Good afternoon, everyone. Welcome to the last session on Pediatric Body MRIs. My name is Govind Chavan. I'm a staphylogist at Hospital for Sick Children in Toronto, Canada. This is the last session but not least. We have a gamut of very interesting aspects of the Pediatric Body MRI in these four talks. To kick-start this session, it is my great pleasure to invite Dr. Andrea Doria, a staphylogist at Hospital for Sick Children and professor at University of Toronto and my colleague. She will be speaking on whole body MRI in pediatric patients, clinical applications and up-to-date protocols. Andrea. Okay, so this is my talk about whole body MRI in pediatric patients as Govind mentioned and we're going to discuss clinical applications. So I'm going to talk about techniques for whole body MRI from the perspective of non-cancer disorders, indications and protocols. We're going to focus on four questions. If the whole body MRI is useful in chronic non-bacterial myelitis, if it's useful to evaluate temporal mandibular joints, if contrast is useful and if diffusion-weighted imaging is useful. And in terms of cancer disorders, also I'm going to discuss indications and protocols from the perspective of patient positioning and additional sequences. The learning objectives of this session are to understand the clinical indications for whole body MRI for cancer and non-cancer disorders. We're going to discuss techniques and protocols for different conditions, recognize the appropriateness of contrast-enhanced whole body MRI and evaluate the value of novel clinical applications of whole body MRI in early cancer discovery. So whole body MRI is a technique that is excellent for assessment of the entire body. That's why it's so important for a guidance of therapy and follow-up imaging. On the other hand, depending on the protocol, it can be very long to scan, can be costly, especially in some systems where each part of the body is caused separately and is charged separately, and low access for some populations. Typically, we use torso body coils that cover head to toe, as you can see in this picture, and the spatial resolution for whole body MRI is lower compared to a dedicated coil, as we can see here. So this patient has chronic non-bacterial stromylitis. On the left side, we see involvement of the bone of the proximal clavicles with soft tissue changes, and you can see the difference in the spatial resolution between whole body MRI and dedicated MRI. In this case, we also injected contrast materials, so you can see more detail in terms of the soft tissue changes adjacent to the bone marrow edema. In terms of utilization of whole body MRI, there is a survey that was distributed among 471 physicians from this SPR membership. Most of them, 93%, practice in academic institutions and had used whole body MRI for less than six years. And based on this survey, the disorders that people use more commonly whole body MRI were CRMO, chronic recurrent multifocal stromylitis, or chronic non-bacterial stromylitis, and cancer predisposition disorders. And these are the entities that we're going to focus in this session. So what are the non-cancer whole body MRI indications? There are many papers that have really evaluated that, and focusing on this one that was published in Pediatric Radiology in 2018, more than 70% of the indications are for two pathologies, anthocytes-related arthritis and chronic recurrent multifocal stromylitis. So what is a CRMO or CNO? This is an entity that is characterized by a fluctuating course of episodes of pain, and typically distinct from chronic pyogenic osteomyelitis, we don't find a causative organism. And we don't see abscess formation, fistulas, or sequestra, as we see in the pyogenic pattern. And in up to 20% of the cases, we can see skin disease. On the other hand, ERA, anthocytes-related arthritis, is a subtype of juvenile idiopathic arthritis, and have some characteristics that can overlap. So not infrequently, we see a superposition of patterns of CRMO, CNO or CRMO, and anthocytes-related arthritis. In this case, for example, we can see that there is the flame-shaped appearance of bone meridema in the metathesis of the long bones. We also see changes along the sacroiliac joints. In this case, we didn't have involvement of the spine, of the vertebral bodies, but we had involvement of the anthocytes at the level of the greater trochanter. So in this case, we have superposition of two entities. And this has an impact on the protocol that we're going to use, as we're going to see. Also, an important thing is that to use whole-body MRI for follow-up of therapy. In this case, for example, this patient had chronic non-bacterial osteomyelitis, and you can see multiple abnormalities, bone meridema and degeneration of the end plates. In the pre-treatment whole-body MRI scan of the spine and post-treatment, some of them have improved. And in our study, we have shown that in 34% of the cases, one-third of the cases, there is improvement or total resolution, but we can also have some lesions that have progressed. And in 8% of the cases and in 16% of the cases, we have new lesions. So whole-body MRI is extremely helpful in evaluating response to therapy. So now let's discuss the protocols for chronic non-bacterial osteomyelitis and ERA. So this protocol that is here, that includes coronal inversion recovery of the entire body and sagittal stir of spine, bilateral knees and ankles, and coronal oblique views of the sacroiliac joints is usually used for ERA, and the scan time is 45 to 60 minutes. On the other hand, in our institution, we use for chronic non-bacterial osteomyelitis one protocol that only includes coronal stir and only sagittal images for the spine, which reduces this scan time to up to almost 15 to 30 minutes. So there is no free lunch in radiology. If you want to have more imaging, it takes more time. And this is what you can see with the whole-body MRI. We see multiple stations with different parts of the body, and there are some stitching of all the parts into reformats. In the ERA protocol, we have additional sequences in addition to the evaluation of the spine, including the sacroiliac joints, knees, and ankles, and this increases the time of scanning from 25 minutes to 45 minutes at least. So now the position of hands. This is very important because usually we ask the patients to place their hands on or behind their thighs and or above the head, but it's difficult to be above the head because it's very tiring. We try to avoid having the position of the hands along the body because it difficults the assessment of the small bones of the hand, including carpal and metacarpal bones. So now the questions. Can whole-body MRI be useful in chronic non-bacterial osteomyelitis? So this is a case of an 11-year-old boy that the X-ray did not show any abnormalities in the sacroiliac joints, and we conducted a whole-body MRI. And we could see a focal area of bone meridema along the right sacroiliac joint inferiorly. And also interestingly, with the inversion recovery sequence, we could also see abnormal signal in the left talus and the stone metatarsus. So for a biogenic osteomyelitis, so having multiple places is less common. And we also had an incidental finding of a cystic lesion in the left lower renal pole. So what are the findings that we can see on whole-body MRI? It's excellent for assessment of bone meridema, enthusiasis, fat infiltration, ankyloses when you have dedicated coronal oblique views. And it's less helpful without chondrus enhancement, as we're going to discuss, for assessment of synovitis. And other dedicated imaging is needed for assessment of cartilage loss and subchondrus cysts. So now the second question. Can whole-body MRI be used to evaluate temporal mandibular joints? The answer is no. I think it's very poor for assessment of the chondius. This patient had superposition of CNO and ERA. And you can see here that this is a lesion with remodeling of the left mandibular ramus. And this is related to CNO. But the patient also had TMJ. And you can see that within a short period of time there was degeneration of the mandibular chondi that would not be able to be seen with whole-body MRI. So there these are some indications specifically for dedicated imaging. Question number three is chondrus administration needed for whole-body MRI. So initially people used to inject gadolinium for assessment of a whole-body MRI. But because of the very erratic distribution of chondrus material nowadays we don't use. And multiple studies have shown that we can evaluate bone marrow abnormalities as well with STIR as we would do with chondrus enhanced imaging. Exception is just for dedicated imaging that may require chondrus. Diffusion-weighted imaging. When do we use and what it is? So this technique provides qualitative and quantitative functional information about the microscopic movements of water at the cellular level. When we have high cellularity with a specific structure we have increased B-value, B-value images signal and decreased ADC values. And when is it useful? So for example in this case nine-year-old girl with a radiolucent lesion in the left femur, in the right femur we can see that there is a long list of differential diagnosis including a pyogenic and non-pyogenic CNO tumors and also UBC. And on you can see here that this case we had the inversion recovery imaging was more helpful than the diffusion-weighted imaging. And in this case here we're seeing two different locations when we do a whole body MRI. So this really lead us to think about being CNO. By comparison I think in this in our case we had I think no restriction and the ADC map was 1.5 times 10 and elevated to minus 3. And usually if you have pyogenic osteomyelitis we expect to see I think in the chronic phase lower ADC map values. In this case of pyogenic osteomyelitis was in acute phase and it was very similar the ADC map values. So in this case the IR was inversion recovery sequences were more helpful than the diffusion-weighted imaging. So now I think for cancer what are the indications? There are multiple clinical indications for death focusing on cancer predisposition disorders. We use in different, there are different protocols that people have. In our institution we only use coronal STIR images. Other groups use also axial images and other sequences including T1 and T2 weighted imaging. And in terms of diffusion- weighted imaging based on the literature probably out of six groups, different groups, only three use diffusion-weighted imaging as part of the protocol of surveillance of pediatric cancer. So this is a meta-analysis conducted by our group and in 477 patients from different institutions. And we try to see what was the diagnostic test performance of different techniques for for whole body MRI for assessment of primary cancer. As you can see here for inversion recovery the diagnostic performance is very high. Sensitivity, pulse sensitivity of 98%, specificity of 98% as well compared to diffusion-weighted imaging which in general is much lower. The sensitivity is 76%, the specificity of very variable with 24%. And PET MRI has very high sensitivity specificity but it's not very used in many centers, especially whole body MRI using PET. And for metastatic disease I think most studies I think have shown very high sensitivity and specificity for whole body MRI as you can see here. So now I'm going to show a different case that the diffusion- weighted imaging was very helpful in the surveillance of pediatric cancer. So this is a patient that was a 10-year-old patient with diagnosis of leukemia and here we can see some lesions that were seen only on the diffusion-weighted imaging the kidneys of the patient. And this patient was also under surveillance with ctDNA which is a new technique for assessment of circulating DNA. So initially when the titers of ctDNA started to increase the whole body MRI was completely normal. And then there was another blood work I think in between two MRIs that it continued to rise and four months after the first MRI we saw complete abnormality of increased signal intensity in the bone marrow of multiple places of the body. So this is an example how the ctDNA together with the whole body MRI can be used for very early diagnosis of cancer especially in patients with cancer predisposition disorders under surveillance. So the problem, one of the problems of whole body MRI in the surveillance process are the false positive results that may cause anxiety in the family. So in this case we evaluated I think this increased signal intensity in the right public brain was multiple times the patient ended up having a biopsy and it was negative. And so this is something to keep in mind. So in terms of management based on whole body MRI I think if ever we don't see abnormal signal not to worry we do annual surveillance in patients with cancer predisposition disorders with whole body MRI. And if we see some areas I think we may need to do a very close follow-up and if there is suspicion of being a tumor biopsy or surgery. And I think these were the learning objectives that I discussed. I hope they were accomplishing this presentation and I would like to thank all the collaborators. Thank you so much. Dr. Ethan Smith from Cincinnati Children's. He is speaking on pitfalls with MR entropy beyond inflammatory bowel disease. All right well thanks Govind. Thanks everybody for sticking around to the very very end. I'm going to be talking about pitfalls with MR entrography beyond inflammatory bowel disease. So my objectives the two things I want to sort of touch on here are inflammatory bowel disease mimics. So things that we might see on MR entrography that can confuse us for active inflammation in the bowel. This includes bowel wall thickening and some other bowel abnormalities and then also some confounding things for perianal disease. And then the second part of the talk will be focused on something that I sort of consider a pitfall. Sometimes we get very obsessed with looking at the bowel and we kind of miss extra intestinal findings. So I want to go over a couple common extra intestinal findings that can be seen in patients with IBD. So the typical bowel wall findings of IBD on an MRI I think we're all pretty familiar with these. We'll often see bowel wall thickening. We'll see hyper enhancement edema in the bowel wall which manifests as increased T2 weighted signal. If we do diffusion we'll see restricted diffusion and then you can see like in these examples here a lot of neural irregularity, ulcers, pseudosacculations. And then if you do do motility imaging you'll also see diminished motility in those affected segments. Specifically with bowel wall thickening is probably the most common finding we're going to see in IBD with active inflammation but we have to remember it's not always specific. There are a lot of other causes of bowel wall thickening which include things like just normal findings, bowel loops that are decompressed for example, technical factors, maybe you didn't get a great oral prep, infection, other types of inflammation and even neoplasms. So one of the most common sort of pitfalls that we see are these sort of what I call normal or technique related abnormalities. So if we have a really nice oral prep like we have in this image on the right then that's nice and easy. Usually the bowel is pretty easy to look at but there can be some confounding things especially in younger kids. Normal jejunal folds can look really prominent even if the bowel loops are well distended but that gets accentuated if the bowel is under distended. Just collapsed bowel loops in general can often be quite difficult to evaluate. And then a lot of patients especially younger kids and teenagers will see prominent lymphoid tissue in the terminal ileum which can kind of confuse us for active inflammation and we'll talk specifically about that. So this is an example we might typically see in practice. This is a 14 year old boy who comes in for his first MR enterography, has chronic abdominal pain and so we have t2 imaging of the coronal axial planes here. And if you look at the bowel loops kind of in the left upper quadrant they kind of look a little bit thick. They're really not all that well distended. Maybe there's a little bit of t2 hyperintensity on that image on the right there but it's a little hard to tell. So just as an aside at least in our practice it's relatively common to have a suboptimal oral prep especially in younger children. So the typical thing you'll be sitting there reading cases technologist calls and says the next patient says they only had one one bottle they were supposed to have two and they can't finish the rest. So what do you do with those patients? In our practice we typically just have them do their best and then scan anyway. There's no other way to really get oral contrast into these patients so it's really the best shot you have at it. So we just say well if they did one bottle that's better than nothing let's go ahead and scan. And then we can use some other techniques we'll talk about here even if the bowels a little bit under distended to try to determine if it's active inflammation. We have found in our practice and there's some data to back this up that switching to Bariso is better tolerated than Volumin. So if you're still using Volumin it's consider potentially switching to Bariso. It's a lot easier for the kids to tolerate. So what do we do if we have these bowel loops that look kind of thick but we're not sure if they're normal or not if it's just technique related? Well in order to distinguish normal bowel from true bowel wall thickening meaning inflammatory bowel wall thickening some of the things I look for is does it persist over time. The nice thing about an MR entorography is you have several imaging time points you can look across unlike compared to a CT entorography for example. So you can look at your first sequence in your last sequence and see if that bowel loop that you were concerned about has actually changed over time which is not uncommon. Sometimes they'll kind of open up you'll see oh that looked funny on the initial t2 and then I'm looking at it later and it looks completely normal. Diffusion is actually extremely helpful especially in the jejunum. There's a good paper recently in Pediatric Radiology out of CHOP about this using diffusion weighted imaging to sort of determine normal jejunum even if it's under distended. What I call true enhancement so that's not just like it looks a little bright but like really seeing enhancement would be something that would convince you that you're probably dealing with active inflammation as opposed to just under distention. Similarly bowel wall edema that means really bright signal in the bowel wall would lead you towards this is probably active inflammation as opposed to other just other under distention and then just other signs of IBD. If there's decreased motility in that segment if there's a lot of fiber fatty proliferation it kind of shows you this patient's had long-standing disease certainly you can use those things to your advantage. So we go back now to our 14 year old chronic abdominal pain and on the left there you see the BTFE image and it's sort of a corresponding t1 post-contrast weighted image where I would argue there's really no enhancement or no abnormal enhancement in those bowel loops that look a little bit thickened in the left upper quadrant. Similarly when we compare a similar level of our t2 weighted imaging non fat set on the top there with the diffusion weighted image you can see there's really no restricted diffusion. So in this case we were able to pretty confidently say that those are just probably under distended jejunal loops this is an active inflammation. So specifically looking at the terminal ileum the terminal ileum has more lymphoid tissue in its wall than than other bowel loops. I think we've all seen this before when we used to do upper GI small bowel follow-throughs and you see that little nodularity in the terminal ileum that could be normal these are Peyer's patches they're just normal lymphoid tissue. But just like in the appendix sometimes we see you can actually get lymphoid hyperplasia in the terminal ileum which can cause bowel wall thickening it's usually mild but it looks like thickened bowel. It can even cause some increased t2 weighted signal and enhancement and may even restrict diffusion. So how do we distinguish this normal sort of lymphoid hyperplasia from Crohn's disease? Well unfortunately it's not always possible. This was an adult study that was done in 2014 where they looked at cases that were biopsy proven Crohn's biopsy proven normal and then biopsy proven lymphoid hyperplasia and they found out of the nine cases of biopsy proven lymphoid hyperplasia that they had their readers when they had gone back and read blindly had actually called it active Crohn's disease with quite a bit of confidence. So that's you know we're not very good at this unfortunately. Typically this paper described it has less severe findings so maybe it's more mild bowel wall thickening. The t2 hypertensive signal may only mildly increased and they're really only mild enhancement but that's really subjective. I would argue that clinical factors can also help quite a bit. For example if you have a patient who has no other findings of inflammatory bowel disease just a little bit of thickening in the terminal ileum maybe a little bit of restricted diffusion. You look in their labs their fetal comprotectum was normal. Chances are you're just looking at lymphoid hyperplasia. Whereas if you have a patient for example who had severe terminal ileal Crohn's disease two years ago has been on treatment now we know this patient's coming back for re-imaging for a reassessment and there's a little bit of thickening there. It's probably related to their inflammatory bowel disease but difficult to tell but clearly their pre-test probability is so much higher that I would probably go ahead and call it in that situation. So now we're going to move on and look at a couple IBD mimics or things that mimic IBD based on causing bowel wall thickening. So this was an eight-year-old male who came into our hospital with acute onset of abdominal pain and diarrhea and you can see on the coronal t2 weighted image the arrows are pointing at the colon which is extremely thickened. On the axial images likewise the arrows are pointing at the colon but notice there's a whole bunch of edema and the mesenterium and this patient looks very sick they've got body wall edema etc and then actually on post contrast imaging we saw little wedge shaped areas of the kidneys that weren't enhancing very well and this actually ended up being an E.coli infection. So again bowel wall thickening but in the different clinical context where this was E.coli not inflammatory bowel disease. Another patient this was a 17 year old female with chronic abdominal pain. They had done colonoscopy which was normal but here we have coronal and axial post-contrast imaging you see there are some mildly thickened bowel loops and left upper quadrant. I would argue they are enhancing quite a bit. They're a little bit under extended so it's a little bit hard to tell but as this patient went further through her clinical workup and had some laboratory testing etc was found that she actually had celiac disease and not inflammatory bowel disease. So this was actually reactive bowel or inflammatory bowel wall thickening from celiac not from inflammatory bowel disease. Another patient this was a ten-year-old female with a history of a renal transplant who came in with abdominal pain. You can see the renal transplant in the right lower quadrant there on that image on the left which is a coronal t2 weighted image. There's this really pretty robust bowel wall thickening but the bowel loops all look kind of like they're almost matted together. You don't see individual bowel loops necessarily but there's clearly a lot of bowel wall thickening and then you look in the axial plane similarly you can see there's a lot of sort of matted kind of thickening of these bowel loops and left upper quadrant. When we do our diffusion weighted imaging with B800 imagery you can see there's just robust restriction of diffusion here. So this ended up actually being lymphoma post-transplant lymphoproliferative disorder given her history of a renal transplant. You know the really the mesenteric thickening and all the matted look to it really isn't something we typically would see with Crohn's although certainly you could if there was penetrating disease. You can see how that confusing but in this case this ended up being lymphoma. And then finally this is a 14 year old male who came in with relatively acute onset of abdominal pain it was sort of acute on chronic and also had a fever. And so the circle there is just pointing out these kind of matted bowel loops in the left upper quadrant where it's really hard to sort of distinguish what's going on there. And then on the image on the right you can see the two peripheral errors that are actually pointing out there's quite a bit of peritoneal thickening here and peritoneal enhancement. So it looks like there's peritonitis. And then that third arrow up in the corner here is actually showing a focally thickened bowel loop with a hole in the wall of it. And we were able to determine that that was colon. And so this was actually gone on went on to colonoscopy and biopsy. This actually ended up being a colonic adenocarcinoma with a focal perforation. So obviously very very uncommon but just something to keep in mind that there are other things that can look quite a bit like penetrating Crohn's disease that aren't penetrating Crohn's disease. So a couple of mimics for perianal disease. So things that can mimic perianal fistula and abscesses include things like pilonidal cysts especially if they get infected. Other types of infections. So you could have a diverticulitis for example that's way down near the rectal sigmoid junction and we have seen diverticulitis in teenagers before. Or maybe now you have a perf diverticulitis that makes it look like you have an inflammatory bowel disease process going on in the lower abdomen. Even skin infections could get enlarged enough that you might think they're actually a perianal fistula or perianal abscess. And kids with really bad chronic constipation solitary rectal ulcer syndrome can sometimes cause things that look like Crohn's disease in the rectum. And then last but not least trauma certainly is something to think about if you have someone with a perianal abscess. It's unexplained especially if it's a younger child. You know you could consider non-accidental trauma or some sort of abuse if there was instrumentation down there and trauma to the rectum that caused a perforation and an abscess. So this is an example of a 16 year old male that showed up with drainage and sort of pain posteriorly. We did our MR enterography protocol but also did a dedicated perianal fistula protocol. So on the image on the left there that's a t2 weighted image you can see there's this focal fluid collection restricts diffusion and has some peripheral enhancement. But when you go to the sagittal images you can see that this is actually located quite high up and posterior to the coccyx. So this was consistent with the pilonidal cyst that was probably infected. Given the location you're probably not going to confuse it too much with perianal Crohn's disease but certainly something to think about. All right so moving on for the last minute and a half or so we'll talk about some extra luminal manifestations. So it's important to remember that patients with Crohn's disease both ulcerative colitis and or excuse me inflammatory bowel disease both Crohn's disease and ulcerative colitis are at risk for a lot of extra intestinal findings as well. In the hepatobiliary system, obviously primary sclerosing cholangitis is something we're really gonna scrutinize for, but they're also at risk for gallstones and gallstone-related complications. In terms of pancreatic abnormalities, they're at risk for pancreatitis in general, but specifically autoimmune pancreatitis. In the urinary tract, they're at increased risk for renal and ureteral calculi, and obviously musculoskeletal, things like sacroiliitis. So this is just an example of a seven-year-old girl with pancolitis from ulcerative colitis that had dilated bile ducts consistent with PSC. And the right-sided bile ducts are maybe a little bit dilated, don't look that bad, but hopefully you can see on that MIP image there on the right that the left-sided ducts are beaded and a little bit irregular. And this is enough to call this PSC in this patient with really bad ulcerative colitis. Just a different patient now, a seven-year-old who had severe Crohn's disease, actually had already had an ileocecal resection, and just showing some filling defects on the gallbladder consistent with gallstones. The thought is they probably are at higher risk for gallstones because you interrupt this sort of normal bilirubin circulation as you take out the terminal ileum, potentially, or there's inflammation in the terminal ileum, which can increase their risk. Pancreas, these are two different patients with autoimmune pancreatitis. So on the image on the left, it's a T2-weighted image. You can see that kind of sausage-like enlargement of the pancreas. There's potentially a peripheral rim of T2 hyper-intense signal, which can be a finding associated with autoimmune pancreatitis. And then on the image on the right, it's a separate patient. You can just see that very large, kind of sausage-like enlargement of the pancreas on this T1 post-contrast image. And then finally, musculoskeletal. I always tell our trainees to scrutinize the SI joints. You have T2-weighted fat-saturated imaging, and you have post-contrast imaging, so certainly you can evaluate these sacroiliac joints and see if there are changes of sacroiliitis. So in summary, the pitfalls that we discussed. Probably the most common one is just under distention and normal bowel folds. It's frequently technical, but not much you can do about it. And to sort of problem-solve for that, you can look at multiple time points across your examination. And then also diffusion tends to be very helpful, especially in the jejunum. Other causes of bowel wall thickening, infection and inflammation, neoplasm, frequently the clinical history and labs may be helpful to help you distinguish between those. And then don't forget to look for extra-luminal findings and look beyond the bowel wall for known complications of Crohn's that occur outside of the bowel. All right, and thank you. So next up is myself, and I'm going to speak on rapid MRI on the body side in children. So when I got this topic, I thought I'll collect some cases and do the pictorial of uses of the rapid abdominal MRI. But we still haven't rolled our clinical rapid equidermal. So I'm going to do this as a concept of rapid MRI in children. So my objectives are to discuss how rapid is rapid MRI. Why do we need rapid MRI? What are the requirements for it? And I'll discuss rapid MRI, abdominal MRI as a rapid MRI as a prototype protocol as a concept to the rapid MRI. And then in the end, we'll discuss some other potential areas where rapid MRI can be applied. So how rapid is rapid MRI? Basically, when you compare any speed, you compare with something else. And obvious modalities seem to be ultrasound, X-ray and CT. And if MRI goes close to that, probably you can call it rapid, but there's no strict definition for it. But the other attributes with which the rapid MRI is typically these MRIs use anywhere between three to five sequences, and they are 10 to 15 minutes long, and usually less than 15 minutes all the time. So what have we done? But as the time goes and the advances in technology happen and overall the community of images get used to it, I think the time is going to go down. Now why do we need rapid MRI? We need it to reduce the time, improve the patient comfort, reduce image interpretation time for the radiologist, and that in turn will improve the MRI workflow, reduce cost, meet the increasing clinical demands for the MRI. And also recently there has been emphasis on patient-centered care and radiology and development of efficient and precise protocols. So rapid MRI kind of fits with that aspect of the radiology as well. So what are the requirements for the rapid MRI? The main rationale behind the rapid MRI is to reduce the number of sequences as compared to the standard protocols. So usually you reduce the redundant sequences without compromising the diagnostic information, and that has been the main basis of rapid MRI. But if you look overall at the rapid MRI, there are many other things you can do to make it faster, and especially in children, if you eliminate the need for sedation, need for contrast injection, need for scan check, or you eliminate the need for the longer motion-sensitive sequences, then also you can make this MRI much faster as compared to the standard MRI. So how do we do that? For eliminating the need for sedation, you can use the distractions like goggles and movies. If available, you can use the child lab services and use the feed and sleep. For eliminating the need for contrast injection, it is a tricky part, but you need to choose the right indications for rapid MRI because you have to be careful with that. You cannot use a rapid MRI with all the indications. And for the eliminating need for scan check, you need to build the robust balance protocol and educate your technologists properly. And once they go through that learning curve and more comfortable, then you can probably avoid the scan check completely. And for having the motion robust, faster sequences, there have been recent advances on the MR technology side. So for the T1-weighted imaging, there have been approaches to reduce the motion artifacts by using the radial case-phase filling. For improving the temporal resolution or speed of the sequences, there are many techniques are used in the sequences like partial case-phase filling, view sharing, or parallel imaging. Or there have been approaches which combine both of these factors and get the advantage of both, like increase the speed and reduce the motion. And sequences like 4D fast breathing from Philips or graphs like sequence from Siemens are doing that. And essentially, these sequences will enable us to have the free breathing dynamic protocol in the future. They are in early stages now. For the T2-weighted imaging, similar approaches. Again, use the radial case-phase filling. The example is propeller, blade, or multivane sequence. It is quite robust motion-wise, but doesn't reduce the time at all. It, in fact, can take more time than the standard sequences. The other approach is using the 3D version of the T2-weighted imaging. And the attraction is that you acquire image in one single plane of volumetric data and then construct the other two planes retrospectively. But again, this sequence is there since almost a decade or more now with all three vendors. But the reconstruction are not that up to the mark, so still not up to the level of diagnostic quality. And then there is a synthetic MRI, which is basically you acquire a single sequence of four to five minutes and get the different weightings in the same sequence retrospectively, like T1-weighted imaging, T2-weighted imaging, flare, and even diffusion nowadays. And it is mainly being used in the neuro side currently. But if it becomes robust in the technique, that could be one of the approaches for the faster MRI, just acquire one sequence and get all the weighting in the same acquisition. This is a study which came from Phoenix Children's Hospital where they compared the radial approach T1 and T2-weighted images with the standard sequences in infants and neonates. And they found that the quality was comparable and it reduced the need for sedation and intubation in these babies. So this is quite early and all these approaches are still in early stage, but in the future they will help in making the MRI protocol much rapid. So I'll discuss the rapid abdominal MRI as a prototype for all these rapid MRI on the body side. So it basically started in adults, in pregnant women, to detect the appendicitis because you cannot have a CT scan. And from there, this concept of rapid abdominal MRI and especially rapid MRI for the acute abdominal started and eventually passed to the other areas and in pediatrics. And since then, it has consistently shown that the sensitivity and specificity of these rapid MRIs is in the range of 95% in both pediatric and adult population. Not only it is good in the detection of the appendicitis, it also diagnoses alternative diagnosis in the abdominal pain in 20 to 37% cases. For example, this is a patient who presented to us with right lower coronal pain and this rapid acute abdominal MRI was performed. And what we found was there was a omental infarct which was diagnosed non-invasively and she was treated conservatively. So what are the attributes of the rapid abdominal MRI? And does it fit with the qualities we discussed earlier with the rapid MRI? So typically so far in pediatric population, the reported studies have used three to five sequences, mainly the detection sequences like we do at imaging. They all have been below 15 minutes time. And if you get used to it, the time reduces as shown in Cincinnati children they have. They are well under 10 minutes and some of them they are saying between six to 10 minutes they are doing the rapid MRI for the appendicitis. All of them do not use any sedation and they are mostly free breathing, no breath holds or in respiratory triggering sequences. And that's why they have been used in children above five years. And how commonly it is used in the pediatric population? This is a survey which was done by SPR in 2020 and of those who responded, 38% said that they use rapid MRI in acute abdominal sitting. And based on this survey which also asked other questions like what sequences they use, what services, availability is there and et cetera. So based on that survey, literature review and the consensus discussion between the two committees of the SPR, this consensus protocol was developed which basically recommends these four basic sequences for the acute abdominal protocol which include axial and coronal single shot or HAST. Then axial HAST single shot with a FATSAT and axial diffusion. So those four core sequences are essential for the acute abdominal protocol. This statement will be published soon in the Pediatric Edology. This is, we are also rolling our rapid abdominal MR protocol and Andrea is leading our initiative. She started this with her research study and gradually we are trying to offer our clinical service. This has a clinical study. So we are having some different aspects to it. So we are trying to roll it not just as acute appendicitis protocol but as acute abdominal protocol. And as we saw that there are diagnoses which are alternative to the appendicitis in 20 to 40% cases. So it is important to roll it as a more robust acute abdominal protocol rather than just appendicitis protocol. And it is easier for the team to get that robust protocol and roll over. And it's also good for the patient flow and patient diagnosis that they get faster diagnosis. The system moves much faster. So for that reason, some of the things we are covering are our coverages from diaphragm to pubic symphysis. And apart from those four sequences we had in the consensus protocol, we are adding one more T1 weighted imaging in Dixon or in and out phase imaging. This is a very fast sequence, 20 to 30 seconds. But it provides a lot of information about that short time because it gives a T1 contrast to diagnose the fat like we see in this case in the dermoid with ovarian torsion. Or it provides the signal intensity for the hemorrhage. It is T1 hyper intense as we commonly see in the hemorrhagic ovarian cyst which is a common differential for ovarian torsion and both lower cord and pains. And in some other studies they have shown that in and out phase imaging helps in some cases to localize the appendix because sometimes on MR it is very difficult to localize the appendix. And with in and out phase imaging if the normal appendix contains air, it will bloom on the in phase imaging and that has been shown in other studies. So it is a very useful sequence just for 20 to 30 seconds in your protocol. The other areas where it has been reported useful is this is a study from St. Louis, Malincroft where they use this three sequence protocol including single shot T2 stir and diffusion in perforated appendicitis. And they use this protocol to try the patient to differentiate between those having phlegmon needing the medical treatment and those which can be discharged versus those having the frank abscess which needs the admission and drainage. So a very short protocol was useful in that study. The other areas where outside the abdomen where the rapid MRI has been reported in pediatric population is rapid MSK infection protocol for osteomyelitis and this is routinely used in Lurie Children's and Emory Atlanta. They use these three or four sequence protocol and they're using it regularly in the clinical practice. On a neuro side, there are reports of rapid brain MRI for detection or assessment of hydrocephalus and rapid spine MRI for syringomyelia, CRE and position of the ponus. So in pediatric population, it is still limited to the detection of the pathology or kind of just seeing where is the abnormality. But in adults, the abbreviated or rapid protocol goes one step further. They're using it for the characterization and the examples are SCC screening or prostate cancer screening where they just use two sequences. So for SCC, they inject the hepatocyte contrast outside the scanner and at 20 minutes, they acquire two sequences to detect any new nodule which could be SCC and in prostate cancer, they just use a non-contrast axial T2 and diffusion to detect the prostate cancer. And there are also other examples like fatty liver protocol which includes the elastography. So what are the challenges in performing rapid MRI in children? So we have age group from six months to four or five or six years where in most children, we need anesthesia or sedation and in that group, probably we cannot apply the rapid MRI because once you put a child in anesthesia, you cannot get over the short protocol. If anything is needed, then the repeat anesthesia is not a good idea. So that age group probably will be difficult to apply this rapid MRI or difficult to justify the rapid MRI in that age group. Not all conditions can be assessed that rapidly so you have to choose the indication carefully. And despite advances, battling motion in body imaging still a challenge. Also, on the body side especially, the number of body MRI in children are significantly less as compared to the neuro MRI or the adult MRIs. So it may not have that impact on the workflow or cost saving in terms of rapid MRI in this particular population. So in summary, rapid or abbreviated MR protocols are coming. They will carve their roles in clinical care in children and currently in children, they are used for acute abdomen osteomyelitis, hadrosephalus, rapid spine protocols. And this is just a random thought. So neonatal NICU MRI, MRI scanner itself in the NICU unit is at very few places. One of them is Cincinnati. So in case in future they become widespread, you may not need to send that unstable baby to the fluoroscopy unit for half an hour. You can just do the two sequence like this actual proof is for actual to do. It shows SMH in relationship, D3 going behind it and also an entire duodenal CP is nicely seen here. And just two sequences, five minutes, you'll have the diagnosis of normal versus the malrotation or any other complications. So this may be the future of the rapid MRI. And thank you very much for your attention. So with that, it is my great pleasure to introduce last speaker, Dr. Sherwin Chan from Children's Mercy Hospital, Kansas. He's talking on template reporting. I'm Sherwin Chan, I'm a pediatric radiologist at Children's Mercy, Kansas City. And today I'll be talking to you about structured reporting. I did wanna specifically acknowledge some friends and colleagues, Dr. Jonathan Chung from UCSD, Dr. Chris Walker from KU Medical Center and Dr. Aya Kamiya from Stanford who were instrumental in helping me put this talk together. And I'm specifically gonna highlight research that was done by Dr. Chung and Dr. Kamiya. In summary, I wanted to talk about structured reports and my take home points are that they're preferred by clinicians and patients and I'll present evidence for this. This disease specific templates such as those in CNO can really aid in quality improvement and also help remind the radiologist to include all of the ancillary findings in the dictation. And then finally, I'll touch on the topic of this RSNA which is how to lead through change. So before we talk about the benefits of structured reports, I wanted to review the different types of radiology reports out there that you'll see in practice. Most reports may be broken up into three different types, individual templates or PROS, structured reporting or combination of structured reporting and PROS. So for these early examples, I'm gonna pick a simple report, chest radiograph report to make my points. This is an example of a chest radiograph finding section using PROS as the reporting style. PROS is loosely grouped information often based on radiologist search pattern. In a large group of radiologists, you often have considerable variation in the reports and reporting style. Note that in this report, actually they represent all of the information. So they talk about the central venous catheter, they talk about the bones, the lungs, the pleura, the mediastinum and the heart. If we look at this PROS report on the same case, the information is really similar but it's presented in a different order. Also note that here the radiologist didn't talk about those osteostructures. In this last example, we can see the radiologist decided to combine everything, the findings and the impression into one section and simply stated negative chest. This was common back in the day and hopefully everyone can appreciate that if a clinician's looking at it, it may be difficult or take more time to find what they're interested in looking at, especially if they're looking for information about the lungs or pleura or potentially the mediastinum. So let's contrast that with a form of reporting called the structured report. This is a structured report from the same case. You can see that each section, including the support devices, lungs, pleura, heart and mediastinum and osteostructures have their own heading. It's also standardized across the radiologists in the group. Structured reporting is particularly helpful for junior radiologists because it helps them establish a search pattern and we do have to remember that ultimately this is our work product. This is what we're creating for our referring clinicians and our patients and so we have to ask, what do the patients actually prefer? What do the providers prefer? Luckily, we have evidence on this. So there's a large study that looked at reporting styles from the patient perspective. So over 5,000 patients reviewed two identical chest radiograph reports. One presented in prose and the other was structured reporting. They basically found that structured reporting was easier to comprehend and the patient felt less likely to take unnecessary action in the event of a negative report. So this is great because we want our communication styles to be clear. And there's actually been several studies looking at how clinicians perceive different radiology reporting styles. So this study was performed at a large academic center in Ohio and the study divided clinicians into three different groups. A novice group consisting of medical students and interns, an intermediate group which consisted of PGY2 to 4 trainees, as well as mid-levels and advanced practice providers and an advanced group that consisted of PGY5 trainees and above as well as all the attendings. Again, each participant in the study reviewed two radiology reports containing identical information. You can see that the one on the left used structured reporting and the report on the right used prose or freestyle. You can see the information in both reports is identical. The only difference is that the structured report contains section headings such as mediastinum and bone. Importantly, the impression is actually identical between the two reporting styles. So what did the results show? So they first assess which reporting style was easier to read by the clinician. And basically you can see that this structured report had better readability or similar readability compared with the prose or paragraph style reporting. Next, they looked at the clinical utility of the structured report compared to prose. And again, the structured report was perceived as equal or better in the vast majority of cases. Also note that the results were really similar when looking across the different groups of clinicians involved in the study, ranging all the way from medical student to attending. So next, let's talk about disease-specific reporting. We are in an MR session. So with disease-specific reporting, you tailor your structured report to the condition you're reading to ensure the completeness of the report. This is especially helpful as it reminds radiologists to discuss the important information pertaining to that disease that might be forgotten in a busy day, such as including all the image-defined risk factors in a new diagnosis of neuroblastoma. Disease-specific recording may be used in a wide variety of cases, ranging from neuroblastoma staging to whole-body MRI for CNO or cancer predisposition syndromes. Here's an example of fields that could be incorporated into a structured report for CNO. In the structured report, the overall study quality is described, and I'll show you why this is important later. And then clinically relevant characteristics of each lesion are described. So you can see that the lesion location is described and then different lesion characteristics about where they are in the bone and different descriptive characteristics are here. This information is from a multicenter imaging registry for CNO. It was performed actually by a really talented medical student, or sorry, undergraduate student at the University of Toronto. And this registry has aggregated reports and images from four different children's hospitals in North America. So this is the data that I want to present from the study, and it's around how structured reporting actually makes the reporting of lesion characteristics more complete so that our clinicians can make the correct clinical decisions. The gray bar represents incomplete reporting. So you can see this is two of the four hospitals that are represented in our study, and there's a huge difference in the reporting of the lesion characteristics. Hospital one, which included lesion characteristics as part of their structured template, was 100% in reporting all of the lesions. Hospital two also used structured reporting, but didn't actually include lesion characteristics as one of their fillable fields. And you can see there's a lot of missing information for the hospital two. Another example is where a hospital actually changed their structured reporting practice. So you can see on this hospital, they looked at PECT reports, and they changed from a non-structured report to a structured report. And you can see that there was a baseline of 69% of those reports being complete. And then 92% after the change was made. Something else that this PE structured template allowed them to do was actually track PE data longitudinally from month to month, and a tracker in the PE tracker, which is an IT tool that they implemented. So one of the variables they actually tracked was the indeterminate rate of PE studies. So studies that were technically inadequate for PE evaluation. The rates typically oscillated from one to 3% every month. However, since they were able to track this, they noted that there was a monthly analysis of that PE data, and there was a huge uptick in indeterminate rate for about four to 6% over a several month period. And that was a standard deviation above what they had observed before. So looking at that change in the non-diagnostic rate, what they did is they implemented a PowerPoint-based education system for their technologists. They standardized the CTPE protocol across their whole system. And then they had a subsequent improvement in their non-diagnostic rate back to two to 3%. So basically, by ensuring that each exam had a discussion on the quality of exam, which is why I think this is important, using the structured template, they were able to quickly detect a change and then implement an improvement. So if they'd been using a prose-style report or they hadn't included this field in their structured report, this information wouldn't have been available to them. Another example from the literature, which I like, and this one is more important to how do you do structured reporting across multiple institutions and why is that important? So structured reports are actually key to enabling large retrospective research and quality studies. For example, the adoption of a standard ultrasound LIRADS criteria allowed Dr. Fetzer at UT Southwestern to show that poor visualization of the liver was actually highly correlated to sonographer experience, which we would expect, and patient location, which they actually didn't expect from their initial analysis. This actually allowed revision of the new ultrasound LIRADS practice guidelines at multiple centers to discourage ultrasound LIRADS in certain settings, and it emphasized the need for sonographer training. Another example from the ultrasound LIRADS criteria is initially the management of LIRADS to ultrasound was continued close surveillance. However, this work actually suggests that the initial management plan was probably a little overaggressive and a little overzealous, so future clinical guidelines will reflect this evidence and allow for less aggressive follow-up. Again, this is all enabled because there's so much structured reporting in this field. Finally, structured reports actually enable large multi-center studies, and I would argue this is the exact type of study that we need to change practice in pediatric settings because we all have such small patient populations. So these structured reports allow studies to be performed with little to no funding. For example, this adult study reported on 2,050 patients over only a nine-month period from five large academic centers, and this study is helping to inform the new generation of the ultrasound LIRADS criteria. So again, successful adaptation of the structured reporting has been instrumental in the fast iteration of clinical guidelines and the subsequent large improvements in clinical care. So hopefully now you're all convinced of the benefits of structured reporting. So finally, let's touch on the theme of this RSNA conference, actually leading through change. So you've probably heard many talks on this before, but I wanted to briefly give you my two cents. So I think change management's hard. It's not something that's commonly taught during our training. If you're listening to this talk and you feel the need to implement structured reporting at the institution, realize that this is the distribution of people that you'll likely be dealing with. And if you look, a reasonable measure of success would be to get that fat middle, so the majority on board with your plans. Again, there's lots of different change models that are out there. This is the one that's preferred at my hospital. So this is one of the models we use for change management. You can see that it really emphasized engaging with all of our stakeholders who will be affected by the change. So they're going to move along these different stages of change management. And so the stages and the nature of the engagement will actually change over time. And your message really should be tailored to where your organization and where individual people are along these stages. Additionally, I did want to highlight that bottom stage, the reinforcement, because that's probably going to be the longest stage for you in your change management. And it's critical to creating enduring change and critical to changing culture in the organization. So again, in summary, the structured reports are really preferred by our clinicians and patients. I've presented evidence to convince you of that. Disease-specific templates, aiding QI, and also help remind radiologists. And then finally, please be really intentional as you lead your group through any changes such as structured reporting. And thanks for staying for the last talk of the last session of the last day of RSNA. And please feel free to contact me with any questions.

Pediatric MRI

whole body MRI

chronic non-bacterial myelitis

MR enterography

inflammatory bowel disease

rapid MRI

structured reporting

diagnostic quality

patient care