All right, so the next 15 minutes will be a whirlwind view of just a couple of spine cases and going through some differentials. I have no disclosures. So the first, well, one thing about spine history is not particularly useful most of the time, so neck pain and weakness, you'll see that's a variation, a 41 year old. So again, my topic is gonna be spine tumors. So take a look at this. T1 on the left, T2 in the middle, T1 post-contrast on the right. Digest that a little bit, what you think. Okay, there's the axial view at the level of the lesion and then right at the inferior margin of it. Okay. So first of all, what I list on the questions here has really little to do with the actual differential that you might put forth for intramedullary tumors. So these, you gotta have some discriminators here. So do you think that's an ependymoma, a hemangioblastoma, an AV fistula, a meningioma, or radiation myelitis? So you just think about that. I can feel the vibes coming to me here. So the answer to that one is ependymoma, right? So spine is easy. You only need to know three spaces, right? You need to be able to place it extradural, intradural, extramedullary, and intramedullary. And then you can narrow your differential and maybe if you know three things for each of those spaces, so maybe you need to know nine things and you can get away with it pretty well. So if you were actually sitting there and had an intramedullary lesion, your differential would be ependymoma, astrocytoma, hemangioblastoma, and metastatic disease because you've covered 95% of the lesions in that space. This is a test question, so it's not going, it's only partially correct. So ependymoma is the first thing and then we'll look at these other ones that are the distractors for this question. So ependymomas, kind of two varieties in the spine. You've got myxopapillary ependymomas. This is not a myxopapillary ependymoma. That's the one that's located along the phylum terminale. Myxopapillary ependymoma is recently upgraded by the WHO. It was always a grade one, now it's a grade two, but we're not talking about that. This is a classic ependymoma of the spinal cord. These are either two grade two or three. Ependymomas, we lag behind kind of the genomic features and prognosticating features that you see, say for brain tumors, particularly like glioblastoma. That really doesn't quite exist for ependymoma, still a fairly enigmatic tumor. There are a lot of different histologies of this, but they really don't have much in the way of prognosis. So we have a lesion and we're not certain how the patients are going to do, except for this more recently one type, which is the MYCN amplified. So this genomic variant is thought to be poor prognosis. This is a lesion that is treated by gross total resection. So find a good surgeon, that's the best way to approach it. These tend to be soft and friable lesions. You can have polar cysts, cranial and caudal cysts associated with the lesion. They can have chronic hemorrhage and hemocytin deposition. So they can be a fairly complex signal intensity lesion. Here's the obligatory schematic for an intermediary tumor for an ependymoma showing the lesion with a cyst along the cranial aspect. And the histo slide on the right, just to point out that these lesions can be fairly big and look fairly ominous, but they actually have a pretty well-defined margin, unlike say a more infiltrative astrocytoma. So even if the lesion looks quite large, shouldn't preclude the surgeon going in and trying to get a gross total resection. Okay, here's another ependymoma. Again, big lesion. You see the nidus kind of intermediate to low signal on T2. You've got cysts both above and below the lesion. And you see kind of the syrinx pattern of these housetral markings down below. And then following contrast, you see fairly robust enhancement of the tumor itself. And hopefully you'll be bored by these when you'll see so many. Here's another ependymoma, another lesion with a cyst, cranial and caudal. Interesting on the T2 to separate out the enhancing component of the tumor, the bright kind of more bulk water cyst that is associated with it above and below. And then you have this flame-shaped focus above and below extending into the cord, which is the cervical cord or the cord equivalent of vasogenic edema with sparing of the periphery. And then you have the enhancing component in this ependymoma. And guess what? Another ependymoma. So we have a mass with very heterogeneous signal intensity. We have areas of kind of methemoglobin with a little bit of high signal and a lot of hemocytin deposition along the margins and some faint enhancement within the lesion. And then the T2 shows all this hemocytin staining of these small hemorrhages that have occurred related to the tumor as it tracks within the substance of the cord. So this really isn't necessarily tumor. That is just old blood byproducts getting on the highway and then tracking up along the spinal cord. Mangioblastoma was in the differential. So these can be isolated lesions or they're certainly the association with VHL, von Hippel-Lindau, which you all know the relationships better than I do, but they can be both cystic and solid. You're gonna look for that robust enhancing nidus. These lesions can be called out because a fairly small lesion can have a large amount of edema within the cord. Hemorrhage is very uncommon, even though it has hemangio in the name, they don't hemorrhage. And these are WHO grade one. So focal lesion with enhancement, large syrinx associated with the lesion, bulk water now. So such bulk water that you see the dephasing of the pulsatile CSF within the syrinx associated with the lesion. And then down below you merge some of this as bulk water, but then you have that edema within the cord, that flame shape inferior margin. Vascular malformations, that's an easy one, didn't look like that. I didn't see flow voids on it, so no one's gonna pick that. If you say vascular malformations in the spine by default, if you just pick dural spinal fistula, you'll get most of them. So just as a complete, here's what a dural fistula looks like, usually mid and lower thoracic cord. You see the edema pattern, again sparing of the periphery and you see the punctate flow voids calling out that the abnormality might be a vascular malformation. And I also had meningioma there just to double check that everyone knows that we're dealing with the images of an intramedullary lesion and this is an intradural extramedullary lesion and so should not be the correct answer. So you see the broad dural base margin compressing the cord and just like the old myelography days, come into the lesion, you see expansion of the CSF space as you come into the lesion, a hallmark of an intradural extramedullary lesion and you see the defacing of the CSF related to that. Radiation myelitis was part of that as well. This is hard and very nonspecific but it mainly relates to prior history. I didn't give you a history of cancer or anything like that with radiation therapy but here you see the pathologic fracture prior radiation, the irregular enhancement and edema within the cord. All right, number two, 20-year-old with back pain. All right, so it's abnormal, I'll give you that. You know, 90% of what you look at, you have to figure out, is it normal or abnormal? I gave you the 90%, it's abnormal. So we have a mass, we have a spiculated or calcific matrix within the lesion, a very kind of odd pattern of scalloping and destruction over one, two, three, maybe four vertebral bodies with some sclerosis. Here's the MR, parasagittal views post contrast is on an odd enhancement pattern, bright on T2 with the septations or spiculated pattern. So there's the axial, very large lesion, posterior mediastinum creeping in the neural foramen and then the cord is displaced off to the left. So what do you think? Is it a giant cell tumor? Is it a malignant peripheral nerve sheath tumor? Is it lymphoma or is it chordoma? Let's just take hands here in the room. Who thinks it's a giant cell tumor? Malignant peripheral nerve sheath tumor? Okay, thank you. Lymphoma? Okay, last one, chordoma? Okay, strong response, okay, excellent. Just checking, yes. So chordoma is the correct one, mainly based on the T2 signal intensity, very bright on T2, the spiculated pattern. And again, the questions are not the differential. The differential diagnosis in real life is, I see this, it's going to be chordoma or chondrosarcoma. Those are gonna be the main two that it can both be bright on T2. The rest are going to be distractors, but we'll go through them as well. Chordoma, again, two to 4% of primary bone tumors, most often sacrococcygeal, rises from notochordal remnants. And surgery, again, is the best option. They don't respond very well to conventional XRT, but certainly proton is being used on these lesions at our institution. This is a different patient with a thoracic chordoma. Another young male, for whatever reason, you see that scalloping of the vertebral body, giant lesion, very bright and weird looking on T2. This kind of puddles of enhancement post-contrast. That was a chordoma. This is a chordoma. So you see a lesion that looks like a MET, but it's very bright on T2, as this one is, think chordoma. How about this one? Another lesion, think it's a MET, very bright on T2, kind of a funny enhancement pattern, think chordoma. How about this one? Hope we're bored now seeing these. Very bright on T2, it's a chordoma of the sacrum. And then this outlier, which I think is an important one to recognize, this is also a chordoma. This is in a child, and it almost looks like CSF. It almost looks like some type of lymphatic malformation or weird arachnoid cyst, because there's really not much in the way of enhancement along the margins, but this is a solid mass. It just doesn't enhance very much. It's been there for quite a while. You see the expansion of the neural foramen, and this is another chordoma. So cervical chordomas in the paravertebral region, another area. And then another one, bright on T2, you got the message. Giant cell tumors, sacrum's most common site. Again, these tend to be low in T2 signal. Look for those secondary aneurysmal bone cysts with blood fluid levels or fluid fluid levels. Thing about giant cells, you do have a chemotherapeutic agent you can use now that can be very effective in treating these non-surgically. So here's the axial view of that lesion, kind of intermediate T2 signal intensity, not super bright, not speculated, and homogeneous enhancement. Malignant peripheral nerve sheath tumor or malignant nerve sheath tumors, that's going to be based on history. You had a lesion like a schwannoma, something changed. There's new pain, new growth, or you have a patient with new symptoms, particularly who has NF1. And then you're going to be thinking about that. You can stratify those based on the FDG PET uptake. So here's a malignant peripheral nerve sheath tumor, very high uptake on PET. You know, the lesion that I showed has calcification in it. You know, a lot of these differentials that I showed on the question, it's not going to have calcium in the lesion, doesn't have a matrix. Malignant peripheral nerve sheath tumors don't have a matrix, calcific matrix. So that's not going to be part of the problem. And then lymphoma. That's a go-to diagnosis for nearly everything, except this case, because that doesn't calcify. And it's low in signal on T2. So here we see lymphoma, trans-spatial, vertebral body, paravertebral, within the epidural space, pushing everything over, and then extending into the posterior elements, and then this broad area of enhancement. Now I have 30 seconds, maybe I'll take a minute 30, give you this last one, chronic leg pain, just to point out another very nonspecific pattern. Lumpy, bumpy enhancement within the fecal sac, long segment, there's the axial. Which one is least likely? Carcinomatosis, fungal infection, TB, Guillain-Barre, or sarcoidosis? Right, which one is the least likely? Right, so, well, the answer's going to be Guillain-Barre, just because the differential in reality here is do I see lumpy, bumpy, or do I see smooth enhancement along the cataquina? If it's lumpy, bumpy, I've got a big, long differential of everything listed there, from tumor to chronic infection, fungal infection, TB, sarcoid. If it's smooth and linear, I'm in another differential for viral infection, or Guillain-Barre. So Guillain-Barre, you know about, that smooth enhancement of the cataquina can be both ventral and dorsal. Here's a post-vaccination Guillain-Barre that we saw recently, again, all the roots enhancing. And then finally, I'll just stop with this and point out that we've had another influx in Arizona of West Nile, which you see the cord involvement in this patient with West Nile, but also a very striking ventral root enhancement. And, you know, things are bad enough with the pandemic, but recently, I would say within the last three weeks, we've actually had more patients in the ICU with West Nile than we did with COVID, which is kind of crazy. So my time's up. I'll leave you with that. Thank you very much for your attention. All right, so as I mentioned, we're gonna talk about spondyloarthropathies. And I don't want to really go over the, you know, classic, this is all the findings that you see on spondyloarthropathies. The objective of this talk is really to emphasize the images so that you at least put spondyloarthropathies in your differential when you see a case on your PACS of a patient that comes in with neck or back pain. And then think about some of the mimics of spondyloarthropathy, some of the other entities that may have similar imaging findings. So if we think of spondyloarthropathies, inflammatory processes that involve this axial spine, we can think of two big categories. We've got rheumatoid arthritis, and when we think of spine involvement, it's predominantly cervical. And then we have the seronegative spondyloarthropathies, and those are the five major types. There are ankylosing spondylitis, which is our main prototype, and then reactive arthritis, psoriatic arthritis, enteropathic or inflammatory bowel disease-related spondyloarthropathy, and the undifferentiated, which, truth be told, I almost always kind of lump everything into that and let everybody else figure it out. But this is our first case. So we see radiograph, and it's really hard to tell what's going on. There is some suggestion that there's widening of the atlantidental interval. The MR on the same patient shows this soft tissue eroding the tip of the dens. And this is an example of rheumatoid arthritis. As I said, this bundle of arthropathy involves the cervical spine. And the key feature with rheumatoid arthritis, just as a reminder, is the erosive synovitis. When you come across a study of a patient with neck pain, you may not get that classic pencil erosive dens that we're all taught. It's just because a lot of the immunomodulatory drugs nowadays don't give us that classic appearance. But we will see the widening of the atlantodental interval because of the ligamentous subluxation. Sometimes clues and other imaging on that same patient, such as the temporal joint prosthesis, because RA involves the temporal joints, can be helpful. And we really want to look for that inflammatory involvement of synovial joints so we can see enhancement of the atlantodental articulation here. There's fluid. And there may be widening of the atlantodental interval, which we see in the supine CT. Flexion extension radiographs really exemplify this laxity. We sometimes do balanced steady state free procession at our institution, which will show widening of that atlantodental interval and can even show narrowing of the spinal subarachnoid space. There's one very rare but important complication of rheumatoid arthritis that you should be aware of. Like I said, with immunomodulatory medications, we don't see it as often. But when we get this erosive synovitis at the C1-2 articulation, the calvarium can actually sink upon the cervical spine so that you'll get cranial settling. And this can cause the dens to rise up into the cranial vault and cause impingement of the brainstem, as you can see in this case. All the synovial joints can be involved. So this process can involve the oncovertebral joints and you'll get this coronal deformity, which is unfortunate and functionally debilitating. Here is a complementary case, but not the same. It's a mimic. So this patient has fever, neck pain. You can see extensive soft tissue at the cranial-cervical junction around the dens. And what's very helpful in this case is there are calcifications associated with the soft tissue. This is an example of crown-dens syndrome. So we have acute onset of pain, elevated inflammatory markers due to deposition of the calcium pyrophosphate dehydrate crystals within the retrodental soft tissue. The actual deposition is quite common. It's something like 12.5% in the population. And this increases, this deposition increases with age and you'll see it quite commonly. So it only becomes symptomatic when the thought process is when the crystals get into the joint space and create an inflammatory reaction. But patients can present with acute pain and elevated inflammatory markers. Here is another example of CPPD deposition. And it was very, very bizarre. And I wanted to share this case with you because we were scratching our heads. So this patient also had CPPD and pretty extensive inflammatory changes involving the subaxial cervical spine. You can even see inflammation in the paraspinal soft tissues, epidural phlegmonis, kind of reactive engorgement of the phlegmon. And what was really helpful for troubleshooting in this case was dual energy CT, which was helpful in discriminating by showing the uptake. Sorry, I keep on calling this a nuke study, but showing the crystals on this dual energy CT. So that was a nice troubleshooting tool. Degenerative soft tissue, we see this all the time and this is due to segmental micro motion at different levels. You can see it at the cranial cervical junction. In this case, it's because this patient has altered biomechanics due to lentocipital assimilation. And you can see a retro dental synovial cyst, which impinges on the cervical medullary junction. The cord is squooshed and the patient presented with myelopathy. So these are important things to keep in mind. So just a kind of a conglomeration of things you may see when you are thinking of inflammatory spinal arthropathies involving the cervical spine. You have the cranial settling that makes you think of rheumatoid arthritis. I didn't get into the infection, but infection, particularly in immunocompromised patients can involve the cranial cervical junction. You'll see an inflammation, maybe even a fluid collection. You'll notice the extensive osseous involvement with enhancement. CPPD, as I mentioned, it can be very challenging in this scenario. The CT is very helpful because it'll show you the crystals, the calcification in the transverse ligament. And degenerative changes, very, very common, but can have more extensive degenerative sequelae such as synovial cysts that can impinge upon neural elements. Here's our next case to make us think about spinal arthropathy. So we have a 49-year-old male with low back pain. And you'll see on the sagittal stir, not too much, maybe a little edema along the end plate. Scrolling down, and fortunately, the axial images actually went through the sacrum. We see edema in the sacroiliac joints, a little bit of enhancement in the inferior portion of the SI joints. And this was a patient with sacroiliitis. So now that in and of itself is not indicative of a spinal arthropathy, but at least should make you think. It's in the differential diagnosis when you have a patient under 45 years old, in the 40s with greater than three months of low back pain that has imaging evidence of sacroiliitis and other spinal arthropathy features such as HLA-B27, things like that. So what we're looking for in MR are active inflammatory lesions with kind of these subchondral areas of bone marrow edema, synovial enhancement. And as time goes on, you'll see subchondral sclerosis. You may see blurring of the iliac margin of the joint. And the iliac margin is involved first because the cartilage is thinner there. Then you'll see these thin sclerotic erosions that look like perforations on a stamp. And then they involve both sides as time goes on. And it looks like a rosary bead. That's my illustration skills there. So over time, as the joint fuses, you'll have the bone metaplasia and bony bridging. And on MR, the joints will appear T1 hyper intense. And in radiographs, they're called phantom joints. All you see is this like shadow of the previous joint. And as I said, to emphasize that imaging evidence of sacroiliitis is actually in the criteria for diagnosing spinal arthropathy. So it's important to think about that when you see imaging evidence of that. And then have the clinician kind of follow up with all these other clinical features. Perhaps they have IBD. For example, in this case, this patient had known IBD and presented with back pain. And you can see these more chronic appearance of erosions along bilateral sacroiliac joints. And you can see inflammation and thickening of the rectum and truth be told, I didn't pick up on that, the fellow did, because I don't pay attention to that. But another important, very interesting concept is the fact that there are thought to be biochemical and molecular links between the intestinal and synovial inflammatory process. So there may be some kind of spectrum of autoimmune that's involved in all of these processes. So here's another kind of gestalt of things that can involve the sacroiliac joints and mimics. So we have bilateral sclerotic lesions of the sacroiliac joints in our axial spondyloarthropathy. This was actually ankylosing spondylitis. Then we have this very common triangular sclerosis along the medial margin of the ilium, osteitis, condensans, iliae. DISH, keep in mind that the calcifications occur along the superior aspect of the joint. The synovial portion, which is the inferior third, is not involved in DISH. And here's an unfortunate case of bilateral infection of the sacroiliac joints. You can see these peripherally enhancing fluid collections with an enhancing fluid collection anteriorly. This is a patient with mid back pain. You can see these areas of hyperintensity in the mid thoracic spine. And these are our classic Romanus lesions. So it's inflammation at the enthysial attachment of the annulus fibrosus to the vertebral corners. And on MR, initially, there'll be hyperintense on STIR. Over time, there'll be fatty marrow replacement. And you'll see these corner lesions, usually the anterior superior inferior aspects. These corner erosions have periosteal bone repair. And you have the squaring appearance of the vertebral bodies that's classically described as the shiny corners on radiographs. So CT and MR play very complementary roles. That's the emphasis, with MR showing more structural lesions, earlier inflammation, and CT showing the osseous lesions much better. What role does spine MR play in the diagnosis of seronegative spondyloarthropathies? It's supportive, highly suggestive. But at this point, it's not in the criteria. So you can suggest, if you see spondylitis in this case, that do some tests. This patient may have an inflammatory spondyloarthropathy. It has been used to predict which patients will be responsive to treatments. Those patients with elevated inflammatory markers and these spondylitic changes actually do really, really well over two to three months with treatment. So this can be really helpful. Look for additional areas on your spine MR that you may not readily think about, some enhancement indicating inflammation of the costovertebral joint. On the CT, there's blurring of that joint. There may be inflammatory changes of the facet joints, as well as the interspinous regions. But you really want to look for more than that, the osteitis, the edema of the adjacent bone, where the entheses attach. That's really important. Some mimics would be SAFO. But you'll notice that the edema and inflammation is much more extensive along the end plates. SAFO, as you can remember, also involves other joints, such as the costosternal joint. In this case, there was uptake at the sternomanubular joint as well. So this is a little bit different, but can have that kind of mimic. I'll skip over the degenerative end plate changes that we know so well. And then this is the composite. We will see how ankylosing spondylitis has these corner lesions. SAFO, much more extensive. Degenerative changes, what's very helpful is loss of the disc height, subchondal sclerosis. Osteomyelitis has a little bit more extensive edema and inflammation of the vertebral body. If you will allow me, Dr. Shah, I'll go a few more minutes. I just want to emphasize a couple more important points that should make you think of spondylitis. So here's a patient with chronic back pain. You can see very obvious ankylosis of multiple vertebral bodies on the correlative myelogram. You can see these thins and desmophytes. And there's duralectasia, which can be seen with ankylosing spondylitis due to the whole systemic inflammatory process. The thins and desmophytes have vertical growth. They're vertical ossifications of the outer lamella of the annulus fibrosus. And that's in contradistinction to traction osteophytes, which grow horizontally. So one gestalt number to keep in mind, traction osteophytes grow greater than 45 degrees relative to the vertebral body, while thins and desmophytes are less than 45 degrees. Just a general rule to help tell the difference. Psoriatic spondyloarthropathies have much more bulky syndesmophytes. They're called parasyndesmophytes. They tend to be asymmetric and sometimes unilateral. And here's a nice composite showing the thin syndesmophytes of ankylosing spondylitis compared to the relative preservation of disk spaces and flowing osteophytes of dish, and then the horizontal osteophytes of osteoarthritis. The ossifications can involve the paravertebral and interspinous areas, giving us the very commonly known bamboo spine. This is quite vulnerable to complications such as transverse fractures at the areas of greatest biomechanical load. You can see right below the fusion in this patient with underlying ankylosing spondylitis. And lastly, last imaging feature to keep in mind, when you see these areas of irregularity and sclerosis in the thoracic, thoracolumbar region, think Anderson lesions. There are two different types. There is the acute type, which, sorry, there's type one, which is a more inflammatory type. And it can look very confusing, almost like diskitis osteomyelitis. But you have these preserved cortices, the dark cortices on MRI and the sclerosis on CT with multiple levels of involvement. And then you also have the type B, which is a chronic pseudoarthrosis, either because there may be a skipped area where there's constant motion, or it may be a true fracture that has healed with pseudoarthrosis. And here's a nice composite showing the difference between the Anderson lesion versus a diskitis osteomyelitis, where you lose that cortex. That was the point I was trying to make. And some of the other things that you may encounter and put in your mimics, such as erosive osteochondrosis. And here you have the classic loss of diskite and end-place sclerosis. And with that, I appreciate your time. And thank you for a little bit of leeway there. Thank you. I'm Vaneel Shah. I'm a neuroradiologist at UCSF. And during my talk, we'll continue our journey of looking at painful spine lesions, no pun intended. For my talk, I have nothing to disclose. And I did want to say that I will be using the audience response system for my cases. So I will put up that link shortly. And you can pull out your smartphones. And whether you're in the room or logging on virtually, you can access the ARS. OK, there we go. OK, so here is the link to participate in the audience response questions. It's polev.com slash RSNA S402, which is the room number. So polev.com slash RSNA. Don't forget that S and 402. OK, here is the first case. What is the likely cause of this patient's right S1 radicular pain? A, is it a perineural cyst? B, a nerve sheet tumor? C, a sequestered disc? Or D, an epidural abscess? Okay, we'll call it here. So the majority of the attendees thought that this was a nerve sheet tumor. Well, I'm glad I showed this case. This is actually a sequestered disc herniation. And here is the axial T2, T1, and T1 post-contrast images. And we note that in the lateral recess on the right side, there is a T2 hypo-intense, T1 iso-intense, peripherally enhancing lesion with a central area of non-enhancement. And this is a pretty characteristic look for a sequestered disc fragment, and the enhancement characteristics are typical. Now remember that gadolinium is not routinely administered in patients who are imaged for low back or radicular pain, in whom systemic disease is not suspected. However, gadolinium can play a useful problem-solving role, particularly in the setting of unexplained radicular pain and in patients with post-operative pain. And both gadolinium contrast and fat-saturated T2 or STIR images can depict the physiologic parameters of inflammation. If gadolinium has been administered, as in this case, it is instructive to observe that most of the soft tissue seen with the disc herniation on the unenhanced images and thought to actually represent the disc is largely enhancing inflammatory tissue surrounding a small non-enhancing disc fragment. So at the time that we're imaging after contrast administration, the disc is not enhancing, but the surrounding granulation tissue is enhancing. And the inflammatory response that is associated with the herniated disc that we can see here on this post-contrast image can incite a chemical radiculitis around the adjacent exiting or traversing neural tissue, resulting in pain generation, and in this case, right S1 radicular pain. So it's important for all of us that read lumbar spine CTs or MRIs to be familiar with the updated lumbar disc nomenclature. This is version 2.0, which was a joint task effort of the ASNR, the ASSR, and the North American Spine Society. Remember that herniated discs may be classified as protrusions or extrusions based on the shape of displaced material. This distinction is not generally clinically significant, however, it is notable that in normal patients, extrusions are much less common than protrusions in asymptomatic patients. It's also important to be aware of the zones of the spinal canal, and this is the terminology we should use in our reports. I do want to highlight that the lateral recess is that specific portion of the subarticular zone that is medial to the pedicle. So here I've shown you two images, one at the pedicle level, one at the foraminal level, and the lateral recess refers to the subarticular zone level that's medial to the pedicle. And so it refers to the entire cranial caudate region that exists medial to the pedicle, where the same numbered lumbar nerve root travels caudally before exiting the foramen under the caudal margin of the pedicle. So the lateral recess is a specific part of the subarticular zone. Here's another example of a sequestered disc, and this is a far lateral disc fragment, and we note this low T2 signal disc fragment, that interestingly, this is at the L3-4 level, but the patient actually presented with L2 radicular pain. And we can see here that there is room between the exiting L3 nerve and the disc, but this disc is actually contacting the lumbar plexus and the L2 nerve, and hence causing L2 radicular pain. And we note on the STIR images, again, highlighting my previous point, that there is a fair amount of edema, inflammation surrounding this sequestered disc fragment. Inflammation around discs, sequestered discs or disc herniations, is what is primarily responsible for pain generation, and is the reason why injected corticosteroids around these disc herniations help decrease pain, because they decrease inflammation. Here's another example of a far lateral disc herniation. Here's a normal patient, here's a patient with a large intraforaminal disc herniation that is displacing the L3 nerve posteriorly and superiorly, and this patient did get contrast because they had a history of an underlying, they had a history of a brain tumor, and they were looking for mets, and we noticed that there are no mets here, but there is this disc herniation in the foramen, far lateral, with a lot of surrounding soft tissue that is actually not disc material, but the enhancing soft tissue is granulation tissue, inflammatory tissue that's inciting a chemical radiculitis. Here's another example of a far lateral disc herniation. Notice quite subtle here, but well seen on the axial image, on the axial T1 image. Here's the disc, and here is the nerve that's displaced by the disc. Now it's important to note that the small foraminal disc herniations, particularly far lateral or sequestered discs in the lateral recess, can be easily missed, and the axial T1 images are particularly helpful for identifying these discs, because the sagittal MRI images typically only cover to the lateral aspect of the foraminal zone. Additionally, because disc extrusions or sequestrations are typically brighter in T2 signal than the parent disc, they may not be well visualized against the high signal FAT on axial FASPIN echo T2-weighted images. On T1-weighted images, however, the low signal intensity of the disc contrasts nicely with the bright foraminal FAT, allowing for easier identification. Okay, here's case two. What is the most likely cause of the intrathecal enhancement at L4-5? A, carcinomatosis, B, intradural infection, C, transdural disc herniation, or D, spinal stenosis. Okay, we'll stop the poll. So the majority of you got this answer right. The answer is spinal stenosis. Remember, in this case, we see multilevel spinal stenosis, but at L4-5, where there is severe stenosis, there is intradural enhancement. I don't see any intradural disc herniation, but there is enhancement at short segment at the level of the stenosis. Now, in older patients, particularly spinal stenosis is common, and in some patients may cause the clinical syndrome of neurogenic intermittent claudication. Spinal canal stenosis is necessary, but insufficient by itself for pain generation. Neurogenic intermittent claudication pathophysiology is thought to be a result of venous congestion from compression with the resultant inflammatory reaction and malaria and degeneration that produces blood nerve barrier disruption and intra-radicular edema, which may account for this enhancement. And gadolinium enhancement, therefore, in these patients may provide more specificity in correlating imaging findings with the clinical symptoms in the setting of neurogenic claudication. Now, one more point to make here is that in the setting of multilevel canal stenosis, clinical studies have shown that symptoms most closely match the radicular distribution of the more caudal levels compromised, of the more caudal of the levels compromised, which in this case would be L4-5. Here's just the axial limit showing the severe canal stenosis at L4-5. Here's another example of a disc extrusion here. We can see that the posterior longitudinal ligament has been uplifted by the cyst extrusion. There's severe canal stenosis. Post-GAD imaging shows that there is disruption of the blood nerve barrier and intrathecal enhancement. Do not mistake this for infection or tumor. This is related to the lumbar stenosis at this level from venous congestion and inflammatory changes. So again, inflammation is key for pain generation, even in patients with spinal stenosis. The stenosis by itself is not sufficient for pain generation. Okay, here's case three. What is the cause of this patient's left L5 radiculopathy? A, a facet cyst, B, a discal cyst, C, a cystic schwannoma, or D, a perineural cyst? Okay, so the two most common choices selected were a discal cyst and a perineural cyst, and the answer is a discal cyst. So here we see a cystic lesion with a low signal intensity rim on T2 within the ventral epidural space at L5. The superior margin of the cystic lesion is closely approximated to the posterior margin of the L4-5 disc space. And on the post-contrast images, we see a thin rim of enhancement surrounding the lesion, and on the axial image, we see that the cystic lesion is in the lateral recess. Remember, we're at the level of the pedicle here and medial to it, and it's compressing the L5 root, which is posteriorly displaced. Note that the lesion does not communicate with the adjacent facet joint. This lesion was surgically excised and confirmed to be a discal cyst. L4-5 is the most common level. These are not very common, and as I mentioned, they do communicate with the intervertebral disc. They are thought to result from an injury to the annulus fibrosus with fluid accumulation and formation of a pseudomembrane. They may spontaneously resolve, but may require surgery or percutaneous aspiration if large or impinging on neural tissue. Here's a recent case that I treated percutaneously. This is a patient that developed a left L5, sorry, a left S1 radicular pain syndrome after a L5-S1 discectomy, and we see a similar-looking lesion in the L5 lateral recess impinging on the adjacent S1 root, and we see some inflammatory changes from the recent discectomy. This patient was symptomatic, so we offered a percutaneous treatment. So here's my needle entering the cyst. I first aspirated the cyst, and then I injected a small amount of medication steroid anesthetic with a little contrast to pacify the cyst and then rupture the cyst, and you can see the ruptured contents injected into the cyst freely in the epidural space, and this patient had a really excellent pain relief from the injection. Okay, here's case number four. What is the most likely diagnosis in this patient with back pain and colon cancer? A, a small node, B, sarcoidosis, C, osteomyelitis, or D, a metastasis? Okay. Great, so the majority of you got the correct answer. This is actually an acute small node. This is a symptomatic acute small node that was mimicking metastatic disease, and in fact, on multiple subsequent studies, this was called a metastasis, and it came to our table for a biopsy when we said, just wait a second. Let's look at this a little bit more, and we deferred the biopsy, and we recommended a two-month follow-up. So what do we see? So on the initial set of images, we see a lesion with a central area of low signal, surrounding high signal, and surrounding low T1 and T2 signals, so there's some sclerosis, there's edema, and there's ill-defined enhancement within the lesion. If you look carefully, you notice that there is a defect in the inferior end plate as well, and after two months, the edema and the sclerosis have improved, and there's a more well-defined defect in the inferior end plate. So this is an evolving Schmaltz node. Here's what the CT looked like, and remember, for a metastasis, this is very unusual to see all the sclerosis around the primary lesion, and we note here that this might have been the reason for the node, is there's some calcification in the disk space, and that may have resulted in formation of the Schmaltz node, and we see calcium extending into the node itself, and these scan in the acute setting show uptake on FDG PET or bone scans. So these are uncommon, and like in this case, they may be mistaken for more sinister abnormalities. In the acute setting, we might see surrounding edema, sclerosis, or enhancement, which really indicates vascularization and inflammation, and in that setting, the disk may peripherally enhance, followed by more homogeneous enhancement for six months or more, and over time, there's decreased edema and a well-defined end plate defect, and CT is very helpful in confirming the diagnosis, and I'm almost out of time here, but I'm gonna show you my last case really quickly. Which space is the injected contrast in? Is it, this was during an attempted interlaminar injection. Is A, epidural, B, intradural, C, subarachnoid, or D, retroligamentous? Very good, I think whoever answered the question, or how many of our people answered it got it completely right. This is a retroligamentous injection. This was an attempted interlaminar epidural injection. Here's the needle, but when contrast is injected, we see that there are bilateral facet arthrograms, and this needle was in the retroligamentous space, or sometimes called the retrodural space of OCADA, and this is a potential space that can act as a conduit for the spread of inflammatory or infectious processes, and it's more commonly seen in the setting of PARS defects or facet arthrosis. It can connect ipsilateral adjacent facet joints, as well as contralateral adjacent facet joints, and even the spinous process adventitial bursa, and we can sometimes see it in the setting of inflammation in this space on MRI, like in this case, and we may even encounter it during transforaminal injections, as in this case, and during interlaminar injections under CT guidance with injected air in the facet joints and not in the epidural space. So it's important to be familiar with this space in order to consider an inflammatory process in this space rather than infection when you see edema here on MRI, and also critical for the interventional radiologist or neuroradiologist to recognize this pattern. So in summary, I hope I've convinced you by showing some of these cases that inflammation plays a key role in low back pain syndromes. It's important to be familiar with spine nomenclature, be aware of uncommon locations or unusual imaging presentations of disc herniations, and gadolinium can have a useful problem-solving role. Thank you very much for your attention. Hi, my name is Levi Chazen from the Hospital for Special Surgery in New York City. I'm very happy to be with you here today to talk about spine trauma with a focus on some cases. I have no relevant financial disclosures. So spinal fractures make up about 5% of skeletal injuries. They most commonly occur at the thoracolumbar junction between the relatively rigid thoracic spine and mobile portions of the upper lumbar spine. However, cervical fractures, which are the second most common location in the spinal axis to sustain a fracture, have a much higher association with spinal cord injury, not surprisingly over half of cord injuries are associated with cervical trauma. There are about 12,000 cases of spinal cord injury per year in the US. They're more common in males, and the average age is around 40. Most commonly, those are secondary to motor vehicle collisions followed by falls, acts of violence, and sporting related injuries. Here's the first case for you to consider. So we have a lateral radiograph and an open mouth odontoid view. So I'll give you a moment just to take a look at these pictures. The lateral view looks relatively normal. There's maybe a little bit of widening of the lantodental interval. But when we take a look at the open mouth odontoid view, we can see that the rule of Spence is violated and the lateral masses of the C1 vertebral body are displaced laterally relative to the C2 vertebral body. So that's very abnormal. The patient had a same day CT scan. And here you can see much more obviously that there is a C1 burst fracture, often known as a Jefferson fracture. So what's the best way to categorize these? Well, AF Spine has an excellent classification system for different parts of the spinal axis. So what I'm showing you right now is the upper cervical injury classification system. And there are three buckets you can put these into. One is occipital condyle, a perineal cervical junction injury. Two is C1 ring injury, which we obviously have in this case. And three is C2 and C2, C3 joint injuries. And then based on those categories, you can say whether it's a type A, B, or C injury. Based on the degree of ligamentous disruptions, we can see in our case, we have at least a type B injury with a multi-part fracture. Revisiting this case, one other point that I'd like to make is that while the findings can be seen on an X-ray in this particular case, I would argue they're quite subtle relative to the CT scan where you can obviously see there's a fracture. So this has been borne out in a number of different studies. A study from 2005 looked at lateral radiography and found it only had a sensitivity of 39% for cervical fractures and 52% for unstable injuries. So really no better than a coin flip. CT imaging in that same study had sensitivity and specificity approaching 100%. Another meta-analysis of seven studies, again for cervical spine injury, also confirmed that radiography really only has a sensitivity of around 50%, whereas CT approaches 100%. So what do we do with this information? Well, the first consideration is whether to get imaging at all. And there are two well-validated clinical tools that you can use. One is called the contingency spine rule and the other is called NEXIS for patients to determine if they're low risk and whether imaging is indicated in the first place. So here are the NEXIS criteria just as an example. If the patient has no midline tenderness, no evidence of intoxication, they're alert, their neurological examination is normal and no painful distracting injuries, you actually don't need to get imaging at all. One study looking at almost 9,000 patients and evaluating the contingency spine rule found that it had 100% sensitivity for excluding cervical spine injury. So cervical spine can be clear on clinical grounds, but if you're gonna get imaging, it's best to consider cross-sectional imaging because radiography has a very low sensitivity and a lot of injuries may be missed. A risk benefit and cost benefit analysis found that CT outweighs radiography, both in cost and theoretical radiation risk in low and high risk trauma patients. So if you're gonna get imaging of the cervical spine instead of trauma, consider cross-sectional imaging. Okay, let's take a look at this next case here. So we have a sagittal T1 sequence. You can see that there's abnormal interolasthesis of C67 and there's soft tissue here in the ventral epidural space. We see that again on the sagittal T2, there's an injury here with anterior lasthesis, subluxation, there's widening of the interspinous space. And here's the synaxial section. Same patient had a cervical spine CT. And as is often the case when imaging the spine, the MRI and CT finds a really complementary when there's significant trauma. Here you can see that there's a common fracture of the facet joint and it's anteriorly sublux with a Persian distracted facet joint. So this would fall in the AO spine subaxial injury classification system, which again, I think is very helpful to categorize these types of injuries. Type A are compressive injuries. Type B is a tension band injury. Type C is translational injury. There are some special types like bilateral injuries, if there's a facet related injury as in our case. And there's some modifiers you can use too based on the neurological status and some other special modifiers, for example, if there's vertebral artery anomaly. There's also a nice flowchart here if you like flowcharts. So basically, you can say from the starting point, is there displacement and dislocation? As in our case, there is. So that would be a type C injury right off the bat. If not, is there a tension band injury? If not, vertebral body fracture or vertebral body process injury? And if you say no to all the questions in the flowchart, then it's nice because you end up with no injury. Okay, here's a companion case. So we have a C6 birth fracture with retropulsion and a little ventral tear drop fragment. You can see there's abnormal alignment both above and below the fracture. And here's a corresponding MRI. So what does this add? Well, obviously, this cortisone abnormality is something that we can only detect on MRI, and that's going to have prognostic information. So what are the benefits of MRI over CT? It illustrates the ligaments, so you can tell ligament and disc-related injuries on MRI. You can see epidural hematomas much more readily. And the setting of sclera, which we'll talk about a little bit, spinal cord injury without radiographic abnormality. MRI adds a lot of useful information. If there's vascular injury, so vertebral artery or carotid injury related to trauma, you can check that on MRI. And most importantly, perhaps, is the integrity of the spinal cord. One of the worst prognostic indicators that's seen in this case is if you have T2 hypo-intense signal within the spinal cord, nicely illustrated in this gradient echo sequence here, that's indicative of cord hemorrhage, and that's a very poor prognostic marker. When is cervical MRI not indicated? Well, there have been a number of different studies on this, and if you have a patient who has blunt cervical trauma, but a normal CT and a normal neurological exam, and they're an adult, you can rest assured that MRI has not been shown to give additional information in that setting. This study concluded that it would be at a lower health benefit, concluded that it would be at a lower health benefit and higher cost when followed up with MRI compared to no follow-up after a normal CT. And this bottom study here, more from a surgical perspective, concluded that the addition of MRI to CT and the evaluation of subaxial cervical spine fractures did not seem to affect surgical management. Take a look at a companion case here. So here are right and left parasagittal views on this fat suppressed sequence. Give you a moment to consider that. And here's a corresponding axial image. So what do we see here? The vertebral artery flow void on the right is preserved and nice and dark. This is a T1 fat suppressed image. And on the left side, we have lost our flow void, and there's some T1 hyper intense signal corresponding to the vertebral artery. So this represents methemoglobin in the setting of vertebral artery dissection. You can see that we've lost our flow void here on the left side. This patient had a same day CT angiogram. And illustrated by the arrows, you can see this vertebral artery is small and it kind of tapers out and then is completely occluded, compatible with vertebral artery dissection. So traumatic vertebral artery injuries, they're not often seen when we consider all types of blunt cervical trauma. But if the fractures involve the transverse foramen, the rate goes up quite significantly, not surprisingly. The prognosis with these is relatively good. Over 90% of stenotic dissections will resolve within several months. Typically, the treatment is just an antiplatelet such as aspirin or NSAIDs. And over half of occluded vessels will recanalize. When there is an infarct related to vertebral artery injury, it typically involves the pica territory as shown here. Not surprisingly, because the pica arises from the V4 segment vertebral artery itself. You can get some cross fill from the other vertebral artery, but not always if there is a dissection. These dissections can also be graded. The mildest grade 1 is less than 25% luminal narrowing. And the most significant grade 5 is when you have arterial transsection with active extravasation. Okay, let's take a moment to consider this case here. This is a pediatric patient with a fall from crib. So you can see there's an abnormality of the alignment at C2, C3. And there's kind of what would be an unusual appearance for an adult patient with this kind of rounded off vertebral body morphology. So these are, of course, both normal variants. There's an entity called pseudo subluxation that can be seen in children. It's most commonly occurs at the C2, C3 space, but can also be seen at C3, C4. It should be relatively straightforward to differentiate from a true traumatic subluxation by looking at the posterior cervical line of Swiss chuck, which I've drawn here. So if it's pseudo subluxation, that posterior cervical line should be normal. And if it's a true traumatic subluxation, that posterior arch of C2 will be displaced eventually. In one study showed that it occurred about 50% of patients under the age of eight who were otherwise normal and had cervical spine imaging. And this wedge shape appearance of the vertebral bodies is normal in a pediatric patient related to the ossification centers. Spinal cord injury in children is seen for a number of reasons. The cervical spine is the most mobile portion of the spinal access. Children have increased head weight relative to their body. Their neck muscles are developed. Their facet joints are shallow and horizontal. The SNA process is hypoplastic until about, or underdeveloped until about 10 years of age. And the vertebral bodies are wedge shaped with growth plates. So a number of things set them up. In kids younger than eight, the upper cervical spine is the most common area to be injured. And in adolescence, it tends to be more cervical thoracic as we see most commonly in adults. So there's an entity called sclera spinal cord injury without radiographic abnormality. It's a little bit of a historical designation in the era of modern cross-sectional imaging. So this is first described on radiography in a paper from 1982. And it was observed that some patients who have normal imaging went on to develop neurological abnormalities, including paralysis. So some people use the term true sclera. There's actually absolutely no neural imaging abnormality, and that's much less common. There's an entity called sclostet, which I think is a mouthful, spinal cord injury without CT, evidence of trauma. But what's the take home in these patients? Kids with blood trauma and even transient neurological symptoms. So if they had some neurological abnormality in the field, even if their neuro exam is totally normal by the time they arrived to the hospital, really deserve to have a MRI to look for a ligamentus or some type of spinal cord injury. So I'm going to show you some examples here. This patient had a normal radiograph, and you can see there's a preclival hematoma here. Also normal radiograph. This patient has disrupted their cruciate ligament here, as well as both anterior and posterior lentoccipital membranes. This patient normal radiographs and has a very large epidural hemorrhage. And another patient here who did not sustain a fracture, but did sustain a cord contusion that you can see here with abnormal T2 hypertensive signal within the spinal cord. So again, the take home point for cervical trauma in a pediatric patient is have a lower threshold to obtain an MRI if there's any neurological symptoms, even if they're transient and they're resolved when they present to the radiology suite. Okay, let's take a look at this case here. So we have AP and lateral radiographs of the lumbar spine. Give you a moment to consider that. And as you can see, there's some height loss here of the L1 vertebral body. And that's all I would say is that there's height loss. And I wouldn't call it a compression fracture. I'll show you why. Here is the sagittal and 3D reconstructed CT image. And you can see the fracture goes to both the anterior and posterior portion of the vertebral bodies with some retropulsion. And this patient also had an MRI. So we can see this fracture here goes to both the anterior and posterior portions of the vertebral body with retropulsion. And there's some edema extended to the pedicles, but there's no frank destruction of the posterior ligamentus complex. So thoracolumbar injuries, as I mentioned, the most common portion of the spinal axis to have a fracture. They're more common than cervical fractures. And if you cast your net between T11 and L4, you'll catch about 90% of fractures. A simple anterior wedge fractures make up about half of those. And these occur in two groups, young patients who have severe trauma and older patients with osteopenia or osteoporosis and poor bone density. About one in five of those patients are going to have fractures at multiple levels. So important to keep in mind in terms of satisfaction of search. If you see one fracture, keep looking for additional ones. Usually they're the superior end plate. An isolated inferior end plate fracture may on occasion be associated with pathologic fracture. A burst fracture, like the case I just showed you, is associated with axial loading, often as retropulsion. And you can see some of the other findings that occur with severe axial loading, such as lumbar sleep or calcaneal fractures and sacral fracture. Treatment for these is a little bit controversial, whether it's vertical fixation, vertebral augmentation, or maybe even bracing in some cases. So here's a companion case. What's similar to the case I just showed you? Well, there's a fracture of the vertebral body here, and it goes through both the anterior and posterior portions of the thoracolumbar junction. But what's different? The signal abnormality extends through the posterior elements. You can see there's disruption here of the interspinous ligaments. So we have a true three-column fracture. The eponym for this is a chance fracture. Or probably more helpfully, we can look at the aospine classification system for thoracolumbar fractures. So type A are going to be compression injuries. The first case I showed you would be either an incomplete or complete burst fracture, depending on if there's involvement of the inferior end plate. And this case here would be a distraction type injury. A B1 versus a B2 depends on whether it's a purely transosseous disruption or if there's ligamentous disruption, as in the case I just showed you. So that would be a B2 fracture. And again, there's a useful flow chart here you can use to categorize your fracture type with some modifiers. So it really is a helpful system. So you and the surgeons and neurologists, et cetera, are all on the same page in terms of what fracture type is present and what the management might be. So just some take-home points. For cervical trauma, clinical tools are very useful at clearing the cervical spine. And if a patient satisfies either the contingency spine or the nexus criteria, no imaging is necessary. If you are going to image the cervical spine instead of trauma, consider cross-sectional imaging, because radiography has very low sensitivity and specificity. MRI is somewhat controversial, but probably not needed if you have normal CT and normal neurological examination. Look for concomitant vertebral artery or extra spinal injuries. In the setting of pediatric cervical trauma, get an MRI if there are any neurological symptoms, even if they are transient outside of the hospital. For thoracolumbar trauma, that's where you're going to see fractures most commonly, look for integrity of the posterior ligament that's complex. And the AO spine classification schema can be helpful for categorization. Thanks very much for your attention.

spine tumors

ependymomas

degenerative conditions

spondyloarthropathies

ankylosing spondylitis

spinal trauma

AO spine classification

MRI imaging

SCIWORA

diagnostic features