Good afternoon, everybody. My name is Thiru Anaswamy. I'm a professor at UT Southwestern in the PM&R department and a staff physician at the VA in Dallas, Texas. It's my pleasure to bring this session to you on value of additional imaging evaluation for your spine patient going beyond diagnosis. Fellow faculty on this course with me are Kiana Jenkins, Dr. Haewon Lee, Jayesh Vallabh, and myself to round out the speakers. The purpose of this talk is to discuss how beyond the initial diagnosis, diagnostic studies can be valuable in the management of your spine patient. Each section will have a disclosure before each section and a list of citations peppered in and at the end of each section. And if any of you would require a comprehensive list of references, they can be provided upon request. Without much further ado, I will turn over this talk to my colleague, Dr. Jen Lee. Okay, great. Well, thank you for the introduction. It's very nice to be here today. So over the next few minutes, I'll be speaking on how additional CT imaging may add value in the evaluation of our spine patients. So I have nothing to disclose and if we can move on to the next slide and the next slide after that, actually, thank you. Okay, so the first type of imaging modality that I will speak about is SPEC and SPEC-CT. So a SPEC scan is a nuclear medicine scan. So the images are taken after an injection of a radionucleotide isotope, and this is most commonly technetium. This study is sometimes obtained as a standalone study. CT images in this context are obtained for anatomic localization. So a SPEC-CT is where the images from these two different types of scans are combined. So if we can move on to the next slide, we'll see that combining the information from a SPEC scan and a CT scan allows information about body function or metabolism from the nuclear medicine scan to be overlaid on the information about where and how the body structure looks in the CT scan. So here is an example of how SPEC and CT images combined can help with diagnosing a metabolic lesion. The planar image on the left shows patchy areas of increased uptake, but the location of the uptake is not entirely clear. In the middle panel, you'll see the CT scan of the same region with a notable defect at the inferior end plate of presumably L1 and the superior end plate of L2. But as we know, a CT image does not demonstrate metabolic uptake. So then with fusion of the planar and CT images on the right, we can see a clear correspondence between the abnormal scintigraphic findings and the defects seen on CT. So the pathophysiologic status of the affected region is clearly demonstrated. And in this way, in this case, the diagnosis of dysgitis with associated bone embodiment was made by using both modalities. So certainly, we may come across dysgitis in a patient who presents with back pain, but this is not a typical diagnosis. So how can SPEC and SPEC-CT be applied to other more common problems that we treat in spine and pain? If we move on to the next slide, this is an example of a patient that we may typically see in clinic more often. So this was a 68-year-old female with degenerative spine disease. And her pain had been present for many years and was gradually worsening. She reported pain primarily over the low back. And she did not have leg, bowel, bladder, or balance complaints. She reported severe stiffness in the mornings. And most activities worsened her symptoms. On exam, she had some midline and paraspinal tenderness. An AP radiograph shows spinal deformity with diffuse spondylosis on the left. And then we also see diffuse findings on a T2-weighted MR sagittal image. So it may be reasonable to suspect that this patient has multiple sources of back pain. And there may not be complete agreement between clinicians of the patient's main source of pain simply based on the information that we already have. So if you want to move on to the next slide, we can take this next image in contrast. So this is a SPECT CT in a different patient that clearly shows increased uptake at the right L3-4 facet joint. In one study, there was a high level of agreement among physicians in identifying the area of greatest metabolic activity of facet joints on SPECT CT. So then the question becomes, is this uptake diagnostically relevant? So in the next few minutes, I'll go over a few studies that explored this topic. And the following studies looked only at SPECT scans. So if we want to move on to the next slide and the slide after that, we can look at this first study that sought to determine the predictive value of the appearance of facet joints on SPECT imaging in diagnosing facet syndrome. They found that seven of 43 patients who met the criteria for final diagnosis, which were an initial clinical presentation consistent with facet syndrome, abnormal MR, X-ray, or regular 2DCT imaging findings, in addition to SPECT findings, and a positive response to a facet injection with bupivacaine and steroid. So in this study, the sensitivity of SPECT imaging was found to be 100%, and specificity was found to be 71%. And it was theorized that SPECT offers low specificity because this type of imaging detects small increases in bone turnover. The negative predictive value of 100%, however, suggests that SPECT can be used to select appropriate patients for facet injections and to identify patients who would be less likely to benefit from facet injections. The study also noted that 16 patients, either because of negative response to their injection or because of abnormal uptake on SPECT of a different anatomic area, 16 patients were diagnosed with pain of a non-facet joint etiology. And this included degenerative disc disease, spinal stenosis, and lumbar strain. We can move on to our next study, and this identified patients with presumed facet joint pain based on local tenderness, as well as pain with extension and exercise. So 58 patients met these criteria, and 22 were found to have abnormal facet joint uptake, while 36 had negative scans. These patients all received facet injections of steroid and lidocaine. At the one-month mark, 95% of patients who scanned positive had a statistically significant improvement in their symptoms. And these were measured by the McGill Pain Questionnaire, the Present Pain Intensity Score, and the Visual Analog Score, the VAS. At the three-month mark, 79% of patients who scanned positive had a statistically significant improvement in their McGill Questionnaire answers only. And then at the six-month mark, there was no statistically significant improvement in any of the three measures, although mean results were noted to be lower than pretreatment values in patients who had positive scans. Patients who scanned negative were found to have no statistically significant improvement in their symptoms at any of these time points. The next third perspective study identified 47 patients who were referred for facet joint injection based on their usual clinical evaluation. And these patients were randomized to undergo or not undergo SPEC scanning. So in the group that was randomized to undergo SPEC scanning, they were divided into two further groups. So patients in group A1 were found to have abnormal facet uptake on SPEC. And their injection target was based on these abnormalities on their imaging, even if the target was different from the levels initially planned and ordered by their referring provider. Patients in group A2 were patients who also underwent SPEC scanning but were found to have negative scans, meaning there was no abnormal facet joint uptake appreciated on their studies. So these patients were injected at the targets determined clinically by their referring provider. Patients in group B never underwent a SPEC scan. They received their facet joint injections at the level determined by their referring provider. At one month, 13 of the 15 patients in group A were found to have improvement in their pain score by greater than one standard deviation. And the change in pain score at three months after treatment was significantly higher in this group than it was in the other two groups. However, just as in the other last study, the difference in pain scores between the groups was not significant after six months. Notably, in patients with a positive SPEC scan, the number of facets treated with injection was decreased from 60 to 27. So as the injection plan was guided by SPEC findings, there was some cost reduction calculated by the authors, roughly about $330 per patient. The authors concluded that selection of levels based on SPEC seems to lead to better response to injection with steroid and decrease the number of levels to be treated. So these were trials looking at SPEC only. We can move on to two other studies that looked at SPEC CT imaging for facet pain. So this first study, which was a double-blind study, randomized patients into a group that underwent SPEC CT scans and either received medial branch blocks for increased facet uptake or intra-articular injections if they were found to have SI joint uptake. In the bone scan group, 80% of patients reported greater than 50% pain relief and 70% reported 70 to 100% pain relief. Also in the bone scan group, 23 patients had a change in their clinical diagnosis based on SPEC findings. So this altered their injection plan, whereas in the control group, which did not undergo SPEC CT, they were treated based on clinical diagnosis alone. And in this control group, only 53% of patients reported greater than 50% of pain relief, and then even a smaller percentage, 25%, reported pain relief of 70% or more. So this study concluded that SPEC CT can help in making a correct diagnosis in facet and SI joint pain. So in this last study that we'll look at today in this section, 29 patients with clinical presentation consistent with facet joint pain underwent a series of three medial branch blocks with lidocaine, bupivacaine, and saline on three separate occasions at the levels prescribed by their referring provider. These patients also underwent SPEC CT prior to their diagnostic or placebo injections, but their injection plans were not influenced by the results of the SPEC CT initially. So what would happen is a patient would undergo a series of these three blocks. If a patient had at least a VAS reduction of 70% or a VAS of less than 4 after a block with lidocaine or bupivacaine, the block was deemed to be positive. If the block was negative after the first series of infiltrations, but looking under SPEC CT, a patient was found to have active uptake in a different facet joint that was not targeted in the first series, a second series of infiltration would be done, and this would target the corresponding facet joint that would light up on their SPEC CT. This study found a modest correlation between SPEC CT scan findings and the result of diagnostic blocks. The response rate increased from 17% to 24% with SPEC CT, and there was a sensitivity and specificity of 0.57 and 0.77, respectively. So this study, like prior studies mentioned, also noted a decrease in targeted levels when guided by SPEC CT findings, and theirs was an average of 3.2 levels targeted to 2.3 levels, although we should note that there were only six patients who underwent a second series of injections guided by SPEC CT in this study. So the question at hand would be, should SPEC or SPEC CT scans be used to help guide facet injections? So overall, there is some moderate evidence that SPEC and SPEC CT may be helpful to identify target levels. There is some evidence that suggests that these scans may help to avoid unnecessary interventional care and can potentially reduce healthcare costs. In terms of safety, bone scans are relatively safe procedures. The technetium has a half-life of about six hours and decays into a non-radioactive isotope, and hypersensitivity reactions to this are very, very rare. We should be aware that SPEC CT involves exposure to radiation. The estimated effective radiation dose to a patient undergoing a bone scan is about 2.5 millisieverts, and this is comparable to the annual background radiation of one to two millisieverts a year. The effective radiation dose to a patient undergoing a lumbar spine CT scan in contrast is about 5.9 millisieverts. So we can just slightly switch gears, and in this next section, I would like to look at multi-planar 3D reconstruction and its potential use in spine injections. So to be clear, I'm speaking about reconstruction in 3D, so I'm talking about surface rendering or volume rendering of a CT scan in order to provide a three-dimensional view, which is derived from multi-planar reconstruction. And we should note that this type of imaging is already used in trauma and surgical planning. But the question is, is there an application in spine medicine? So we can look at this first pilot study on the next slide. This looked to demonstrate the feasibility of segmenting lumbosacral nerves and bone on CT with an automatic deep learning computer program, specifically to reconstruct Kambin's triangle, which is often a preferred target for transforaminal epidural steroid injections. So you'll see in this figure an image of an L5-S1 level. This appears to have an inaccessible trajectory to the safe triangle on an axial CT slice, as depicted in panel A on the top. However, with 3D reconstruction, we can see in panel C and panel E that actually there may be an accessible oblique trajectory to the safe triangle. If you take a look over at panel F, this panel overlays the oblique CT plane that demonstrates the path detected on the 3D model, which was not obvious on 2D CT. So this is an example of how 3D reconstruction of the spine can improve injection planning. Currently, 3D reconstruction of a multi-planar CT can be completed manually. However, this manual segmentation is a skill on its own, and even in patients who have multi-planar CT scans available, we would likely not want to do this for every patient that undergoes an injection. So this study showed that deep learning-based segmentation of lumbosacral nerves can potentially make 3D CT reconstruction for injection planning more widely accessible in the future. So in a slightly different application of 3D CT reconstruction, another group introduced the concept of merging multi-planar reconstructed CT scans to create 3D ultrasound images in real time. In this method, an electromagnetic sensor is used to track an ultrasound transducer. The images are then correlated to a preoperative or pre-procedural multi-planar CT volume of the area, and this can then create a 3D ultrasound image. So the spatial relationship between a patient's anatomy, the needle, and the ultrasound transducer are continuously tracked using the electromagnetic sensor, and are displayed with respect to the CT volume. So this image demonstrates the user interface of the system. So on the right, we can see on the top a slice of the CT, and then right below that there's reconstructed ultrasound volumes corresponding to the bottom picture of the live ultrasound imaging. And then on the left, you can see the position of the ultrasound transducer, the needle, and real-time reconstructed images. So this can facilitate the injection, theoretically reducing procedure time and the number of needle adjustments, and improving the accuracy of the spine injection. However, this technology, which was actually published in 2010, has not yet replaced the usual conventional approaches. This does require a more complex workflow, and it does require additional tracking hardware. But perhaps with more advances in the future, this type of injection guidance will become more accessible and allow us to experience the benefits of real-time 3D ultrasound imaging. So this brings us to the end of this overview of some of the advanced CT imaging studies that might be helpful in addition to the usual workup in spine evaluation. In summary, there is moderate to fair evidence based on a handful of prospective and randomized control studies that SPEC and SPEC-CT can help identify treatment targets in lumbar facet syndrome and may be useful in patients with challenging diagnoses. And as we look to the future, there may be a role for advanced CT and ultrasound modalities to help guide injections in both routine and complex cases. Thank you. And I'd like to introduce my colleague, Dr. Jay Vallabha. He will be speaking about MRI modalities. Thank you. We can go to the next slide. I have no relevant financial disclosures for this talk. So what we're going to do is we're going to do a couple of cases here. We're going to look at where MR neurography might be useful clinically. We're going to take a peek at some literature regarding dynamic MRI and also a case where repeat MRI imaging was helpful for us and some related literature related to that. So if we can go to the next slide, we can get our first case going. And so this first case is a 47-year-old female with one-year history of left gluteal pain. After lifting, went to the emergency department. Plain films were normal for her age. They ended up getting a lumbar MRI, which was read as normal for age. But we did not have these images available. Ended up seeing sports medicine for a possible proximal hamstring pathology, got an MRI of the leg, which was normal for age as well, and presented to our spine clinic. When she presented, she was fairly exasperated, no clear diagnosis, had evolving mood overlay. She was becoming a chronic pain patient. And so we really decided to check an EMG here to figure out if we could clean any pathology. This did show an L4-5 pattern to innervation, but did not have paraspinal or more proximal abnormality, so it wasn't fully diagnostic. So here we have radicular pain, potential root abnormality, and a normal MRI. So we felt the next best step was to obtain an MR neurography. And so on the next slide, we can see the scan. And so the MR neurography did show some asymmetrical enhancement of the left L4 nerve root here. So we ended up doing multiple transpiramidal epidural steroid injections. She completed further PT. But despite all this, she continued with radicular pain and was referred to one of our surgeons for surgical evaluation. So here, our surgeon, and actually previous slide, our surgeon did not feel comfortable planning surgery around this MR neurography alone because we still could not obtain her original MR imaging. So the surgeon decided to obtain a preoperative MRI. And so this is actually just a normal MRI scan. And on the left, we see a T1 sagittal view of the L4 neuroforamen. It really doesn't look too bad. The MR neurography tells us there's some neuritis there, however. And on the right, you see on the axial cut, there is some subarticular disc abnormality on the symptomatic side. Eventually, the patient did undergo TLIF at L4-5, which significantly improved her leg pain. And so if we go into the literature on the next slide of when could this be useful? So the question is, what's the overall utility, right? Should more patients with radicular symptoms just get MR neurographies or not? I don't think the evidence suggests that MR neurography should usurp traditional imaging, but does show a potential role in these scans. So Chabert et al explored a small group of patients with radicular pain, 50% with motor symptoms. All of them had what were felt were to be non-contributory MRI findings, and subsequently underwent MR neurography from T12 to L1 to the lesser trochanter for evaluation of performers as well. In this study, the radiologists were blinded to all the clinical information on patients, including their prior imaging. These authors reported that with non-contributory lumbar MRI imaging, most of their patients showed extra spinal pathology, although some did have root level changes as we had in our case. If we look at the next study from Lewis et al, they looked at 14 more patients in similar situations, and they showed a high proportion of these scans did show pathology with regards to extra spinal abnormalities. Chaz and et al did a retrospective review on the next slide of lumbar sacral neurography over a three-year period. They looked at 300 plus scans and clinical information, pulled out about 64 of these scans. Patients who also had EMG and nerve conduction studies done within three months of symptom onsets. They found significant association between neurography findings and EMG abnormalities, but interestingly not physician reported signs of weakness, sensory abnormalities, reflex abnormalities, straight light rays, and intraneural signal abnormalities. So these authors noted that the neurography correlation with EMG might support potential clinical utility in patients and may help guide therapeutic options similarly to EMG in some scenarios. And one thing I kind of like doing as predominantly clinical faculty is what's the realistic, is it realistic for me to obtain these scans? So on the next slide, we just kind of briefly look at some insurance coverage, small sample here, but mostly non-covered as experimental. Evercore seems to be most generous and would cover this in instances where more traditional evaluations are non-diagnostic or equivocal as in our case. And from what I could tell, Medicare doesn't have a specific LCD on MR neurography either, but I welcome comments at the end if anyone has any different information on that. And so moving along, you know, what about dynamic MRI? You know, what is this? Why do this? Well, this is pretty much an MRI scan in various positions. Why would someone do this? Well, the main argument would be that we are dynamic, we move, we have pain with walking and standing, sometimes supine imaging may miss relevant pathology. And this is really shown in the next slide by Michellini et al who did a literature review. They report that the utility of standard imaging is often a viewing in patients in a position of rest or a position of no pain, and that that does miss pathology. So they did a pretty reasonable review, broke this down by body area. So the next slide, we can see the cervical spine, and they reported on their review of studies that found, you know, C2 through CIS discs, six discs tend to move posteriorly with motion, C7 tends to move anteriorly. As you can see pictured here in the flexion versus extension view, there are more posterior disc herniations with the extension view, and you can also see some ligamentous-based stenosis at C5-6 level. And so this is a kind of a clear depiction that, you know, when we move, there are parameter changes within our canal. And so on the next slide, they show us how this works in the lumbar spine. So they show that loaded or standing position alters the lumbosacral angle. They show that canal surface area and neuroforaminal diameter may decrease with loading. And so you can see there that even the facet is, which has fluid in it, gets compressed. This also narrows the lateral recess on that side and also the neuroforamen on that side. In the next slide, Zhou et al took this a little bit further, and their study was a little, about a year after Michellini's review, and they looked at standing MRI images, and they really showed that dural sac cross-sectional area can significantly change in some folks. And you can see this on the left, where you have a fairly normal cross-sectional area, and then in a standing image, this decreases significantly. And so this becomes interesting when we start thinking about, you know, what are the findings that can cause symptoms, and are we seeing patients who have symptoms but our normal imaging might be missing? And so, again, I always find it useful, what does insurance think about dynamic MRI? So on the next slide, we find that, well, really, typically there is no approval for multi-position MRIs. You can certainly get an open MRI or a standing MRI, but most payers, from what I could tell, are not gonna pay for really two views, essentially, again, paying for two scans. And so this is something that is interesting, but as far as something dynamic, getting something supine, getting something standing, might not be realistic for those of us who are predominantly clinical to start using this. On the other side, if one were curious, you could potentially get an open MRI and just have a standing MRI as your initial evaluation. Moving on to kind of our next case, this is where repeat MRI was really helpful in our direction of treating this patient. And so here's our case, very active young lady, several weeks of right reticular pain in an L4-5 distribution. Pain was refractory and conservative management, PT, NSAID. Subsequent MRI did show a right subreticular disc herniation which we felt corresponded to her symptoms. She underwent multiple epidurals, incomplete resolution of her symptoms. Ultimately, she opted for cervical management due to ongoing limitation in work and activity. And our surgeons, again, ordered a repeat MRI. And on the next slide, we can see that this patient, her disc herniation regressed nearly completely despite ongoing symptoms. And overall, over time, her symptoms did improve, but this was important as she was really on the cusp of getting a discectomy here. And this repeat image really shows there was nothing to substantially fix for the surgeon and that we just counseled her on further time and she did improve. So on the next slide, we really, the question is, other than with changes in symptoms, new symptoms, new neurologic abnormalities, should we regularly repeat these scans? The answer is kind of messy there. It's maybe, maybe not, depends, kind of wishy-washy. The when we should do that is a question that maybe we can explore. And so kind of how it went about this was looking at two main types of pathology that we might treat, particularly the lumbar spine. And so on the next slide, there was a study that looked at non-surgically managed symptomatic lumbar disc herniations and looked at when do these tend to regress, how frequently do they tend to regress and at what time points, which can maybe help us with that when. And so it's kind of a long and busy study by weighing it out, but they looked at the incidence of regression for disc herniations. They did not separate an extrusion versus a protrusion versus something sequestered. They just looked at the literature as a whole. And because of that, they noted a lot of heterogeneity within the studies. And a lot of them were also not clear on symptoms. So certainly some limitations with this. They did find that there were wide differences in when these scans showed regression, somewhere between 45 days and 77 months. It's a huge window. They ended up doing a pooled regression rate, found out that to be about 63%. So patients who have a symptomatic lumbar disc herniation, they felt that about 63% of these will regress at some time point. And that most imaging is likely to improve within a year with these disc herniations. And that repeat imaging prior to surgery seems reasonable. What's that timeframe? Well, they say that repeat imaging might be helpful for our surgical colleagues after the four month time point of symptoms, because that's when the first group of these tend to regress. And then certainly by the 10 month or 12 month time point, you'll see more regression there. And so what about patients who don't have disc herniations? What about patients who have degenerative lumbar spinal stenosis without disc herniation? And the next study really goes into that. And that was by Li et al. And Li et al looked at 103 patients and 167 lumbar levels, and they compared initial and pre-op MRI scans. And I think the takeaway here is that most of these scans did not change within a year. So although our lumbar disc herniations, more than half of them may change within a year, with our degenerative lumbar stenosis patients, less than, I should say, more often than not, those are not changing within the year. And that repeat scans might be helpful for surgical planning after the one year mark. And so again, guidelines are very kind of broad on this, and insurance criteria is very broad on this. And so I think it's important to note that the American College of Radiology appropriateness criteria is reasonably conservative. There's been studies that show that patients who have had recent lumbar MRIs and new symptoms may also not need MRI scans. And so apart from surgical planning, it seems that repeat MRIs are not useful within certain time periods. And so I know that's fairly ambiguous. In our group, we tend to repeat scans fairly frequently within one or two years, particularly when we're considering spinal interventions. That, based on what we've reviewed, seems fairly aggressive, actually. And so that's it, thank you. And then without further ado, Dr. Anaswamy is going to talk to us about ENGs. Hi, it's my pleasure to take the baton from Jay. And today I'll be talking about a case where EMG and biomarkers and a EMG technique was useful in management in spinal stenosis. Disclosures are that some of this work that I'll be talking about was supported by the AANEM Research Foundation Grant, Foundation for PM&R Grant, and departmental funds from UT Southwestern through the Shark Tank Research Funds. The objectives of this case is to review a case-based diagnosis using electrodiagnostic study information to help management. The diagnostic prognostic utility of EMG nerve conduction study, munix and motor unit nerve index and biomarkers and research implications and clinical implications will be discussed. The case is that of a 60-year-old man with hypertension, hyperlipidemia, and prediabetes. He presented to us with low back pain and predominantly left leg pain more than right leg pain and paresthesias. These symptoms worsened with walking and standing and relieved by sitting or leaning forward. And there were no red flag symptoms to worry about or be concerned about. So the clinical diagnosis of lumbar spinal stenosis with neurogenic intermittent claudication and a left greater than right lumbar radiculopathy was suspected. Patient also was taking diglufenac gabapentin methocarbamol with moderate to minimal success. And he had previously tried other treatments such as physical therapy, chiropractic care, TENS unit, and traction without any sustainable or sustained benefit. So the MRI of his L spine showed multilevel mild to moderate central canal stenosis at L3-4, L4-5, and L5-S1 with associated lateral recess and foraminal stenosis as well. We did an EMG nerve conduction study which showed significant abnormalities on the left side corresponding to distribution of his symptoms. This included active denervation in the tibialis anterior, gastrocnemius, and with chronic re-nervation findings at multiple levels. And this corresponded to polyradiculopathy acute on chronic primarily involving the L4 and L5 roots. So the question that we came back with is in order to help with his ongoing radicular pain plus neurogenic claudication symptoms, is it appropriate to consider a lumbar epidural steroid injection? Several studies, including a large multicenter randomized control trial from Friedli et al indicated that epidural steroids in the case of lumbar stenosis is unlikely to be of benefit compared to epidural lidocaine alone. So the overall, since that study and other studies that have come out since then, the utilization of epidural steroids for stenosis cases have come down appropriately so, but is there an exception in this case and how do we use EMG to potentially figure that out if there is an exception? EMG is just to take a little bit of a step back to give a background. Diagnostically, the diagnostic role in this case has been more than well-established. It is more specific than MRI in diagnosing impairment. A normal EMG can rule out a neurological process and point you more in favor of a musculoskeletal mimicker. And it is two to eight times more likely that it can confirm your diagnosis. So it's very clear, diagnostic utility is established. But prediction of outcome has also been studied. There are several studies, including one that we did and published back in 2012, that showed that a positive EMG or abnormal EMG, those patients tend to be more likely to respond to an epidural injection. Likely that the analytical framework of this is that a radiculopathy, which is abnormal on EMG, indicates that there is nerve inflammation and nerve injury resulting in axonopathy, which shows up on the EMG. And because it establishes a clear nerve injury by means of the neurophysiology, it is likely to therefore respond to the anti-inflammatory treatment. So there's kind of a pathological, a pathomechanistic basis behind it. The conclusions of these studies, including ours, is that the clinical presentation can be heterogeneous, and therefore an abnormal EMG will help you narrow down those cases that are more likely to respond to an epidural steroid injection. Therefore, based on this kind of information, based on these kinds of research studies, we decided an epidural steroid injection is more likely to benefit than not. But are there other indicators that you can use to make that decision? So we did a study where we looked at biomarkers, biomarkers that are indicative of inflammation. We looked at a series of cases. This was a pilot study, so the N is not very high. We performed EMG on all of them, as well as obtained serum for biomarker analysis, cytokines, and on several patients, also obtained an epidural lavage. So we injected saline and aspirated it right out, and then analyzed for biomarkers. The biomarkers that we analyzed was a cytokine panel, including these markers that's on this slide. And many of them got blood and epidural markers, but all of them got epidural markers. And this is a list of 11 patients on which we studied and these are the demographics and EMG findings. As you can tell in our 11 patients, more than half had either plausible or clear evidence of radiculopathy. So that was the kind of population we were dealing with, stenosis with radiculopathy. If we looked at a more traditional biomarker of inflammation, a high sensitivity CRP, only two to three patients had elevated levels. So this is not something that you can get from a traditional clinical test. So looking at the laboratory research biomarker levels, this is just a snapshot of slides that gives you a visual of the LAVAGE biomarkers. And then this one is the serum biomarkers. I don't, I'm not going to belabor the details here, but the point is several of these biomarkers levels were elevated in the serum and some of them were also elevated in the epidural lavasate. And when we looked at the outcome of the epidural injection, there was a tendency to improve from baseline to one month and sustained at two months, except for the PDQ, the pain disability questionnaire, and the Swiss spinal stenosis questionnaire that regressed back a little bit to the mean. The correlations of individual biomarkers and the EMG showed promise that there were some correlation between outcome improvement after an epidural injection and the initial baseline levels of the biomarker or the abnormality noted on the EMG. So based on this pilot study, we found that there are some associations that correlates with improvement in pain and function after steroid injection. And these preliminary data, this pilot study indicated that biomarkers, in addition to EMG, can be promising and may be indicative of this kind of phenotype. The phenotype being that the lumbar degenerative pathology results in not only a mechanical nerve injury, but also an inflammatory response, which can be measured and is correlated with improvement in an outcome when you treat them with an anti-inflammatory sort of treatment. A fully powered study would be, therefore, however needed before we would advocate or recommend any of these kinds of laboratory tests that are at the research stage to be used in a more widespread clinical means. So right now, it is not something that we would recommend to be used clinically. So based on this, do we need to get cytokine levels? Not right now. So perhaps in the future, we can talk about a more clear phenotype profile that can be informed by biomarker analysis. We also looked at whether or not we can longitudinally evaluate patients like this. Is there a more quantitative way to assess axonopathy rather than a qualitative way based on diagnostic EMG? We looked at three groups of patients, also a pilot study. This was published by Cai et al. And a shout out to Daniel Cai, who was instrumental in completing the study, getting it funded and completing and publishing it. The three groups, the study purpose of the study was to look at discriminatory ability of MUNIX, which is a motor unit nerve index that quantitates axonopathy and axonal loss, which is useful for neurodegenerative conditions like ALS. So we explored the possibility whether this could be useful in stenosis, which is also a progressive axonopathy or neurodegenerative condition. And we had two groups, one was symptomatic stenosis, one was asymptomatic anatomical stenosis, and the third one was no stenosis, but peripheral neuropathy or other neurological symptoms. The details here are not very important, but we obtained a measure, an index of MUNIX at multiple nerve muscle combinations. We grouped them according to the predetermined group and we looked at baseline symptomatology and functional impairment. As you can tell, group A, which is a symptomatic group, was significantly, statistically significantly more impaired in their Swiss spinal stenosis questionnaire compared to the other groups. So clearly we did our job in grouping them appropriately. And then when we looked at the, however, when we looked at the MUNIX values, they did not discriminate between those three groups. So we had a negative results in a sense. The patient that we presented with, that I started this case discussion with, is patient number 12, and he was part of this study group as well. So the utility of MUNIX, it's a little bit early. So perhaps it has utility in the future, but based on this small pilot study, it did not seem to discriminate. So we discussed potential models of clinical presentation of the phenotype of how MUNIX could be utilized. So we think in lumbar spinal stenosis, one of the reasons why MUNIX did not discriminate these three groups in this particular study is that denervation-renervation in spinal stenosis is kind of a intermittent process. It's not enough of it happening long enough or severe enough to where it could impact the MUNIX level. Also, some of the symptoms in stenosis could be due to a dynamic conduction block, not necessarily axonopathy. Therefore, in that instance, even though EMG also would be negative, MUNIX would clearly be negative or would not be affected if their degree of axonopathy does not, is not enough to merit a quantitative decrease in the number of motor units. However, if this patient continues to worsen or deteriorate, and if he is reluctant to get surgery, but at some point it is, you know, as a physician treating him, if you are considering guiding him towards surgery, perhaps MUNIX could be useful. If it right now was not affected, it did not go below what is considered normal for a patient like him. Maybe in about a year from now, maybe in about a year from now, if his symptoms deteriorate, you get a MUNIX again, and if the MUNIX significantly changes, and based on what Jay presented earlier, if the MRI does not change, but MUNIX changes, you could use MUNIX to be a little bit more specific to axonopathy and loss of motor units, and then more appropriately determined to refer the patient to surgery and counsel the patient accordingly. So perhaps it has some utility there, but right now, based on this pilot study, we did not find it to have a diagnostic utility or discriminatory ability. So to wrap up, what did we do on this case? Well, we actually did a series of epidural injections based on the initial diagnostic EMG, which helped us understand that he had active denervation, and accordingly, his epidural injection provided excellent relief, so we were thankful that we used our diagnostic test to guide treatment in this way. On an unrelated note, while I was prepping the case for this presentation, I was able to also see, unfortunately, the patient had passed due to some unrelated illness in November of last year, not due to COVID, however, at least as far as I can tell, it wasn't that. In conclusion, electrodiagnostic studies can be useful in spinal disorders, not only for diagnostic purposes, but also for prognosis, also to guide therapy, and to help us counsel them better, giving us a little bit more, to borrow the term that our plenary speaker used yesterday, to be comfortable with our uncertainty and to use uncertainty, but also helping us guide, use these kinds of indicators to guide them, even in the ocean of uncertainty. Other psychological, functional, and patient variables will also need to be considered, though, because despite all of these research, we still don't have a confirmed phenotype that will predict outcomes in, you know, more than 60 or 70 percent of the cases. This concludes my presentation, and I'll stop sharing, so we are open for Q&A. I encourage the audience, if you so wish, to please submit your questions on the chat box, and we have about 12 minutes left, and I'll read out the questions, and I'll ask the speakers to chime in, if it is appropriate to them. Jay, first question is to you. What makes MR neurography different than a STIR sequence image on a diagnostic regular MRI? Oh, wow, you're really testing me here. Okay, you know, I don't know the technical details, so full disclosure on that. I would have to defer to one of our radiology colleagues, but from what I'm aware, the MR neurography is a fat-suppressed T2 scan. It uses a higher field magnet, so the slices have a little bit better resolution. It's a little more sensitive to anisotropy within the nerve, and determining like endoneurial inflammation, or endoneurial, I shouldn't say inflammation, but endoneurial signal change. The STIR, as we know, it's predominantly fat-suppressed, very sensitive for water, and I have looked at some things that say, you know, should we be adding STIR coronal views to be able to see part of the plexus in general on our normal MRIs, which can give us some information, extra spinal pathology, just by adding that coronal view, and that doesn't seem to be standard, but to kind of answer your question, I think it's predominantly a resolution difference there. Thank you, Jay. The next question, I believe, is addressed to me. I understand you aren't recommending the biomarkers in clinical practice quite yet. What is the approximate cost of biomarker lab panel for patients? So, what I can tell you, in this study, we used our core lab in order to do the biomarker analysis or assay, and they run a plate, and each plate is about $3,000 for the laboratory, and in each plate, you can run up to 80 or 85 cells, so I guess, you know, per patient, the cost is not bad, but you have to accumulate enough samples to run a plate. Otherwise, you can't keep running a plate for each individual patient. So, unless the way you do the analysis changes to cater to individual cases, right now, it is just not something that we can convert the cost to individual cases and apply it to clinical practice, but that's a great question. Thank you. All right. I have one question from Jacob Strong. Actually, it is addressed to multiple people, but maybe Jen can answer first. How about whether SPECT-CT offers more than MRI STIR images when trying to identify pain areas? And Jay, you can chime in as well after Jen. That's a great question. I actually hadn't come across anything looking at that specifically. Certainly, the advantage of SPECT-CT is that it offers information about metabolic activity. So, whereas MR-STIR might be able to demonstrate areas of inflammation, to a certain extent, SPECT-CT can show bone turnover, which is why it's a particular interest in spine and facet disorders. Jay, anything to add to that? No, I would agree with that. I think just in general, it's, you know, we all know it's hard to look at these scans and say specifically, you know, just based on the result of the scan, what's causing someone's pain. But, yeah, I agree with what Dr. Lee said. Great. Thank you. The next question is from Patrick Cordobin. Regarding EMG utility for ESI, treatment is typically multifactorial for those patients. So, how does that play into decision making to perform the EMG or not? Most appropriate first step would be, seem to review MRI to ascertain if fully concordant with patient's history and physical exam. So, true, whether we like it or not, the patient probably already has an MRI by the time they see you for potential management. So, really, the decision comes down to whether or not to do an EMG. If the diagnosis is pretty clear and it's concordant between the history and physical exam and the MRI, the EMG is really not very, it's not something that you would normally lean towards to diagnose the case or to determine ESI. You would probably go ahead and do it. But if the MRI is equivocal, or if perhaps you may have done an epidural before they even got to you, or if the history and physical is also kind of questionable, the patient has a strong pain behavior component where they are resistant to move and you can't even get a reliable straight leg raise on the patient. You don't, you can't get a reliable tendon reflex on that patient. So, if the physical exam is unreliable because of the way the patient presents to you, and the MRI is unclear or nonspecific, maybe in that case, EMG would be a little bit more useful than if it's a clear-cut case. So, that's probably how I would answer. I hope I answered your question, Patrick. So, while we wait for another question to pop up on the screen, we have five more minutes in the presentation. I have a question for Jen. How about insurance approvals for SPECT-CT, and maybe if you can answer that about, you know, CT-guided injections as well. Yeah, that's always the elephant in the room question, and sometimes just the main question. So, you know, for surgical planning, for trauma, for, you know, questions about infection, you know, these CT scans, SPECT-CT, are covered by Medicare and by most insurances. There, I could not find any information about, you know, whether or not this would be approved for specifically injection planning. I would suspect that might be a little difficult to get by, might be dependent on who is on the other side reviewing the case, if that ever came into question. But for injection planning specifically, you know, if there was a, you may be able to get it done if there was just a question of diagnosis. Thank you, Jen. Just a quick check on the chat box. Okay. So, let me ask Jay a question about the user inter-rater reliability and the user training and influence on how someone reads an MR neurography. What have you seen, whether it is a very user specific or is there some inter-rater reliability? So, in the studies that I reviewed, they all noted that inter-rater reliability was very, very high. Now, a lot of us interventionalists will review our own imaging. I will. I'll look at my own MRIs. I won't write a formal read by any means, but I know some people do that. My ability to go through an MR neurography scan, unless there is that clear, obvious abnormality like there was in our case, it's really challenging for some of us who don't have that training. But for the radiologists in these studies, the inter-rater reliability was high. Now, they could have been chosen to participate because they had that experience. And so, gauging that at your institution with regards to the acumen of your radiologists and trusting that is something you should look into. It's easier for me to say that in a big academic institution. Thank you. A question to Jen would be about CT guided injection. Have you found it to be useful in conditions like Z joint synovial cyst when you try to do percutaneous rupture? Yes. So certainly, there are folks who are doing inline CT guided injections. And the benefit is that compared to regular fluoroscopy, you do get much better visualization. And when your target is something like a Z joint or synovial cyst, the difference in that visualization can make a big difference. Unfortunately, the jump to using 3D remodeling is not widely done, to my knowledge. And it has been done really just in studies. Thank you. And Patrick Corbin put in a referral or a citation there for a review. Thank you, Patrick. We'll take a look at that. I also wanted to, while we're waiting for more questions, and we have two more minutes, do any of the, did Jay or Jen have any questions for me? I did. We only have a couple of seconds. But so we know that, at least from what I've read, the sensitivity of axon damage from radiculopathy is fairly high with EMG. Now, that's with axonal injury. But the sensitivity of radicular pain in general is probably very poor. So it is very poor. But where would you say MUNIX would sit? Because you're relying on that axonal injury, it seems, right? Correct me if I'm wrong. And is it going to fall into the same group as just normal EMGs with these evaluations? Yes. So MUNIX would be something that you would use when you have a clear deterioration of motor function and an expectation that the number of motor units available in that near nerve-muscle combo keep up with the amount of EMGs that are muscle combo keeps going down. So which is why it has utility in ALS. It's a quantitative way to assess, you know, viability of motor axons. So because stenosis is a variable process, it's not a progressive loss in motor units. It may not have utility in many cases, but maybe in some. So it does not have sensitivity sort of utility. It has definitely more longitudinal utility, I feel. All right. Last question before we end. There are multiple steroids and doses. Many of us use interventional specialists. What specific injectables were used in your studies? In the time we have left, I can say that it was the beta-methadone or dexamethasone. Thank you all for your attention. Thank you, Jay. Thank you, Jen. I appreciate the opportunity and I enjoyed being a part of this presentation. Thank you all for attending. Have a great rest of the meeting. Thank you.

additional imaging evaluation

spine patients

SPEC scans

facet joint pain

treatment targets

multi-planar 3D reconstruction

ultrasound imaging

EMG studies

epidural steroid injections