, possibly as another feature category for BIRADS, and at least two methods are being developed now for clinical use in assessing lesion stiffness.

Here is a coronal view, which shows you long spicules emanating from this cancer. And the architectural change that is affected, and you can see it so well on the coronal view here, a volume acquisition with reconstruction.

And the volume of this benign lesion is eight and a half ml.

Another new development is automated ultrasound with a wide field of view Here. Up at the top left is a scan showing that was obtained with a 14 and a half centimeter field of view.

It is the wide field of view, but it is automated as the transducer sweeps through the breast in mammographic positioning with reconstruction.

As you can see in the other images, here is a coronal view, and the other images are also reconstructed.

The study was done to evaluate a possible area of thickening that may have harbored a mass, and what we see here on the mammogram marked by a little bb, and going to that area with the automated ultrasound is a prominent area of fibro glandular tissue with a very small layer of fat over it.

Its shape and size correlate nicely with the mammogram. There is no mass here going to breast anatomy.

Breast Anatomy

The breast is a modified sweat gland that lives in a fascial envelope. The anterior portion of the envelopes that's just beneath the skin. And the posterior aspect of this fascial plane lies just anterior to the pectoral muscle.

It's the superficial pectoral fascia and Cooper's ligaments that are the connective tissue support for the breast. They straddle the two layers.

You can see here in a diagram that reproduces the ultrasound positioning cut at various planes. The ductal structures usually between seven and 20, very variable, cut at different levels.

Here you can see the lobules and the connective tissue that surrounds each one of these ductal structures that these segments loosely define the anatomy of the breast.

The significant structure, the site of origin is the terminal duct lobular unit, and the blood supply to the breast is through the axillary artery, and the venous plexus is posterior to the nipple with drainage into intercostal, axillary and internal thoracic veins.

Lymphatic drainage, and it's important to know that it's extremely variable is predominantly more than 90% through the ipsilateral axilla.

There is crossover, there are crossover lymphatics to the contralateral side, and they would drain through to the contralateral axilla.

There's also drainage into the internal mammary lymphatics. Another small percentage here is sonographic Anatomy illustrated.

Sonographic Anatomy

The skin is normally a two millimeter complex defined by two echogenic lines, with a thinner hypoechoic layer between them. The total distance is two millimeters, except in the periareolar region and in the inframammary fold area.

Just beneath that, you can see the anterior portion of that superficial pectoral fascia as a almost a double line adjacent to the skin beneath that fat, the fat lobules, which are outlined.

They're frequently oval, and they elongate in another projection interspersed here. And as you look at these images, think to yourself what the mammogram might look like.

Mammographically, this would be a heterogeneous breast because you have fibro glandular tissue interspersed with fat lobules going from the skin all the way down to the pectoral muscle.

The pectoral muscle in this image is not in focus, but what is in focus are these very fine ducts. Here's a branch point of one of the ducts and another one.

This is the terminal duct lobular unit. And what you see here are the acini that live within the lobules, and they're dilated here in little microcystic structures.

The terminal duct lobular unit gives rise to cancers, both ductal and lobular, as well as to benign lesions, cysts, and fibroadenomas.

Let's look at the anatomy that is posterior to the nipple. The nipple and areola are pigmented, and they're stratified squamous epithelium at this site and smooth muscle bundles for contractility.

The areola contains sebaceous sweat and accessory glands of Montgomery. Sometimes they connect to the lactiferous ducts.

The mammary structure itself, 15 to 20 or fewer lobes branching tubular alveolar glands, and each lobe ends in a lactiferous duct that opens from the nipple.

If you scan the area behind the nipple and ways to do it are to move the areola out of the way a little bit, and to scan and angling back behind the nipple so that the crevices of the nipple itself don't obscure the area posterior to.

It's difficult to scan, or use an offset. As you see here, there's a very thin offset.

What you'll find in most patients is fat lobules that flank an area like this, a conical area of fibro glandular tissue that contains ducts. Here is another image as well.

The fat comes up to this sort of triangular area, and you can see it clearly. If you angle back or use an offset, the axilla is very important.

And in patients, certainly those who have newly diagnosed cancers or those who have masses that you suspect are malignant, it's important to look at the axilla.

Axillary Anatomy and Lymph Nodes

The axilla itself is a pyramid shaped space that exists between the upper extremity and thoracic walls. The base of the axilla is fascia, and its apex opens into the posterior triangle of the neck, the cervicoaxillary canal.

It is formed by bony structures. The clavicle anteriorly, scapula posteriorly in the first rib medially, its muscles are pectoralis major and minor, and you can see them cut here.

This is pectoralis minor and over at the larger pectoralis major subscapularis teres major latissimus dorsi posteriorly, and then the serratus anterior which wraps around, and its feathery interdigitation can be seen anteriorly.

Neurovascular bundle is a very important one. The axillary artery vein and vein exist, and you can see them cut here, large structures and the important brachial plexus and long thoracic nerve.

The axillary node levels important to assess are defined by relationship to the pectoralis minor. And level one nodes are lateral to or below the pectoralis minor, level two, deep to or behind the pectoralis minor.

And level three, which is not often dissected, are medial to or above the upper border of the pectoralis minor, deep in the axilla.

If level three nodes are dissected, there's a higher likelihood of lymphedema as a complication. Most often. Now in breast conservation procedures, the axilla is not entered unless a sentinel node is positive.

Here is what a normal lymph node looks like. This is an intramammary lymph node, which has a thin cortex.

It's hypoechoic, and the fat within the hilum of the node is echogenic just as it is within a kidney. And this little node looks just like a miniature kidney.

The fat within the node is different from the fat lobules that you see beneath the skin which are hypoechoic and more similar in their echogenicity to the cortex of the node contrast, the normal node with this very thickened cortex, and a bulge in the cortex where you see the arrow in a patient who has breast cancer which has metastasized to the node.

So a very thick cortex, greater than three millimeters in its thickness here, it's nearly a centimeter. There is still fat in the hilum of the node, so that's not a good sign of whether a node is or isn't infiltrated.

And when you see a cortical bulge and want to biopsy a node either with fine needle or core biopsy technique, go for the area of that bulge.

What you can't tell from an image is whether a node is metastatic or whether it's benign. And hyperplastic here, very abnormal nodes, a huge node in the axilla of this patient.

And this sonographic correlate just shows enormously thickened cortex. Yes, there is some fat still in the hilar areas, and there are two or three nodes that are abnormal here.

These nodes were biopsied. This patient did not have breast cancer, and the nodal appearance was due to tuberculosis. She had active TB.

Correlating Ultrasound with Other Imaging Techniques

Correlating ultrasound with other imaging techniques can be difficult. And it's important to keep in mind. Certain principles look for the same pattern of fibro glandular tissue and fat, the anatomy of the surrounding area.

The sizes of the areas that you are correlating should be similar, whether it's fibro glandular tissue or whether it is a mass, account for change in the position.

As we mentioned earlier, from upright with mammography to supine or supine oblique with ultrasound to prone with MR, where the anterior posterior difference at distance is longest.

Here we have a patient upright in the mammography, and this is her upper breast, lower, she has a linear area of soft tissue density, and then anterior to that a central linear area going down towards the nipple.

There's fat anterior to it and posterior to both of these areas that are linear. And what do we see when we go to ultrasound? We are thinking about possibly a mass, perhaps invasive lobular carcinoma.

But we don't find anything. Can we correlate what we do see in that same area? And I think you can, at least in this image.

Here in the upright mammography image, the head is above and the feet below the image we have on the sonogram head to the left and feet to the right, a 90 degree turn.

What you do see here is a thick linear band of fibro glandular tissue moving anteriorly just the same way that you had seen it on the mammogram is a thinner linear area, and then the fat above it, and just beneath the skin and posterior to it.

And now we know we are down to the pectoral muscle, and we've seen that whole area sonographically. You don't see the pectoral muscle on the mammogram because the anterior posterior distance is longer, it shortens and for shortens with the ultrasound.

Another example is provided by this lymph node. Patient would come for her baseline mammogram and perhaps have two or three little nodules up in the upper outer quadrant of one of her breasts.

You can see that she has a more fibroglandular breast. And here's a nodule. You get extra views and still can't demonstrate mammographically the fat within the hilum of this node.

You think it's a node, but you have to be able to prove it. So you go to ultrasound, and yes, now you find a nodule that's the same size, and it has all of the hallmarks of a node.

The cortex is normal, and within the central portion of it is a fatty hilar area.

How do you know that this node is the same thing as what you are intending to study on the mammogram? Well, look at the area around it.

You should look at the type of anatomy and on the mammogram you have embedded in fat the lymph node, same thing on the mammogram, on the sonogram, and anteriorly is fat and deep to this.

And again, turning 90 degrees from the position that the woman was in for mammography, it correlates with the fibro glandular tissue behind it.

Indications for Using Ultrasound in Breast Evaluation

What are the indications for using ultrasound in breast evaluation, evaluating and characterizing palpable masses and other breast related signs or symptoms.

Evaluation of suspected or apparent abnormalities detected on other imaging studies such as MR or mammography as the initial imaging technique of palpable masses in women under the age of 30, and in lactating in pregnant women, evaluation of problems associated with breast implants.

And now, and this is from the American College of Radiology practice guideline for the performance of a breast ultrasound examination, which was approved in May of this year and is effective as of earlier this month, ten one 2007 currently.

And this was is as a result of practice patterns that are changing with respect to application of ultrasound, evaluation of breasts with microcalcifications or architectural distortion that's suspicious for or suggestive of malignancy when the background tissue is dense, fibroglandular, so that an underlying mass, which may be obscured on a mammogram might be detected with ultrasound guidance of breast biopsy or other interventional procedures and treatment planning for radiation therapy.

Again, however, as not yet indicated are evaluation of the axilla for occult lymph node metastasis in patients with newly diagnosed breast cancer.

And the efficacy of ultrasound is a screening study for occult masses in dense, fibroglandular breasts of high risk women or women with newly diagnosed or suspected breast cancer.

Now, these are areas of current and very active research, and there are many practices in the United States that do use ultrasound for a axillary evaluation, and also for extent of disease in patients who are newly diagnosed with breast cancer, calcifications and dense, fibroglandular tissue can be seen.

Here they are. This is a hypoechoic mass in two projections, and these little echogenic dots are not acoustic speckle, they're clumped or clustered in and around the mass.

And the architectural distortion that is caused by this mass can also be appreciated. The mammographic correlate is this.

This patient had had a lumpectomy and radiation therapy for carcinoma, and three to four years later she developed a recurrence near the site of the lumpectomy.

BIRADS Lexicon

Turning to the BIRADS. It's a lexicon and it is an organized approach to providing a framework for pattern recognition, which is really what we do with all of our interpretations, any modality, any organ system.

The American College of Radiology Breast Imaging reporting and data system was developed for ultrasound. At the same time. The fourth edition for mammography was published in 2003, at that same time that a BIRADS lexicon had also been published for MRI and wherever we could, it was a small committee, and a consensus document that was developed and validated at a number of different meetings.

The we borrowed or took from the mammography lexicon, the terms that would apply to ultrasound as well.

First, we looked at the type of tissue background echo texture, and that correlates with mammographic density. And for ultrasound, we have homogeneous type of tissue, uniformly hypoechoic fat lobules, or echogenic connective tissue arcs, or composed of echogenic, fibroglandular tissue, and a thin layer of subcutaneous fat Here, mammographically, extremely dense fibro tissue in both breasts.

And what it might look like with ultrasound can be seen here. This is up in the axillary region of this patient's breast, and it is a homogeneous area of echogenicity.

There's a little scalloped layer of fat anterior to it, deep to it. You can see some of the muscle. And then a rib posterior to that background.

Echo texture can also be heterogeneous, where you have fatty or diffusely variable areas of different echo textures, many small areas of increased or decreased echogenicity.

And part of this can be due to scanning technique. And one of the questions that we sought to answer with our multicenter trial of breast ultrasound for breast cancer screening in women of high risk with dense breast tissue is whether or not this type of tissue lowers the sensitivity of ultrasound.

The study that I mentioned is ongoing and concluding now, and it involves mammography and ultrasound performed and interpreted separately, with possibility for two follow-up studies at annual intervals for a total of three exams.

And the study is an American College of Radiology Imaging Networks study, with involvement of the National Cancer Institute.

Correlative imaging interpretation is important here as well. This patient has heterogeneously dense tissue that you can see. Mammographically. Does this limit the ultrasound as the dense tissue limits mammography?

Well, if you look at the sonogram, what you see is some fat hypoechoic, and then deep to that interspersed in an area of fibro glandular tissue, as you see here in this mammogram where the patient has been turned, where the mammogram has been turned to show you similar positioning as you would engage the patient in ultrasound head to the left feet to the right, as you would find in the sonogram.

And what you do see is an area that's hypoechoic. It has an irregular shape. If you saw it in another view, you would suspect that you were dealing with a small cancer here.

And yes, possibly it does provide some limitation on the interpretability of the sonogram, but it does correlate well with the mammogram.

So there is a possibility the data are not there yet. But there are probably some inherent limitations to sonography looking at the feature categories that we use for interpretation and for assessing the likelihood of malignancy of a lesion that we are evaluating with ultrasound.

The top three are the top three, and they should be taken together, shape, orientation, and margins. The other feature categories also contribute to the assessment, and they include boundary zone, echo pattern, posterior acoustic features surrounding tissue calcification, special cases, which I'll exemplify and vascularity, starting with the surrounding tissue, because this is where we can see whether or not the mass or other lesion that we are interrogating with ultrasound has an effect on the surrounding tissue.

We will look at some of these effects only two possibilities here. Either there is no effect or else an effect that you can identify where you might have changes in the ducts abnormal caliber or arborization Cooper's ligament changes where the ligaments are pulled in, thickened, straightened by a cancer, perhaps edema, where you would experience increased echogenicity of the surrounding tissue, unrelated to technical factors, reticulation, where a network of hypoechoic lines intersect with each other in a non ductal radiating pattern.

Architectural distortion, disruption of normal anatomic planes, skin thickening, which can be focal or diffuse, and skin retraction or irregularity where the skin surface is concave or appears pulled in.

This patient has a mass, and most of it is hypoechoic. It has a surrounding halo of echogenicity higher than that of the fat, and then a few little focal areas.

Is this a cancer? It certainly is your first choice in terms of diagnosis. This patient happened to have several other areas just like this, and lymph nodes that were enlarged in both axillae and it was sarcoid, but it's a good example of how a mass can affect the surrounding tissue by invading tissue planes.

And here you can see connective tissue being pulled into the mass. This patient has had surgery in this location, and there's a V here at the skin where the scalpel entered, and the path of the scalpel is down here to the tumor bed deep in the breast.

And this shows skin thickening that is focal.

Definition of a Mass

What is the definition of a mass? A mass occupies space and should be seen in two projections. And this is from mammography.

The BIRADS for mammography can be distinguished from normal anatomic structures, ribs or fat lobules, using two or more projections and real time scanning.

Feature analysis is the way to interpret, to analyze masses and to assess their likelihood of malignancy and combined multiple features are the best predictors, don't seize any particular one.

The most important taken together are shape margins. And for ultrasound orientation, when we constructed the BIRADS, and it's a living document that will undergo change fairly soon, perhaps.

But what we wanted to do was to provide enough descriptors and categories to make it reasonable, but not so many that the choices would be difficult.

So the lumpers and the splitters were compromised are there was a compromise between lumping and splitting shape only three possibilities.

Oval, which is elliptical or egg shaped, can include two or three undulations here, gently lobulated or macro lobulated. And we're thinking of fibroadenomas, round, spherical ball shaped circular or globular if it's not round.

And if it's not oval, it's irregular by definition here, an oval mass in two projections. It's oval and parallel, parallel being parallel to the skin.

And of course, you've already looked at the margins and they're circumscribed. This is a benign mass. It was biopsied, it's fibrocystic change.

This is a small calcification that you can see within the mass. It's not shadowing because it occupies too little percentage of the beam.

How about this? An oval mass perhaps, but turn your transducer orthogonally and it elongates. What you are seeing here then is a fat lobule.

And not to mistake it for a mass, and you must see it as a mass in two projections shape.

Here's a round mass in a cystic structure, so round in round. And this was a thrombus in a patient who had a cyst that was had an attempted aspiration refilled, and there was a clot that formed within it.

Irregular, if you wanna call it roundish or or a sort of oval, call it irregular, it should be round or oval. And by definition, again, if it's neither, then it's irregular as this mass is.

Note some of the other attributes of this mass. There's a halo, a slight halo around it where the tissue planes are somewhat blurred.

It's a heterogeneous mass, posterior to it is an area of enhancement, and its shape is irregular.

Orientation

Now, the orientation of this mass is horizontal in its longest axis, which can be a benign feature and is a feature of most benign lesions, but is also present with cancers.

Define, with reference to the skin line. A parallel or wider than tall orientation is a feature, again, of some benign masses. Most fibroadenomas, but many cancers also have this orientation, diagrammatically an oval mass whose long axis is parallel to the skin.

We are imagine a line beneath the title and the anterior posterior axis is shorter as differential considerations. You have fibroadenoma cancer cyst and sebaceous cyst, so both fibroadenoma and cancer.

Here, a mass whose long axis is parallel to the skin. Its shape is oval, and its margins are circumscribed. All benign findings.

This is a fibroadenoma, it's typical of it. And note too, that it's not doing anything to the tissue planes around it. It's just living there without having any effect.

Compare this mass in two different projections, which is sitting just above the pectoral muscle. And there's posterior acoustic shadowing here.

Whether or not the pectoral muscle is affected by this carcinoma is not discernible from ultrasound. You really cannot tell. And in this case, it was not.

But note the shape. It's irregular. The margins here are not circumscribed in any way, shape, or form, but it is also as was the fibroadenoma parallel to the skin.

This is an invasive lobular carcinoma, which proceeds through breast tissue often in single file, and it can hide there until very large masses develop.

The other orientation is not parallel, where the anterior posterior vertical dimension is greater than the transverse. And synonyms here are vertical or taller than wide and round masses are also not parallel.

And that where you have the anterior posterior and the horizontal measurement equal. What's missing from this differential listing? Fibroadenoma, we've taken it out.

Cancer exists in both cyst and abscesses can be present in this differential listing, an ill-defined mass whose shape is irregular eating away at the tissue planes here.

The margin of the fat lobules is blurred as you see over here. And this mass has a vertical or non-parallel orientation.

It's invasive ductal cancer margins. Two possibilities. Margin is the edge or border of the lesion. That's the definition.

And it's either circumscribed or it's not here. A well-defined, smooth, distinct rim around this simple cyst, which is anechoic, has posterior acoustic enhancement, and is of oval shape.

Again, note that there's no effect on the surrounding tissue that is caused by the presence of a cyst.

The not circumscribed possibility includes a number of other descriptors that could also have many, many additional ones. And most masses that are not circumscribed may have more than one of these descriptors that you could use to characterize them.

But the key thing is that the mass is not circumscribed. It doesn't matter whether it's angular or indistinct. It does matter that it's not circumscribed.

And we'll show some examples. This is a mass whose margins in these two projections are not distinct. They are not circumscribed, and there's an indistinct area between the tissue around it, which is being invaded by the mass, and the mass itself.

It's hard to say where the tissue, the normal tissue begins, and the mass ends. So that's poorly defined.

An example of another invasive ductal cancer here with angular margins, and there are some masses that are nearly circumscribed, but have an angle watch out for those, because those need to be biopsied.

They can be beguilingly circumscribed for most of their margins, but if they have any marginal irregularity, it increases the likelihood of malignancy and they require biopsy.

Angular part or all of the margins have acute angles. Microlobulated, mass here, more than three small short cycle undulations are characteristic. As you see in this mass.

This mass was biopsied, and part of it is a radial scar. The other part was fibroadenomatous, the anterior portion where you do see that it is circumscribed.

Again, note that the orientation here is not parallel, and that the margins are not circumscribed.

What about spiculation? We see it well mammographically here. We see it very well sonographically.

And the definition is that the margins are formed or characterized by sharp lines projecting from the mass as you see here. And the central portion of the mass is hypoechoic.

There's a large echogenic border, and its definition is not great between the echogenic halo and the surrounding fat lobules.

This looks like a cancer, carcinoma. It was a primary breast lymphoma. And there are others that we've seen that have a similar appearance with this large area of echogenicity.

Irregular or spiculated masses include carcinomas, surgical scars, fat necrosis, radial scars, abscesses, and the benign and rare granular cell tumors.

Boundary Zone

Another feature category is that of boundary zone. And there are only two possibilities, none where there's an abrupt border between the lesion and surrounding tissue.

And their examples might be cysts and benign masses with this sort of abrupt border echogenic halo, as I've just shown you with the lymphoma, where there's no sharp demarcation between the mass and the surrounding tissue and their cancers and abscesses would typify this type of boundary zone.

On the left is a short axis view of a complicated cyst. Here, low level internal echoes are present. All of the other characteristics of a cyst are present.

It is circumscribed. In the other view it was oval. There is posterior acoustic enhancement.

How about this mass? The anterior portion of this mass, its anterior margin is not circumscribed. There are short spicules and irregularities that you can see.

It is a vertically oriented mass, and it does have an echogenic halo. It's not prominent, but it's there and there is enhancement on your right.

You have a poorly differentiated, invasive ductal cancer on your left a cyst.

The scanning technique has to be meticulous. What about this mass? This mass has an irregular shape. It's orientation is not parallel, and its margin is not circumscribed.

There is an echogenic halo around it. Your first diagnosis or your likelihood of malignancy here is quite high.

This is not a carcinoma, however, this was a foreign body type of granulomatous reaction. Mimicking a cancer absolutely needed to be biopsied.

Echo Pattern

Another feature category, echo pattern and its definitions, anechoic without internal echoes, as you see here in this simple cyst.

Hypoechoic a new term complicated cyst, not complex cyst anymore, but complicated cyst. And the definitions there are low level internal echoes throughout, or a fluid, fluid level.

Here we have low level echoes throughout. These are two cysts. If you put a needle into them, you'd get sort of greenish. You would get greenish fluid out.

And here a small complicated cyst as well. With a fluid, fluid level that's sharply demarcated.

It's hard sometimes to tell those from benign, solid masses such as fibroadenomas. In general, if there are many simple cysts and complicated cysts, we follow those.

If there's a solitary mass with low level internal echoes, we would aspirate. And if there were no fluid obtained, we would do core biopsies.

Here is a mass seen, a soft tissue mass seen in the central breast of this patient. Sonographically. Two images here.

Hyperechoic, hyper echogenicity is a characteristic of fat. Is this mass a benign fat containing mass? No.

You can see mammographically that it's soft tissue density. If it were fat, a lipoma, which it can't be because the margins aren't sharply defined.

If it were a lipoma, it would be of low density mammographically. So the mammogram tells you no, you're not dealing with a lipoma.

And then as you scan through in real time, you see these hypoechoic crescent surrounded by the area of echogenicity. Margins are not circumscribed.

The mass is round and it's orientation is not parallel. This is another primary breast lymphoma.

Be very careful echo pattern. Isoechoic, same echogenicity as fat, and then complex cystic mass.

Here, as you can see, there's a cystic component, an area of hyperechogenicity in another little component that's complex appearing.

And this, as you can see here, is a mass shown tangentially in the spot compression view. And how nicely you can see it on the sonogram.

This is fat necrosis, mixed hyper and hypoechoic, where you have portions of a mass that are hyperechoic to fat and portions that are hypo or isoechoic without a cystic component.

Posterior Acoustic Features

Posterior acoustic features are still another category that we can add to the mix. And options are none where there's no change in the pattern posterior to the lesion enhancement, where there are increased posterior echoes and shadowing, decreased posterior echoes, excluding edge shadows.

Only about 60% of cancers will shadow a combined pattern of a shadowing and enhancement is also seen associated with some masses here.

Several examples, again, don't look so much first at the enhancement pattern, but shape margins and orientation a cancer with a little angulated portion anteriorly located, and a border zone that's echogenic.

Very nice enhancement. But this is carcinoma. This patient has a complex cystic mass. It's a galactocele.

The echogenic portion is the fatty portion of this fluid containing mass. That's milky if you should put a needle in into it and pull the contents out.

Also, enhancing, shadowing two views of a small carcinoma that is invading the tissue around it, blurring the fat planes, and its shape is irregular.

Its margins aren't circumscribed. It's orientation is vertical or not parallel.

Special Cases and Vascularity

Special cases and vascularity, these are perhaps the addendums. The special cases include microcysts, where there if you should biopsy. And one should, if there is a solid component, you'll find perhaps apocrine metaplasia, intraductal masses, masses inner on the skin, foreign bodies, lymph nodes that are either in the breast or axillary lymph nodes.

And then vascularity. And there were only three options there. Can't assess none or present panoramic view.

This wide field of view allows you to see the beating of this duct containing intraductal papillomas here bulging the duct walls. Here's another area of papillary tissue against the wall of the duct.

And at a branch point you can see the papillomas entering another duct foreign body.

This is the signature of silicone in soft tissue. This fade out, this dropout echogenic noise as it goes deep to the area of free silicone edge of the implant.

You see here, and some refraction shadowing. The fat is clearly visible above the implant.

And looking at the anatomy, you can tell that the implant is subglandular. It sits anterior to the pectoral muscle.

Here is the band of pectoral muscle. So this is extracapsular rupture with silicone in the soft tissues.

Looking at the vascularity, there is no way that you would know that this is related to an area of ductal carcinoma. Situ to, there is just vascularity there, here is a calcification and there was a small mass.

The mass was ductal carcinoma in situ two. It was not invasive.

Assessment Categories

Assessment categories, as we begin to put all of the data together, are similar to those that you would use for mammography. And in reporting, we try to correlate and bring together the mammographic and the sonographic findings.

So as with mammography, a zero is an incomplete assessment where you would like to have, if you've done a sonogram, mammography, or possibly MR. Or possibly previous sonograms.

What you see here in this spliced image is a large mass. The rest of it was also circumscribed. It's heterogeneous.

If you've seen a number of these, you might not need mammography to tell you what it is. The findings are benign appearing.

It's a circumscribed long axis parallel to the skin. It's not causing any distortion of the tissue around it. And its shape is oval, but if you don't, you would go to mammography.

And here it was palpable. Usually these masses are soft. And it's a leave me alone. This is a breast hamartoma that contains all of the tissue components of normal breast in an encapsulated mass.

As with mammography, we really need to give final assessments. It's not good enough just to provide a differential listing.

We should provide a final assessment that perhaps would dictate patient management. So we have one which is negative, and their routine follow up for age would be our recommendation.

Two is benign findings such as cysts. Again, routine follow up for age. Three is the probably benign assessment, which for mammography specifies a less than 2% likelihood of malignancy.

We're not certain a hundred percent if this is applicable to what we think are benign appearing masses with ultrasound, but most likely they are, we think of fibroadenomas and there are many practices with what looks like a fibroadenoma, as I've shown you here today on ultrasound, short interval follow up might be the recommendation.

And there, while there is no specific interval that's recommended just by convention at the current time, it should be approximately a six month follow up.

Four and five require some sort of intervention. Biopsy is what would ordinarily be done. And at the current time, again, the biopsy should be done. Percutaneously with imaging guidance, suspicious for malignancy is four, and there it's a broad percentage range from 2% all the way up to about 95% and five highly suggestive of malignancy.

Number six in the final assessments is known biopsy proven carcinoma. And there the patient may be seen following neoadjuvant chemotherapy.

And we may have assessed the response with ultrasound, and with other imaging techniques. Probably for that MRI as best.

The breast imaging report should combine the assessments of whatever imaging studies are done, combining mammography, ultrasound, and MR if it had been performed.

If the most suspicious assessment should dominate in the management recommendation, avoid tunnel vision.

A patient came to the breast center with a new mass. She was in her eighties and was a good historian. And we have on the mammogram reason for her mass.

She has a calcifying fibroadenoma that's fairly large sitting up in her upper breast. But she said the mass was new. We went to ultrasound.

She has dense breast tissue there and saw in the image on the right, a fibroadenoma here at the macro calcifications occupy enough of the beam to cause posterior acoustic shadowing.

Adjacent to it though was a soft tissue mass whose margins looked angular. Here in the short axis view, again, the fibroadenoma where the posterior shadowing, and then in real time scanning through the whole area of mass, we saw not in the area of the fibroadenoma, an irregularly shaped mass with margins that were not circumscribed.

And its orientation is parallel to the skin. The margins and the shape were enough to dictate a recommendation for biopsy, which was done.

And our patient had a carcinoma, and a fibroadenoma adjacent two very common lesions coexisting.

Summary

So in summary, check the technical quality of a scan before interpreting, before attempting to interpret. If you haven't done this, scan yourself.

Appreciate the importance of real time observations in breast imaging and in analyzing a case use feature analysis.

The categories that are most important and they should be taken together are shape margins and orientation. And you have for your use descriptors within each category, integrate ultrasound with the other modalities that have been used.

And also with clinical data. You are dealing with patients not lesions. Reach a conclusion. And that should be a final assessment.

And along with the conclusion should be a recommendation for management. Thank you very much.