All right, is my screen sharing properly? Can't see anything. So I'll assume yes. We see it and hear you as well. We're ready for your presentation. All right, thank you very much. Thank you for inviting me. So we're going over the imaging of common shoulder pain. I have no disclosures, but I'm always happy to change that. So we'll address some medical knowledge and osteopathic patient care for the AOA competencies. Objectives specific to this presentation. By the end, I want you guys to be able to compare and contrast the advantages and disadvantages of the different imaging modalities. We should be able to appropriately select a specific imaging study for a given shoulder complaint and describe basic pathologic findings for the different imaging modalities. Doctor, do you see a black box on the screen that you're sharing? I do not. Okay, thank you. Should I change something? Whatever you did just fixed it. Oh. Thank you, continue on. Oh, I think it was the Zoom chat. All right, so where do we start? So you have a patient who's 50 years old, right hand dominant and right shoulder pain. What are the options? We can do an X-ray, we can do a CAT scan, MRI, ultrasound, none of the above. Let's decide. Okay, so first we're going to go through the X-rays. So advantages, it's fast, it's quick, it's easy to do, relatively inexpensive, you usually don't need to get prior authorization and it provides direct OSCIS visualization. It's a great screening tool. The cons are that it's a static imaging test, meaning you cannot assess mobility. With the X-ray, because it's primarily for the OSCIS structures, soft tissue is usually indirectly visualized. So we may not see the tendon directly, but we'll see associated findings of tendon injury. And of course there's radiation. So more of a concern, especially for the younger population. This is just to review what a normal X-ray looks like. So I have it labeled because I like things nice and easy. So you can see the clavicle, AC joint, here's the humerus, greater tuberosity here, lesser tuberosity, and you have the glenoid. We have other types of X-ray views. So this first one is just a regular AP view. Here we have the Gracie view, which is excellent for assessing the gleno-humeral space. We have the axillary view. We have the scapular view here. You can see the scapula looks like a Y, and this is called the striker view. So advantages and utilities of each. The AP X-ray is good for overall impression, the initial evaluation of the shoulder. The Gracie view, because we can view the gleno-humeral space quite well, it's excellent for assessing for osteoarthritis of that joint. The axillary view. It's great for assessing subluxations, dislocations, looking for Bankart lesions. The scapular view is primarily in trauma. So it's great because there's minimal shoulder movement needed to position the patient appropriately. And if in high velocity, high energy trauma, we could identify a scapular fracture. The striker view, if you're concerned about any posterosuperior humeral lesions, like a Hillsex deformity, this is what you'll look for. So in the normal patient encounter, you start off with the AP, unless you have concerns for trauma, dislocations, et cetera. So here we have an AP of the shoulder, and this is showing evidence of a rotator cuff tear. So this is a high-riding humerus. The humeral head is elevated, meaning that there is no tendon underneath the acromion to act as a cushion. You can also see that there's some remodeling of the acromion, that's called acetabularization. And then you can see some osteophyte formation as well. So there's rotator cuff arthropathy. So again, we don't see the tendon directly, but these findings are indicative of a chronic, complete, full thickness, supraspinatus tear. So this is the gray-she view. So this is looking at glenohumeral arthritis. You can see there's degenerative changes in the AC joint, and there's also narrowing of the glenohumeral joint. There's sclerosis, maybe some small cysts. There's also some enthesopathy, the greater tuberosity. So this is also showing evidence of rotator cuff tendinopathy. So arthritis of both the AC glenohumeral joint and rotator cuff tendinopathy. Probably less common, but here it's just showing that you can see fractures. So if there's trauma, you can potentially have some traction injury from the supraspinatus on its insertion to the greater tuberosity. So here we're seeing a fracture at the greater tuberosity, but we also see a humeral neck fracture here. You can see on both sides. And we have the comparison on the left. So it's a little bit easier to see. Anterior dislocations are much more common than posterior dislocations. So we have the normal imaging on the left here for direct comparison. Over here, this is an AP showing that the humeral head is now subcoracoid. So here's the coracoid process. You can compare it on the normal where you see the humeral head lateral to the glenoid, but here it's basically directly under the coracoid. This is the scapular view, and it's showing that the humeral head is now anterior to the fossa. So I'm going to outline the humeral head over here. And if you're having a hard time seeing it here, I put the red line. So now you can see it's quite anterior. Okay, so for CAT scans, advantages, very fast. Excellent evaluation of the bones, especially the cortices. If there's calcific tendinopathy, you can assess that very well. If an MRI is contraindicated due to claustrophobia, a medical implant, or obesity, CAT scan is definitely an option. Disadvantages, similarly to the x-ray, it's a static imaging modality. It's expensive. Because it's basically like a super x-ray, it's not the best tool to evaluate soft tissue. So generally we need contrast for that type of evaluation. Even with contrast, it may not fully evaluate the tendon. So the contrast would be injected into the joint. And if there is a lesion that is on the bursal side of the tendon, so we have the bursal side and the articular side. If the articular side is intact, but there is a bursal lesion, the contrast has no path to get to the bursal side. So the tendon may appear normal. Additionally, if you add contrast, the contrast itself may actually obscure the underlying osteostructures. So you would gain soft tissue visualization at the cost of some bony assessment. It's not the best to evaluate bone marrow edema. MRI will be much better for that. And of course, being a super x-ray, there is super radiation. So this is comparing a CAT scan with an MRI. So just first thing, so the images on the left here, these top two are the CAT scans. This is the MRI. And then on the right, it's a different patient, top two CAT scan, bottom two MRI. So one thing to note here, CAT scan, here you can see some screws from prior surgery within the clavicle. And you can see them well, and you can see the other structures. But in the MRI, there's a lot of artifact. The other thing we see here, so here the arrow's pointing to the supraspinatus tendon. And it looks good. It looks intact. But if we're looking at the MRI, you can see that there's some hyperintensity in the tendon, which is indicative of a tear. So this is a false negative for rotator cuff tear in this CAT scan. And it's because the pathologies on the upper superior surface of the tendon, the bursal side of the tendon. On the right, we have a complete rotator cuff tear. So with the contrast, now the contrast can go through the tendon. There's a communication from the joint to the other side, to the bursal side. So because of that, we can directly visualize both the tear in the tendon and the communication into the bursa. We can see that here as well. And the MRI is confirmatory. So for MRIs, really great tool for soft tissue visualization. You can visualize bone marrow edema, which is indicative of, usually indicative of some underlying process, whether infection, inflammation, cancer, et cetera. It's the best tool to assess the labrum. So if you're concerned at all about labral pathology, you have to get the MRI with contrast. And there's no radiation. Downsides, it's a slow test. It's also really loud. I don't know how many of you all have had MRIs, but they give you headphones, play music, because it is a loud test. If you have anxiety, you may need some anxiolytics. It's a static test as well. So it may take 30 minutes or more, and you have to stay completely still. Movement may provide or create artifacts. It's an expensive test, relatively, and insurances like to deny them, even if you provide all the necessary information, like UnitedHealthcare and Oscar Health. Oscar does not call me for the scheduled peer-to-peers. I don't know what everyone else's experience is, but I don't like them. And MRIs might be contraindicated in claustrophobia, medical implants, or obesity. So CT scans would be a good alternative there. In addition to the soft tissue visualization of the tendon tears, you can also assess indirect signs of tendon pathology, such as muscle atrophy, fatty infiltration, and you can assess the ligamentous structures as well. So T1 versus T2, here's a good mnemonic. In Terminator 1, Arnold is bad, so water is dark. In T2, Arnold is good. Water is light. In T1, that's called the anatomy. So it's good for visualizing the normal structures. You can also assess the bone marrow. It's fairly easy to assess fatty atrophy of muscles here. In T2, that's assessing for pathology. So it's great if you're looking for any fluid or edema, such as something that would appear in a tendon tear. Proton density images are kind of in between T1 and T2. They have a higher signal to noise ratio and what that does is it allows you to visualize signal poor structures. So for the meniscus, even though we're talking about the shoulder, for something like the meniscus which doesn't really show up so well on other MRI views, the proton density weighted imaging is great for visualizing meniscal pathology. So here's a T1 of the shoulder. So you can see everything looks really pretty, urinal head, glenoid. Here you can see the supraspinatus. It's a little bit hard to see the tendon, it's black, but we would need the T2 to look for pathology. So here we're looking at the three types of sequences, T1, T2, and the proton density. So here's the supraspinatus. We see it attaching onto the greater tuberosity. So if we're looking at the T1 first, you can barely see anything wrong. On the T2 you can see this abnormal signal, and in proton density you see it much more clearly. Excuse me. For labral pathology, the MRI is king, again with contrast. So these three are different types of labral tears. In the first one, we have some fraying on the undersurface. On the second one, it's more of a linear tear, and it's extending into the superior aspect of the labrum. And on the third one, it's more a displacement of the labrum. So we have the contrast going in between the labrum and the glenoid, and you can see that the labrum is being displaced. Okay, so for ultrasound, it's fast. Well, it's relatively fast. If you have the experience, then it's fast. If you're still learning, it can take a long time, but eventually it'll be fast. It's inexpensive. So yes, the machines cost anywhere from a pocket ultrasound being like $1,500 to a top-of-the-line one being more than $100,000, but still relatively inexpensive, and you can use it with almost every patient, and eventually you'll buy it back. It's a dynamic imaging tool. So it's the first one that we've gone over that's dynamic. So it's great if you're trying to assess for an impingement, if you're trying to assess mobility, so you can evaluate, for example, how well the supraspinatus can glide with shoulder abduction under the acromion. You can check for the different bursas or bursae to see if they're getting entrapped. You can see if there's any nerve entrapment, particularly with certain movements. You can look for snapping tendons, etc. It gives very good soft tissue visualization. I would argue better detail than an MRI. Clinical correlation. So sometimes you'll encounter a patient, you'll have your differential, or sometimes you may not have any idea what exactly is going on. But one of my best physical exam tools is just asking the patient to point where it hurts the most, and sometimes they'll point to a random part of their body, and just thinking anatomically, I may not know what's in that area, there's no tendinous attachment there, so I'll put the probe directly on that area, and lo and behold, I will see something. It might be a branch of a nerve, it might be a cyst, it might be some other, like a lipoma, but if even if you don't know what exactly it is, you can put the probe, and if you know what different structures look like, so you know what a nerve looks like, you know what a muscle looks like, you know what a tendon looks like, etc., you can identify it and look it up after and come up with a diagnosis. And there's no radiation, so if you have a pregnant patient with pain, and they don't want to harm the baby, it's easy enough to look at the soft tissue with an ultrasound. You can do it for the SI joint, if you have a lady with SI joint. The downside is there's limited osseous evaluation, so the ultrasound beams don't penetrate the bone. They're good to evaluate the cortex, so you can assess for osteophytes, you can assess for fractures, but anything deeper than the cortex, you're better off with other imaging. The intra-articular evaluation is also limited, so yes, you can assess for the osteophytes at the periphery of a joint. You can assess the periphery of the labrum, but deeper, you'll need other tools. And it's operator dependent, so it's very easy to lose or have a poor quality image due to handling the probe or mishandling the probe. So if the image is not as clear or clean as possible, then you may miss pathology. Additionally, if you do not scan the entire width or length of the tendon, you may miss pathology. So you can think of the ultrasound beam as being the thickness of a credit card. So tendons are going to be thicker than credit cards, but if you stick a credit card on your shoulder, you can see just theoretically how much you may miss if you're not moving that around. So it's not necessarily intuitive because the probes are kind of bulky. Okay, so we'll spend a little bit more time on the ultrasound just because we're at PM&R. This is what our main diagnostic imaging tool will likely be. So I start off with the biceps assessment, have the patient sitting. Some people have them supine. I learned it sitting. I sit behind the patient. I would rest my hand or wrist on the patient's shoulder. So in this neutral position, you start off with the probe directly in between the two tuberosities. So this is a drawing or computer-generated imaging of the anatomy with the probe. Here's an MRI and here is the ultrasound. So we should see the two tuberosities and in between the two tuberosities you'll see a slice of the biceps tendon. And you can just look at these two images here. Here the tendon looks dark. Here it looks bright. That's just because of anisotropy. So you do have to do some tilting to get rid of that. And for those who don't know, anisotropy basically means that depending on the angle of the ultrasound beams reflecting off structure, a tissue may look brighter or darker. So the image on the ultrasound is generated by receiving the beams back. So if you're not perpendicular to a structure, some beams will bounce away from the probe rather than returning back. And you won't get as clear of an image. This bottom picture here is looking at the bicep tendon in long axis. So you can think of it as a stick of pepperoni. If you're looking at in short, you're only looking at, you know, a slice of it, like a circular slice. And then if you're looking long axis, you can see the whole length of the pepperoni stick. So after you assess the biceps tendon, you don't even have to move the probe, okay? Just have the patient externally rotate. So if we're going to go back to the previous slide, here we have the biceps tendon in between the two tuberosities, and we have the subscapularis tendon coming here. So by externally rotating, we're shifting everything left, and that will bring the subscapularis tendon towards the middle. So here's the subscapularis tendon in long axis. Here it is in short axis. Here you can see a nice fibrillar pattern of the tendon. So you see the straight lines going across. And in short axis, the fibers are coming at us. So rather than straight lines, they may look more like dots, and these are just the different bundles of the tendon. So for the AC joint, so that's my third step in the protocol. Again, I start out with the biceps and with the patient in neutral. I externally rotate, with the patient in neutral. I externally rotate, and then I just move superiorly to the AC joint. So you can see the acromion and the clavicle. There's a disc, which is kind of like a meniscus. You have a ligament going across. You can actually see into the joint quite decently, so you can definitely assess for AC joint arthritis. Additionally, if there's trauma such as someone falling and landing on their shoulder, you may, and if you're concerned about a separated shoulder, not a dislocated shoulder, but a separated shoulder, you can assess for widening of the AC joint. After I assess the AC joint, I just move the probe posteriorly, and now it's the glenohumeral joint. So here you can see that there's still good detail. You can see the glenoid. You can see the humeral head. You can assess for irregularities of the humeral head that would indicate arthritis. You can see part of the posterior labrum here, and you can see the joint capsule. So it's not bad. It's not bad. I save the supraspinous for last because that's usually the most painful part of the shoulder, and ideal positioning is also painful. So what we do is we have the patient go into the crass position, which is the same position you would use for the subscapularis liftoff test. You just ask them to put their hand behind their back. If that is too painful, you can do the modified crass position, which is depicted in this figure. You ask the patient to sit on their hand, and if they sit on their hand, it's still going to bring the supraspinatus anterior. So if you're having trouble identifying where to put the probe, it's the same as the biceps. It's the same as the subscapularis. It's anterior. The difference is you have to rotate it 90 degrees to get the long axis view because here you can see the supraspinatus coming under the acromion and then diving down because again we are now in internal rotation. If we were in neutral, then you would be more lateral, but we're bringing the tendon out from underneath the acromion. And here you can see for yourself, this is ultrasound. These are MRIs. You can see the tendon much more nicely in an ultrasound than you can in an MRI, although the MRI does give a nice global picture. So we have the top layer here being the skin, then we have the deltoid, and then here's the bursa, so this black space here. Then we have the tendon. The black is the cartilage, and then the bright white is going to be the bone. So this is looking at how impingement or bursitis would look like on different imaging modalities. So on the shoulder, again, you would not see it directly, so you look for potential causes or associated findings. So there's just three types of the acromions, you have the flat acromion, a curved acromion, and a hooked acromion, which could predispose someone to shoulder impingement or bursitis. Here on the ultrasound, you can see that there is more fluid within the bursa, so more black fluid is black. We also see that there's a diminished fibular pattern, so we don't see those straight lines that we saw previously in a normal tendon, so this is called tendinosis, it's basically worn down or degenerative tendon. Here on the MRI, remember water is white, so this white here is the fluid within the bursa. Again, comparing three imaging tools for supraspinatus tears, here's the high-riding humerus, and you see the acetabularization of the acromion, the remodeling of the inferior surface. Here there's missing tendon, so we see black, right, the tendon should look gray. So this is actually compressible, if I were to push down on the probe, then it would compress this area, and that would indicate a tear. This is another view of the tendon, and we would call this a full thickness tear, both of these. So to describe a tear, full thickness means it goes all the way from bursa to articular surface. Incomplete versus complete would mean if it spans the entire width of the tendon. So this one is a complete tear versus this one is incomplete, but they're both full thickness. And here is an MRI showing you what a tear would look like, so you see white where there should be black or gray. So what if you can't see a tear? It's not always easy, so what we look for are indirect and secondary signs. So here we have the double cortex sign, so cartilage should look black. If there is missing tissue, then there are more ultrasound beams that are penetrating and hitting the cartilage and bouncing back, so the cartilage is actually visualized, but again you have to have missing tissue in order for that to happen. Here is the sagging peri-bursal fat sign, so the shape of the tendon should be contiguous or constant, but here you see there's a little dip, so that happens if there's missing tendon. Compressibility we discussed, muscle atrophy, you can assess the supraspinatus muscle or whatever muscle that you think is torn, look for fatty infiltration or the size being diminished. Other things to look for, if you see irregularity of the cortex, that indicates likely tendon pathology, if you see an effusion, especially around the biceps tendon, immediately you should think supraspinatus tear. So tendinosis, here we lost the fibular pattern, so you see it looks kind of blurry, this can easily be mistaken for a tear, but if you look more closely you can see that there is still tendon inside, it's dark but it's there, if it were completely black then it would be a tear, but it's just severe tendinosis. Here we see cortical irregularity, right, so the tendon itself looks thinner than the other tendons we looked at, so you can assume that there's missing tendon, there's irregularity of the cortex. Here this one is actually pure black, so there's a small tear there, and I'm more confident in calling that because we see a cortical interface sign here, or double cortex. So this cartilage interface sign is corresponding with this finding, so it's confirmatory, and apparently the cartilage interface sign, or double cortex, or whatever you call it, has 100% sensitivity and specificity. So I want you guys to look at the insertion of the tendon here. So this is actually the subscapularis, so you can see the biceps tendon here, this is the subscapularis coracoid process is going to be here. So if we're looking at the attachment, here it looks nice and filled in, here it looks black, so you could easily mistaken this for a tear here, but when you reduce the anisotropy, it fills in and you can see that there's actually tendon in that space. Another way to eliminate anisotropy is with a heel-toe maneuver, so we're looking at this area here, and all I'm doing is tilting or heel-toeing the probe, and that's changing the angle, it's making the fibers more parallel. So you can see the fibers here are more parallel, at the start they're curved down, that's how you eliminate anisotropy. It's very important to scan the length of the tendon, so here we're looking at the biceps, so early on biceps looks pretty good here, right, maybe a little bit of an effusion, but if we scan the entire length, here we can see there's actually a split tear of the biceps and there's more effusion, so if you're only looking in one spot, you can miss it. So here's a patient with acute onset right shoulder pain, tendon looks pretty good. Here this is an isotropy, so you can, it's not a tear, you can see some of the fibers going, maybe a little bit of bursitis, okay, so same patient, and I'm moving the probe maybe half a centimeter, maybe, not even, let's say two millimeters. Same tendon, same patient, now we have this lesion, this is a calcification, and we're pretty zoomed in, so this is a two by four centimeter calcification, and this is why I said ultrasound is operator dependent, if you just take one picture, one quick look at the supraspinatus, this one looks perfectly fine, you move the probe just a tiny bit and you see this big calcification. Here's another patient, first thing I looked at was the biceps tendon, that's my first step, and immediately I saw an effusion, so yes I'm going to diagnose the patient with biceps tendon synovitis, but I'm putting that to the side, I'm thinking all right I definitely have to look for a tear. So when I go to the supraspinatus, this one doesn't look so bad actually, there's no gross tear visualized, right, there's no sagging bursa, the bursa itself, a little bit of fluid but not too bad, but what we see here is that cortical interface sign, that's the cartilage, um so when you see that, you have to look harder for a tear, and again it's 100% specific and sensitive apparently, so you could argue that this could potentially be a tear here, and if you're not sure, sometimes you have to do a diagnostic injection, not for pain, but to assess the tendon. So here I'm injecting something probably dextrose into the tendon, and you can see it's filling in that gap, and when I stop injecting, it collapses. So there's actually a tear here that is probably difficult to visualize without the injection, but if you see this, you can basically assume there's a tear. So here's an algorithm for how to decide what imaging to get, so basically if there's trauma and they're a little bit older, you get an x-ray, based on that you can go to ultrasound, or the x-ray itself may give you the diagnosis. So you can see here, if there's a high-riding humeral head, you already know there's rotator cuff disease, you may not need other imaging, you can see directly a calcification, you can see arthritis. If it's negative, you can use an ultrasound to further assess the integrity of the tendons. If you're concerned about labral pathology, then you should get an MRI, or actually if you have a high suspicion for the labrum, just skip the regular MRI and go for the MR with arthrogram. This is my algorithm, you don't have to use it, but if there's any trauma or red flags, I get an x-ray. If not, then I kind of have to subjectively decide, will therapy be enough for this patient, or do they need some elevated care? If it's enough, then I don't need to get an x-ray or any imaging, because I'm just going to send them to therapy no matter what it shows. If I don't think they can do therapy, or I think they need something more advanced, then I have to decide, is it something surgical or non-surgical? If it's surgical, I get an MRI, because the surgeons don't really interpret ultrasounds, and if they're non-surgical, if I'm going to do injections, then I'm just going to do an ultrasound. I put an asterisk here, because the insurance companies may require an x-ray beforehand for authorization, even though it won't necessarily help me. So some pearls when you're doing, learning ultrasound, I like to tell my residents to practice with the biceps tendon, look at it in short axis, put it in the middle, practice rotating it 90 degrees, get it in long axis, heel toe to get rid of the anisotropy, uh because the the biceps tendon is relatively thin, so it's a good good one to start off with to develop your skills. And for performing injections, the same skills you use to visualize the biceps tendon are the same skills you use to visualize a needle. Do one movement at a time, so you can slide, sweep, tilt, heel toe, whatever you want, just pick one thing, one type of movement for the probe, and if you do that, there's only two possibilities. You're either doing it right, or you're doing it wrong. So if the picture looks it right, or you're doing it wrong. So if the picture looks better, then keep going. If it starts to look worse, stop. If you if you do three dimensions at the same time, you're you're messing yourself up. Scan the entire width or length of the tendon, I showed you some examples of how it's easy to miss pathology. Try to find known pathology. So if you have an x-ray or MRI, look at those pictures, read the report, look for specific finding, and then try to find it with the ultrasound. So that way you can learn how to orient yourself, and you can see what that type of pathology would look like on ultrasound. If you see an indirect or secondary sign of a tear, but you don't see the tear, look harder, because it's probably there. Thank you very much, and then I don't know if we're going to answer questions now, or just start with the other two lectures. There was one question, and I actually think it was more for Dr. Sherman. Let me just pull it up here, but both of either of you could answer. Do you find the combination of multiple tests increases the specificity for diagnosis? Yeah, so I think that you don't have to do all 20 or however many tests there are. It's important to pick ones that you're comfortable with, that you're familiar with, and for me, it's also important that the tests reproduce the symptoms, and that they're also important that the tests reproduce the symptoms, and that they're also reproducing the symptoms that you're concerned about. So, for example, if you do a NEARS, and it hurts the other shoulder, it's not a positive NEARS, because it's doing something completely different, but if you have a positive NEARS, and a positive Palkins, positive MTCAN, and it's all reproducing the shoulder pain, then you can be more confident in your diagnosis.

imaging modalities

shoulder pain

X-rays

CAT scans

MRIs

ultrasounds

rotator cuff tears

diagnostic tests