Correlating Mammography, Ultrasound and MRI - SD

Introduction and Objectives

My name is Dr. Catherine Piccoli, and I am going to be talking about correlating mammography, ultrasound and MRI.

The objectives to this talk are to review how routine mammographic views are required to illustrate the location of the lesions based on the routine mammographic views, and to show examples of how we correlate the different imaging modalities when we see an abnormal lesion.

Routine Mammographic Views

The routine mammogram is made up of two views. One is the cranio caudal view where the CC view and compression is applied from the top of the breast with the detector at the caudal surface of the breast. And this is a schematic of that looking down on the breast. And this shows how that is performed.

The breast is placed on what we call the Bucky, and then the compression plate comes down on top of the breast. And this is a mammogram in the CC projection. The lateral aspect of the breast is always placed on the upside of the image. The medial aspect of the breast is placed on the downside of the image and the marker on the film. The right CC is always placed at the lateral aspect of the breast.

The MLO view is the second routine mammographic image that is performed, and that is done by placing the breast on the compressor with the compression plate rotated at about 45 degrees. And on this view, we see more lateral tissue relative to the CC view. And this is sort of a schematic showing how that is accomplished. And here we see a patient with her breast in compression. And you can see that the Bucky is tilted at about a 45 degree angle. And this is a mammogram showing the MLO view.

And generally on a good MLO view, we can see pectoralis muscle from the top of the image to about the center of the image along the nipple line. And we should be seeing a little bit of inferior tissue as in the inframammary fold region.

Additional Views for Localization

And there are a number of other additional views, sort of tricks that we use to help us localize and evaluate masses. And the ML view is the most common of those, and that is a view which is obtained from a 90 degree perspective.

Abnormalities on Mammogram

Now, there are a number of abnormalities that we may see on the mammogram, and those include masses, asymmetries, focal asymmetries, distortion, calcifications, adenopathy, and even lesions in the skin.

Example of Correlating a Mass

This is an example of a mass, which was found on a mammogram in year two. And we see mirror images of the MLO views year one on the right. Year two on the left, the arrow points to a mass.

Now a mass in BI-RADS lexicon is something that is seen on two views, and here we see on the CC view that that mass is in the outer aspect of the breast. So we would go to ultrasound and take our transducer and place it in the upper outer quadrant until we find a mass that matches the size of what we're seeing on mammography. And here we've found a simple cyst.

When we're going from mammography to sonography, we have to find the mass and we estimate the position of the mass based on the mammogram, and we can use the CC and the MLO view to do that. But it's also very helpful to have a 90 degree lateral view and to understand the distance of the mass from the nipple on the mammogram as well.

But we have to be careful because the estimated area on ultrasound may, when we see it on ultrasound, may not actually represent what we're finding on the mammogram. So we have to be very, very careful that the ultrasound finding matches the site of the mammographic finding because sometimes on ultrasound you'll find more than what you're seeing on the mammogram, but we have to make sure that we see the mammographic finding on ultrasound as well as anything added.

When we label our ultrasound images, we may do it by clock position or by quadrant, and either by measurement or by thirds, anterior third, middle third, posterior third.

This is a mass, which on the CC and MLO views is in the upper outer quadrant. We'd probably say that this is at the junction of the anterior and middle third of the breast.

Now, this was an ultrasound that was performed. Is it the correct mass? What location is it at on ultrasound? Well, you can see that it's labeled three o'clock, but we don't see a label noting the distance of the mass from the mammogram. We see that it is a fairly benign appearing mass, well circumscribed oval parallel to the chest wall, and it's surrounded by a fair amount of fibroglandular tissue. On our mammogram. There wasn't much fibroglandular tissue around the mass, but that wasn't noted and the patient went to biopsy. And what we're seeing is that the site of biopsy was distant from the mass and the mass is still there.

So when we go and take the patient back to ultrasound, we note that it's a fairly anterior lesion and the mass is surrounded by fat. And so we are finding the mass here at 3:30, 3 and a half centimeters from the nipple.

Localizing Lesions Using Routine Views

Now, how do you determine whether a mass is in the upper outer quadrant or somewhere else? Just based on our two routine views, the CC and the MLO view, here we have a schematic showing a lesion in the upper outer quadrant of the left breast, somewhere between 12 o'clock and three o'clock, We have the mass between 12 o'clock and three o'clock.

On our CC view, we compress anterior over posterior and on the mammogram, the lesion will be off to the side, off to the lateral side. On the MLO view, it'll appear to be in the upper breast, but if we go and do a 90 degree lateral view, while it's still will be upper, it may not appear quite as high in the breast as it did on the MLO view.

And here we have a lesion in the outer breast on the CC view. It's in the upper breast on the MLO view. And on the 90 degree lateral view, it is really just about at the nipple line. So you can see that if a lesion in the outer aspect of the breast moves down when you're going from the CC to MLO to ML view, we can do the same thing with a mass. In the lower outer quadrant here we have the mass on the CC view On the MLO view and on the 90 degree lateral view.

And here we have a schematic showing the mass on the CC view looking very similar to what you might find in the upper outer quadrant. On the MLO view, the mass is going to be lower in the breast. But this looks a little different than it would if it were in the upper outer quadrant. And on the 90 degree lateral view, it shows you how low in the breast it actually is.

And so again, you can see for a mass in the outer aspect of the breast, the lesions appear to decrease in their position from CC to MLO to ML. And here's an example of just that we see the mass in the outer breast, sort of mid breasts on the MLO view, and it's very deep, very low on the ML view. So we can remember that lateral lesions lie down.

Inner Breast Lesions

So what happens if a mass is in the inner aspect of the breast? Well, let's look at the lower inner quadrant. Here we have our schematic showing the compression in the CC, the MLO and the 90 degree lateral view. And this is what it's going to look like. Schematically. The lesion is going to look like it rises from the CC to the MLO to the ML view.

And here's an example of a mass, which is loosened. It's actually an oil cyst, and you can see it's in the far inner breast on the CC view, and it appears to move up from the MLO to the 90 degree lateral view.

And finally, the upper inner quadrant. Here's our schematic CC, MLO and 90 degree lateral views. And again, inner breast lesions are going to appear to rise up from the CC to the MLO to the 90 degree lateral view. And here we have an example of a mass that does just that. So we can remember for inner breast lesions, for medial breast lesions, medial moves up.

Central Breast Lesions

Okay? What happens if the lesion is in the center of the breast? Well, in that case, it doesn't tend to move much at all. On the MLO and 90 degree lateral views, it pretty much stays where it is. And here's an example. CC, MLO and ML right smack in the middle of the breast.

Example of a Suspicious Mass

Here's an example of a mass that we have found on mammography. It looks fairly well circumscribed or, or is it really? Do we see some spiculations coming from it? It needs further evaluation. We wanna take it to ultrasound. Where is it located?

Well, it's in the inner aspect of the breast on the CC view, and it looks slightly lower on the MLO view. Well, we don't have a 90 degree lateral view. Where are we going to look for it? We know that medial masses move up, so we're probably going to look for this in the mid to upper aspect of the breast. And here we see the mass. We've localized it at nine:30 at about the level of the areolar margin on ultrasound. So it's in the upper inner quadrant. And this, by the way, was an invasive ductal carcinoma, and those were spiculations that we saw on the mammogram.

Optimizing Ultrasound for Difficult Localizations

Now occasionally we do have trouble localizing the mass on ultrasound, even though we know where it should be, and we wanna optimize all our parameters to be able to see that lesion well. That includes our technical settings on the equipment and positioning the patient in a position where we can best see the mass.

Once in a while, we actually try to mimic the mammographic projection while we're scanning the patient. Remember to make sure that your presets are on your breast settings, watch gain settings, make sure the focal zone is in the region where you believe the mass to be. And you might want to try using some of the advanced ultrasound modes, compound imaging, harmonic imaging or panoramic imaging to view a large area of the breast.

Example: Focal Pain

This is an example of a woman who had focal pain. She could put her finger exactly where her breast pain was. Most of the time breast pain is bilateral or diffuse or over a wide area. But this was different. This patient could put one finger at the site of pain. The only thing we saw on the mammogram was perhaps a little double density here in the region of her pectoralis muscle.

When we went to ultrasound, because the tissue was so thin and up high in the breast, we put a gel pad on the skin expecting it to be a fairly superficial lesion, but we saw nothing. Then we removed the gel pad and set our focal zone for deeper tissue. And what we found was a irregular hypoechoic mass, which turned out to be within the pectoralis muscle itself. And this was a metastatic lesion from rectal carcinoma.

Example: Palpable Deep Mass

Here we have a patient with a small mass, very deep in the inferior aspect of the breast, and she can feel it. So at first we put the transducer down and we see really not much of anything, but we have not set our parameters appropriately. If we place a gel pad over the palpable lump, we can actually see the skin rise up in the region of palpable concern. And with our focal zones set close to the skin, we can see that the mass is just underneath the skin and in fact involving the skin. And this was simply a sebaceous cyst.

Patient Positioning for Ultrasound

Now you wanna optimize your patient positioning. Generally, they will be in a shallow lateral decubitus position with the ipsilateral arm raised up over the head, and this spreads the breast tissue out evenly over the chest wall and tauts it as much as possible.

But once in a while, you'll wanna put her in a semi erect or erect position. One reason is sometimes they can only feel a palpable lesion in the erect position. But the other thing is that the bulk of the breast will commonly drop down in the erect position, and that can help to improve tissue visualization in the more superior aspect of the breast.

But also you may wanna roll her over more either to the more contralateral side or to the ipsilateral side, depending whether you're looking at the deep lateral tissues or the medial tissues.

And we see this patient lying supine with her arm down at her side. If we were to scan this breast, we would find it very doughy and somewhat difficult to scan around. On the other hand, you raise that arm up and roll her over a little bit and it stretches the tissue out and makes it a much firmer sort of platform to perform your ultrasound on.

And this is just an example of how the distance between skin and a lesion might change if you move the patient from a supine position to an erect position.

Example: Difficult Deep Mass

And this is a rather old case showing a good size mass in the posterior aspect of this patient's breast. And initially we tried very hard to find the mass on ultrasound, but it was surrounded by a lot of fatty tissue and it was nearly isoechoic to the surrounding tissue, so it was very difficult to see. We sat her up and although it's still difficult, we were able to find the mass and then we could follow the mass back into a supine position and find it and able to perform a biopsy once we found it.

Mimicking Mammographic Projections

Now occasionally you might want to mimic the mammographic projection. This generally works best in the CC projection because you can hold the patient's breast in your hand. You can measure from the nipple how far back and how far lateral or medial the mass may be.

Another thing you can do if you have your ultrasound equipment handy to the mammography equipment is to place the breast in a fenestrated mammography grid and place and take an image, and then you can see where the mass is in the alphanumeric grid, and then place your transducer over it, such as, in this case, we saw a very small mass and a fatty breast, and we found the mass only by placing it in the grid in this way.

Verification Techniques for Correlation

Now, sometimes we find a mass on ultrasound. We wanna make sure that this is the lesion that we're seeing on mammography, or we wanna find it on the mammogram. One of the things we can do is place a marker on the skin and then go from ultrasound to mammography with this metallic marker.

Another thing we can do and actually works a little bit better than the marker, but it's a little more traumatic for the patient, is to place a skinny needle through the mass under ultrasound guidance and then perform a mammogram with the needle in place.

Also, if a mass is visible on ultrasound and we aspirate it under ultrasound, we can follow that aspiration with a mammogram to make certain that whatever mammographic finding prompted the ultrasound was removed by aspiration.

Example: Skin Marker

So here's an example of a patient with a very suspicious mass on ultrasound, and we've placed a metallic marker on the skin right over the mass and then taken her to mammography so that we could correlate the imaging findings. And here we see sort of a linear irregular focus on mammography just next to that metallic marker.

Example: Needle Localization

This is a mass that we found on Mammography. We did an ultrasound and we weren't quite sure if this heterogeneous mass represented the mass on mammography or not. So we placed a skinny needle in the breast and redid our mammogram. And sure enough, the mass that we see on ultrasound corresponds to the mass that we saw on mammography.

Here's another example of an ill-defined very small mass that we're seeing on mammography. On ultrasound, we see a little hypoechoic focus not sure if this represents the same mass or not. We place a very skinny needle in the breast and repeat our mammogram. And here we see that the needle goes right through the mass. So we're very confident that we have seen the mass in question.

Example: Asymmetry and Needle

And here's a case of an irregular asymmetry on mammography. We're not sure if this really represents a mass or not, or whether this is simply a little asymmetric breast tissue. We do an ultrasound and the only thing we can find is this hyperechoic nodule. Well, this is a very benign appearing finding on the mammogram. It probably doesn't correlate with what we're finding on the mammogram, but we have to make sure and we place a needle through that little hyperechoic mass and repeat the mammogram. And unfortunately, no, it does not correlate with this asymmetry that we're seeing in the inner aspect of the breast. So whatever we do, we have to do based on the mammogram, not on the ultrasound.

Example: Cyst Aspiration

Well, another thing we can do if we feel that what we're seeing on ultrasound is a cyst is aspirate the cyst and then head on back to mammography to make sure that the mass we're seeing on the mammogram is the mass that we're seeing on ultrasound. So step one, we found a mass on mammogram. Step two, we found what we think is a cyst on ultrasound. We're not sure if it's the same mass that we're finding on the mammogram. So we go and aspirate the cyst.

Now, a long time ago, people used to talk about doing pneumocystogram, where they place a little bit of air in the cyst as well. I don't really know anybody that does that, but if you wanna do it, it actually shows up nicely on the mammogram later. So then you repeat the mammogram after the cyst aspiration and see if the mass is still there.

And here's a case where we saw a mass deep in the breast and we found a cyst on ultrasound, aspirated it to completion, and then came back and repeated the mammogram and the mass is gone. So success.

Example: BB Marker

Now here is a mass in a fatty breast very small, and we're a little bit worried that on mammography we're not gonna be able to find this because a lot of times small masses in fatty breasts on mammography don't show up particularly well on ultrasound. The only thing we find is a little hypoechoic mass with what looks like an echogenic halo. We wanna make sure that this is indeed the lesion that we're seeing on the mammogram. So what we did was we placed a BB and redid the mammogram and it's sort of close to the mass on the mammogram, probably close enough.

So the lesion looks somewhat suspicious on ultrasound. So we did an ultrasound guided core biopsy and then placed a clip. And of course we did our post biopsy mammogram, and it shows that the clip is indeed in the same location as the mass was. And what we got outta that, by the way, was a fibroadenoma nodule a benign lesion.

Example: Bloody Nipple Discharge

Here's something a little bit different. This is a patient who presented with bloody nipple discharge and we did a ductogram and just put a very small amount of contrast material within the leaking duct and did a mammogram. And that showed that the contrast only went so far before it came up against an intraductal mass. We then took her to ultrasound and we can see that dilated duct just under the nipple, and we can see that there is material within the duct. And this represented papillomatosis at biopsy.

Correlating with MRI

Now a lot of times we're seeing enhancing lesions on MRI that will want to biopsy, but MRI is an expensive procedure. It is logistically a more difficult procedure to perform. Just getting time on the equipment sometimes can be a problem and it takes at least an hour to do a biopsy, whereas an ultrasound guided core biopsy takes much less time. So whenever we can do an ultrasound guided biopsy over an MRI guided biopsy, we're gonna try and do that.

Example: History of DCIS

This was a patient who had a history of DCIS in her right breast, and we found a sub centimeter enhancing mass in her left breast. This was her mammogram. She has heterogeneously dense tissue. We did not find a mass on the mammogram. And when we did her ultrasound, we found that at one o'clock in the upper outer quadrant of the left breast, six centimeters from the nipple. As we've documented here, we see a hypoechoic mass matching the size that we saw on the MRI.

And let's just go back to the MRI. We see that this mass is in the upper aspect of the breast. It's at about the junction of the middle and posterior thirds of the breast. And if you come to the axial view, we see that it is in the outer aspect of the breast. So if we take her to ultrasound, we know to look in the upper outer quadrant, fairly deep in the breast. And that's exactly where we found this mass. And we did a biopsy on her. We got a fibroadenoma as a result. And of course we did our post biopsy mammogram showing the location of the clip. And I have the prebi mammogram in the inset there. So you can see that it was very difficult to see that mass on mammography.

Example: Strong Family History

Now here's a patient with a strong family history of breast cancer. She has scattered fibroglandular densities in both breasts. Don't really see anything particularly worrisome in either breast, but we did an MRI as a screening procedure in this patient with her strong family history. And in the left breast we see a mass, which is just below the nipple line. On the sagittal view, it's at approximately the junction of the middle and posterior thirds of the breast. And it's just a little lateral to the nipple line on the axial view.

And on ultrasound, we go to the outer aspect of the breast, expecting it to be a little bit lower. And we see the mass at three o'clock, this is the right breast, and we're seeing a mass in the superior aspect of the breast, and it's in the outer breast on the axial image. And we go to the ultrasound and we see at 10 o'clock, eight centimeters from the nipple, a hypoechoic mass, which we biopsied. And it was a fibroadenoma.

Example: Ring Enhancing Mass

This is a patient who also had a family history of breast cancer, and we did a screening MRI and we found a ring enhancing mass in the breast. We see that it's at about the nipple line, perhaps a little bit higher, and it's fairly deep in the breast, but central on the axial STIR image. What we found on mammography in the central aspect of the breast at about 11 o'clock was a thick wall complex cyst. And this matched the ring enhancement of the MRI and we were able to biopsy of this under ultrasound and it simply was a benign complex cyst.

Example: History of LCIS

This was a patient that had LCIS diagnosed in the left breast. We did a screening of the right breast and we found an irregular enhancing mass in association with linearly array enhancement, we took her to ultrasound and we found a very ill-defined focus of hypoechogenicity at 10 o'clock four centimeters from the nipple in the right breast. And we were able to biopsy that and we got invasive lobular carcinoma.

Example: Distortion and Asymmetry

Now, here's a patient who on mammography had a questionable area of distortion and focal asymmetry in the outer aspect of the breast. No matter what they did on workup, they could not find this on the MLO view and scanning. The outer aspect of the breast on ultrasound showed no definite abnormality. However, on a mammogram, there was a very definite focal area of enhancement, which seemed to match the same site on the CC view of the mammogram.

Well, it turned out this patient could not undergo another MRI for reasons which I can't remember, but she could not go back and have an MRI guided core biopsy. So we had to do our best and find this lesion on ultrasound. And knowing the location of a mass on MRI, which was in the inferior aspect of the breast and lateral, we were able to find a very ill-defined hypoechoic area, which showed some shadowing. And we were able to biopsy this and got radial scar out of it or radial sclerosing lesion. And here's our post biopsy mammogram showing the site of that lesion with the clip in place.

Fat-Containing Masses

Now there are a number of masses that we see on mammography that contain fat. And when you see masses that contain fat, generally mammography is the way to go for diagnosis because ultrasound itself can be rather confusing. And examples of fat containing masses on ultrasound include lymph nodes, hematomas, or the fibroadenolipoma, lipomas and oil cysts. And just a few examples of those.

Here we see a fairly typical lymph node on mammography, and this is what it looks like on ultrasound, but this is a fibroadenolipoma. And on mammography it has a very characteristic appearance. It's a mixture of dense tissue and fatty tissue, and it has a partial capsule around it. If you look at this under ultrasound, you're just going to get very confused and worried about the lesion because it's a very heterogeneous appearance. We don't need to biopsy this based on its mammographic appearance. But on ultrasound, if you're worried about the ultrasound appearance, you're likely to want to recommend biopsy, but please don't.

Lipomas are also fairly easily seen on mammography. These are not something that we have to ultrasound. If you can recognize it on the mammogram and on ultrasound, they tend to be fairly homogeneous. But a little bit increased in echogenicity relative to surrounding fatty tissue.

And this is an example of an oil cyst on mammography, very lucent. Maybe as with round hollow calcifications, but on ultrasound it can look very complex and you do not need to put a needle in these or biopsy these if you can correlate them adequately with the mammogram.

Summary

So in summary, we want to determine the location of a mass based on the positioning seen on mammography or MRI, and then go to that location on ultrasound to find it. We need to verify ideally that the mass that we're finding on ultrasound is indeed the lesion in question on the mammogram or MRI. And do remember that there are a few lesions where mammography and sometimes MRI may be better than ultrasound for making a definitive diagnosis.