So my topic is marbles in the chest, differential diagnosis of solid lung masses. So we're gonna move on to first clinical scenarios with solid lung nodules or marbles in the child. So the first scenario is a child who has history of cancer, presents with clear lungs in the beginning, but then develops nodules, and that's typically a feature of pulmonary metastatic disease. The second scenario is of a child with fever with cavitating nodules at presentation, and so that's suggestive of septic emboli. And finally, a patient presenting with subacute or chronic symptoms with multisystem involvement concerning for pulmonary vasculitis. So the objective of this talk is to discuss the differential diagnosis of multiple solid pulmonary nodules. We'll discuss a variety of the marbles, the bad, the changing, the hidden, and the pseudomarbles. It's gonna be a stepwise, case-based approach. I'm gonna highlight the typical imaging features, cavitation, calcification, the halo sign, and also touch base briefly on the incidental pulmonary nodule in children and where we currently stand. And finally, there's gonna be a post-test with two multiple-choice questions using Poll Everywhere. All right, so as far as definitions go, according to the Fleischner Society, a mass is defined as an opacity greater than three centimeters, and a nodule less than three centimeters and I would be pretty much focusing within this category. A pseudonodule is described as an opacity that mimics a pulmonary nodule and could be due to a rib fracture, skin lesion, device on the surface of the patient, as well as the patient's anatomy. So moving on to differential diagnosis of pulmonary nodules. So in the congenital category, we have bronchopulmonary area malformation and sequestration. However, these typically present as single lung masses, so I would not be discussing that. Similarly, in the neoplastic category, primary lung neoplasms including pleuropulmonary blastoma and carcinoid typically present as single lung masses. I would not be discussing that either. What I would be discussing is neoplastic lesions including pulmonary metastases. And in pediatrics, typically that occurs from Wilms, Evings, and osteosarcoma. And just a reminder, neuroblastoma metastases typically do not go to the lungs. In the infectious category, we'll have normal and immunocompromised patients with multifocal pneumonia with either bacterial, fungal, or septic MLI. In the vascular category, pulmonary AV malformation. And systemic or inflammatory disorders, I think are by far the most difficult as far as I'm concerned and cover cases including hypersensitivity, pneumonitis, vasculitis, rheumatoid arthritis, granulomatosis, and polyangiitis, regenerative rheumatosis, and Crohn's disease. And finally, a few cases showing mimics of chest wall lesions and pulmonary nodules. All right, so. All right, the first category of infection. So here we have a six-year-old with fever, productive cough, and increased white count. On the chest x-ray, we see multifocal round mass-like opacities. There is no pleural fusion. And this is typical for classic multifocal bacterial pneumonia. So patient's symptoms progressed and therefore underwent a I plus CT of the chest. We see consolidation bilaterally with a bone program. There's decreased enhancement concerning for necrosis. We don't see evidence for cavitation but suggest lung abscess formation. But incidentally, we did find a cyst next to the G-junction which was suggestive of esophageal duplication cyst. And then since the patient's symptoms did not get better, got an ultrasound of the chest because of evolving pleural fusions. And here we see bilateral pleural fusions with septations that feature suggestive of empyema. So therefore, this patient underwent baths and decortication. And now the patient now has multifocal bacterial pneumonia with complications. All right, so multifocal bacterial pneumonia in an immune-competent host. The most common background is of a child, a young child with severe viral infection with superimposed bacterial infection. Culture's positive only in 40%. And most of the time, it's due to MRSA, E. coli, Klebsiella pseudomonas. And these patients, upon recovery, most often undergo an immune workup to rule out underlying immune disorders like common variable immune deficiency. Prognosis is typically good because of continued lung growth and remodeling. All right, so we're gonna move on to the next case. And here we have a 14-year-old with severe oropharyngeal infection, fever, cough, and chest pain. So if we see the first chest radiograph on day one, we see a right lower zone airspace opacity, suggestive of right lower lobe pneumonia. On day four, symptoms got bad, and the blood cultures came out positive for MRSA. And chest X-ray demonstrated multiple nodules with central leucency, as well as we have a right side peripheral effusion. And therefore, the patient underwent a CT of the chest. And now we see multiple peripheral nodules with and without cavitation. We see the classic feeding vessel sign, which suggests that the vessel is leading directly to the nodule. Highly suggestive of hematogenous dissemination. So this is a typical scenario for septic emboli. So septic emboli is a changing marble. It occurs due to bacteremia by pulmonary artery. Dissemination either in a setting of endocarditis, infected central lines, periodontal disease. Most common organisms include Klebsiella, Staph aureus, or Pseudomonas. Fusobacterium is typically associated with Le Maire syndrome, with IV septic thrombophorbitis can also present with lung nodules. So what are the differential diagnoses of cavitary nodules? So the most common actually turns out to be metastases with squamous cell carcinoma in adults. I'm just talking about the general population, but this does not occur in children. So we typically don't consider metastatic disease in children. Vasculitis, vaginosis is common, and then as well as necrobiotic nodules, either due to rheumatoid arthritis or IBD. So we're gonna move on to a nine-year-old with ALL relapse, with fever, cough, and neutropenia. And here again, we see multiple nodules with ground glass halo. This is a typical halo sign due to hemorrhage from angioinvasion and infarction. And in addition, we also see tiny enhancement nodules in the brain concerning for cerebritis. So in this particular setting, you're really worried about invasive aspergillus infection. Angioinvasive aspergillosis occurs in profound immune suppression in the setting of post-chemotherapy, stem cell transportation, and about half of them can have systemic hematogenous dissemination to solid organs like kidneys and GI and CNS. It's actually halo sign with fungal infections not specific for aspergillus infection but it can occur with candidiasis and cryptococcus infection as well. So we're gonna go over a few scenarios with pneumonias and neutropenia and looking at classic imaging features. For example, here we have a patient who has lobar consolidation in the left lower lobe and this patient was positive for pseudomonas infection compared to this patient where we see multiple nodules without halo sign and the differential for this should be viral infection, either CMV or RSV, in neutropenia compared to this one where we see diffused glandular sapacities with a few nodules and this is suggestive of PJP pneumonia. All right, so next case is an eight-year-old with leukemia with dry cough and shortness of breath for two weeks. So on the chest radiograph, you see multiple reticular nodular glandular sapacities. There's no pleural effusion. There's no lymphadenopathy. The patient underwent a bronchial alveolar lavage and then the diagnosis was consistent with pneumocystis jarvaki pneumonia. So PJP pneumonia is an atypical organism. It's a yeast-like fungus and it's seen only in immune-compromised host, either due to HIV or malignancy, and treatment is Bactrim and steroids. All right, so we're gonna change gears. We're gonna move on to the neoplastic category. This was a case we soon encountered in our institution, a 14-year-old with shortness of breath, weight loss, no significant past medical history. And if you look at the chest x-ray and the CG scan, there are multiple pulmonary nodules and masses. So when you see something like that, it just says cancer, cancer, cancer, really the bad marble, unfortunately, in this patient. We also saw right ventricular irregular thickening and a mass-like enhancement, so we were concerned about ventricular invasion. So the differential at that time was primary cardiac neoplasm, either it's a sarcoma versus cardiopulmonary metastatic disease. Due to involvement of the heart, the patient underwent MRI, and here we confirmed cardiac involvement, we see pericardial diffusion, as well as pulmonary nodules, and therefore, and the patient underwent a biopsy, which was consistent with metastatic sarcoma without known primary. So when you see metastatic disease in children or the bad marbles, you think about either primary tumors, Wilms, Rabdu, Ewing's, or osteosarcoma. Again, neuroblastoma typically does not metastasize to the lung. And in this setting, we had to consider cancer predisposition syndromes or germline mutation, common of which is Leigh-Fraumeni syndrome, which is a 75% risk of cancer risk. Okay, now we're gonna move on to the VASA category. We have a 16-year-old with epistaxis, shortness of breath, and exercise intolerance. Patient underwent a bubble echo, and after four cardiac cycles, was positive. That is consistent with an extracardiac shunt. Patient also had a family history of HHT, hereditary hemorrhagic telangiectasia, and a chest X-ray was done, which is totally normal, hence the hidden marble. C2PA was done to rule out pulmonary AVMs. So C2PA in this patient suggested multiple enhancing nodules consistent with the AVM nidus. The feeding artery is typically smaller in size compared to the draining vein, which is larger in size, and so this is a feature of pulmonary AVMs. All right, so it turns out if you have dual NGCT, it's way more sensitive for vascular enhancement because it has the ability to create pulmonary blood volume maps. So here there's an example for another patient who has a left upper lobe AVM that's surrounding ground glass. Opacities demonstrating increased iodine distribution suggestive of vascular hyperplasia and not hemorrhage. So why is this more important or better than single NGCT is because there's better size estimation, particularly if you have any intervention done, like coil embolization. So pulmonary AVM in children, there's a very high association with HHT on imaging. We see the AVM nidus, which bypasses the capillary bed. And ground glass opacities in this setting should always consider microscopic telangiectasia. And intervention is considered when the feeding artery is greater than three millimeters or the patient is symptomatic. We're gonna move on to the inflammatory category. And I think this is by far the most difficult category. And here we have a 10-year-old with few weeks of wheeze, nonproductive cough, and shortness of breath. The patient lives in a farm, has many animals and pets. Chest radiograph reveals multiple bilateral reticulonodular opacities. We don't see lobar consolidation. There's no pleural diffusion. A few of the tests were done. The white count was normal. PFTs demonstrated decreased diffusion capacities. Patient underwent a CT scan of the chest, demonstrating central globular nodules, as well as scattered ground glass opacities. And the consolation of these findings suggests hypersensitivity pneumonitis. So hypersensitivity pneumonitis is a immune-mediated inhalation-diffused lung disease. Antigens typically less than five microns in size and get deposited in the alveoli, common of which are fungi like aspergillus. On pathology, you see small airway inflammation as well as fibrosis. As far as classification goes, there are two phenotypes. In children's, we see the non-fibrotic form, where we see nodules, ground glass trapping, air trapping, as well as ground glass nodules. However, if this remains uncontrolled, it can end up into end-stage lung disease, typically in adults. All right. So we're gonna move on to the next case of 11-year-old with six months of low-grade febrile anorexia, weight loss, and epistaxis. So chest radiograph reveals multiple nodular opacities. CT scan shows multifocal airspace disease. There's dense peripheral consolidation. The central, within the consolidation, you'll see a central ground glass opacities. Had renal dysfunction, so underwent a CT scan of the chest, of the kidneys, sorry, and then it's consistent with medical renal disease. And going back on the CT scan, with dense peripheral consolidation and central ground glass opacity, that is the reverse halo sign. And on PFTs, there was decreased diffusion capacity. So these findings are consistent with pulmonary vasculitis. All right, patient underwent treatment with steroids, and now we see a chest radiograph that is normal. And therefore, those opacities were most considered with pulmonary hemorrhage due to underlying vasculitis. So renal biopsy done was suggestive of ankyl-associated vasculitis. And on pathology, you see necrotizing inflammation to small and medium vessels. And within this category, you can see Begner's granulomatosis, Choke-Strauss syndrome, which is eosinophilic granulomatosis and polyangiitis, as well as microscopic angitis. On imaging, again, really a range of findings. It's nonspecific. You see cavitary and non-cavitary nodules, airspace disease, ground glass consolidation, and the reverse halocyte, as I showed you, as well as interstitial thickening, which follows the axial interstitium. So again, the most important thing in this category is to really kind of look at the clinical context and presentation. Anything indolent with multi-system involvement, you should always think about pulmonary vasculitis. So reverse halocyte really described in cryptogenic organized pneumonia, but turns out it's really nonspecific and can be seen with invasive fungal infections, sarcoidosis, as well as pulmonary infarction from post-radiation therapy, and RF ablation of pulmonary malignancies. Here we have a 16-year-old who has Crohn's disease with a few weeks of nonproductive cough and shortness of breath. We see faint nodules of chest x-ray. At that time, we were worried about pneumonia, fungal pneumonia, as well as metastases, because three to 10% of patients with Crohn's disease can have, do have a high incidence of malignancies. So the patient underwent a CD scan of the chest. We see multiple nodules with mild peripheral rim enhancement, no calcification, no halo sign, no lymphadenopathy. Again, our concern was metastasis versus fungal pneumonia. An event, FDG PET, and the nodules were avid for FDG biopsy done consistent with granulomas, necrosis, and giant cell. So this is necrobiotic nodules in a setting of Crohn's disease. So Crohn's disease with necrobiotic nodules is actually a very rare form of pulmonary involvement in IBD, and between Crohn's disease and IBD, Crohn's disease and ulcerative colitis. So ulcerative colitis, a high chance of extraintestinal pulmonary involvement in the Crohn's disease. And if you do have lung involvement, the most common manifestation is interstitial lung disease in the form of bronchiectasis, organizing pneumonia, and necrobiotic nodules actually seen in less than 6%. And so why is this connection within GI and the respiratory system is because of shared embryology with the primitive foregut. So again, cancer incidence in a setting of Crohn's is three to 10%. Biopsy is gold standard, and treatment is steroids. Hidden marble, this is a two-year-old with opsoclonus myoclonus. When you see that clinical manifestation, you are hunting for an occult neuroblastoma. Patients' urine catecholamines were high. Chest X-ray, abdominal ultrasound were totally normal. Chest X-ray demonstrated paraspinal subclerode nodules in the lung apex. This finding is consistent with neuroblastoma. All right, so a few examples of pseudonodules, chest wall lesions. Here we have a patient, two patients of neurofibroma NF1 where we see multiple nodules on the chest wall, which can present as multiple primary nodules within the lungs. So think about chest wall disease as well. And here we have a left upper lobe nodule in this patient. Our CT scan is most consistent with an osteochondroma. So you can have pseudomarbles as well. So a few lines just kind of touching base of incidental primary nodules beyond the scope of this talk. But how do we manage patients who have incidental primary nodule in children? It's a hot topic. Pediatric-specific guidelines don't exist. In 2017, we had the revised Fleishner Society Guidelines, which stated that these should not apply to pediatrics and case-by-case approach as lung cancer is low in children. As far as size goes, there is no size criteria. However, Lee et al. did a study demonstrating that a child with a history of cancer and a nodule more than seven millimeters highly suggestive of pulmonary endostatic disease. As far as features goes, peripheral nodules are typically consistent with intrapulmonary lymph nodes. And if you see that, you should totally not follow them up and be very confident in your interpretation. In 2015, SPR Thoracic Society's effort was to develop evidence-based recommendations for incidental primary nodules. And they suggested, again, that asymptomatic children, the classic benign features, like fat, popcorn, calcification, should not undergo any sort of follow-up. And malignant lung nodules in a child are most likely metastatic disease and primary lung cancer. All right, so the take-home message is use a case-by-case approach when you encounter a child with multiple solid lung nodules. Look at the classic features. Sometimes it's very obvious. It could be a pulmonary AVM. Halosign in a febrile neutroplaning patient suggests invasive aspergillus infection. Remember, if the cause is indolent and prolonged, think about atypical infection vasculitis. And biopsy is occasionally needed. For example, in the case that I showed you, Crohn's with necrobiotic nodules. And then, finally, incidental primary nodule in an asymptomatic child without history of cancer and benign features, no need to follow up. All right, so we're transitioning from marbles to bubbles in the chest. I'm going to keep this as focused as I can on three different broad categories, congenital intrinsic lung disease and spontaneous iatrogenic causes of cystic lesions in pediatric patients. And I'm gonna further target that to the perinatal time period. The first that we're gonna talk about is CPAMs or C-CAMs. And I don't wanna get too far into the weeds about the terminology. Everyone has a sometimes very strong preference about which term to use. And I personally don't really care a whole lot, but I might use CPAM because at my institution, that's the more accepted term. So these lesions are extremely variable in appearance between patients. They can be solitary or multiple. They can occur in any lobe of the lung. They can be macrocystic, microcystic. They may have anomalous vascular supply. So you may see them look like a lot of different ways. Here are three different examples in three different patients here. One of the biggest pushes that we've seen in the past decade or two is that we are doing more comprehensive prenatal evaluation of these patients. So I did wanna talk a little bit about the prenatal features in case you are someone who is involved with prenatal ultrasound or fetal MRI at your institution. Here is a patient who had a chest mass discovered in the second trimester that was suspicious for a CPAM. And they underwent fetal MRI at 23 weeks gestational age. The findings are that there is this huge mass, which is slightly heterogeneous, has these more cystic components, but it's occupying most of the left chest, causing severe rightward shift, inversion of the diaphragm, and the patient also developed some ascites. The patient was very closely monitored because of these findings, and steroids were administered. And thankfully, they started to turn the corner. So at the time of repeat MRI at 35 weeks gestational age, you can see the appearance of the lesion has substantially changed. There's a more consolidated appearance of the solid appearing portions of the mass, and the cystic lucencies are a little bit more conspicuous. The diaphragm now has a normal morphology, and the ascites had resolved since the initial study. The patient did amazing in the perinatal timeframe and was delivered without many respiratory symptoms. They underwent routine pre-surgical planning CT, which showed further volume loss and relative shrinkage of size of this mass compared to size of the patient, causing superior displacement of the left kidney. So they had an unremarkable lesion resection, and actually did quite well long-term. So what are the things that we should really emphasize when we are reading these studies and reporting on them? Again, I personally am not so convinced that getting the exact right diagnosis is the most important thing we can contribute. As radiologists, I think sometimes we get tunnel vision that that's the most important thing, and certainly it can be really helpful. But I would emphasize that lesion appearance and the secondary features of the lesion could really be helpful to the folks who are helping to drive management, like the neonatologist and the surgeon. So you definitely wanna talk about location of the lesion, the size and the volume, if you can identify an anomalous vascular supply, as well as the secondary mass effects and any evidence of vascular compression, like development of a pleural effusion, ascites, or high drops. Some types of CPAMs can grow extremely rapidly in the second trimester and really blossom very quickly, so that is something for you to be aware of, that they can change pretty rapidly over time, whereas others seem to be a little bit more static. Sometimes regression or stability of the lesion is demonstrated in the second trimester and postnatally. The high-risk features that have been demonstrated in the literature include a high CVR. CVR is the lesion volume divided by the head circumference of the baby as a relative measure of size of the lesion compared to size of the fetus. When we approach 1.6, that's when we kinda get into the danger zone of an increased risk of high drops, where we need to consider formal evaluation with fetal MRI and a closer observation of the patient with the possibility of intervention. If you can see that the placenta is enlarged, that there's severe mediastinal shift, that the diaphragm has become inverted, that there's obvious lung hypoplasia, or if an echocardiogram is performed and is abnormal, these are all high-risk features that have poor prognosis for the patient. A typical postnatal course, if you're only seeing postnatal imaging for a CPAM, would include something like this. So this child had an OCPAM in the right mid to lower lung, so on day of life one, there is a delayed clearance of fetal fluid from the lesion, which is fairly characteristic. The patient did very well and was monitored on an outpatient basis after discharge. At four months of age, they had a follow-up chest radiograph, which shows this well-defined cavitary or cystic lesion, which is now filled with gas, which is the expected course. And then at nine months of age, the lesion is again static, undergoing routine pre-surgical planning. The area of the lung was resected without incidence, and the patient did very well. What you should be on the lookout for is the reverse of this pattern. So if you see a well-defined lesion, like we can see in this left upper lobe of this child here, that then becomes opacified with fluid that is the wrong way around, right? So this kid came in with fever and leukocytosis, was symptomatic, and on CT, there was complete fluid opacification of that lesion, indicating that there was superinfection. The treatment will be altered. In that case, you need to treat for the acute infection, and then the resection can be quite a bit more difficult after infection because of the scarring and granular tissue that develops. So you can really add a lot of value here by suggesting that superinfection is present. Postnatal imaging, you're looking for air trapping or an enlarging lesion or a worsening mass effect with mediastinal shift, and also for the signs of superinfection. Shifting from the congenital phase to intrinsic lung disease, we're mostly gonna be focused here on bubbly lesions in the setting of prematurity. So congenital lesions are nice, I think, as a radiologist, because you can almost develop a continuity of care and treat a patient over time. These are the ones that are a little bit of surprises, right, so they kinda show up and haven't fully cooked all the way. But one thing that I wanted to focus on this category is how we report lung disease of prematurity and how it relates to bubbly lesions. I see these terms in a lot of pediatric radiology reports, words like respiratory distress syndrome and surfactant deficiency and bronchopulmonary dysplasia. And I'm a little bit of a stickler for reports, so I just wanna go over why I think, I double think whether those words really belong in our diagnostic portion of the reporting, the report. First, I wanna start, though, with a poll of the audience. I'm interested to know what gestational age your institution will offer to resuscitate a baby, because I think this is something that has really changed over time. So at your hospital or your institution, will they offer to resuscitate a baby who's 30 weeks gestational age? Everyone, right, everyone would. Okay, how about 25 weeks gestational age? Go ahead and keep your hands up. How about 24, 23, 22, 21? Okay, good. So it seems as I would expect. At our institution, we offer resuscitation at 22 weeks gestational age, and we have accepted transfers of 21 plus whatever days. So the gestational age that we've been able to get down to has really changed over time and keeps getting pushed lower and lower. This graphic, just a basic overview of embryologic development of the lungs, talks about the major different stages of lung development. And what we used to see when we really got into the term bronchopulmonary dysplasia in the 1990s, we were seeing patients mostly in the 30-something weeks gestational age zone during the alveolar stage of lung development. And now we're pushing into this new era where we're going all the way down to 22 weeks in a lot of places. One thing that's interesting is that surfactant doesn't start to be produced endogenously until at least 24 weeks gestational age. So that's one of the things I think about when I see the word surfactant deficiency in reports. Based on the stage of lung development of those patients, it's not the actual absence of surfactant because they don't even have alveoli to really inflate or deflate, right? The type of lung injury in those patients is quite a bit different at a much earlier stage of disease. So just some food for thought to think about when you are reporting on those patients. Bronchopulmonary dysplasia, it's an evolving disease. Definition has been in flux over the past 50 years. And as we've seen, what the type of medical management that we can offer for premature infants has also changed significantly over time. So the old definition of bronchopulmonary dysplasia consisted of exposure to oxygen for at least 28 days with abnormalities on chest radiograph. In other words, the imaging was a critical component of the diagnosis using that specific terminology, bronchopulmonary dysplasia. The newer consensus statement, which is now 20 years old, but we know there's always kind of a delay in getting things from the literature to clinical practice, right? The 2001 Consensus Conference of the US National Institute of Child Health and Human Development created some new criteria for definition of BPD. So let's take a look at those. I'm not gonna go into the detail of the nitty gritties of this, other than to point out that the definition is based on the time point of assessment and the gestational age of the patient, as well as the concentration of oxygen that is necessary for respiratory status. It can be divided into mild, moderate, and severe cases of BPD. But notably absent from this graphic is any mention of imaging. So imaging actually doesn't factor into the diagnosis of BPD according to these consensus guidelines. So again, just something to think about when you're using that term. I think everyone understands what you mean when you say BPD, but it may not fit exactly into the standard definition. Well, here's an example of what I would think about for that old BPD definition. This was a child born at 35 weeks who no maternal steroids were able to be administered before birth. And on day of life two, they have pretty mild pulmonary opacification. It's not, doesn't stand out too much. Hopefully you can see it on the big screen there. But by four months of age, they had developed the pretty classic interstitial coarsening, hyperinflation. And then on the CT at four months, the same time as the chest radiograph, that you can see these scattered little small cysts, which is why it fell into this talk, because those are sort of the classic features that you would see for chronic lung disease or prematurity. What we see now I feel like more frequently at our institution is sort of the newer stage of BPD. This premature infant was born at 24 weeks. Gestation, things looking pretty good on day of life one. We have mild to moderate diffused ground glass opacification and maybe some interstitial thickening. But by day 11, we're getting some more coarse interstitial thickening with scattered tiny little cystic lucencies, some hyperinflation. But then we kind of turn the corner and by six months of age, things don't look too bad. Just mild findings. We can tell it was a preemie because of the ductal occluder clip and the tracheostomy tube. But things can kind of change pretty rapidly over time. With these babies, and the imaging appearance doesn't always necessarily correlate with their respiratory status. So how do I handle reporting of these patients? Well, I tend to use umbrella terms. And I have two main ones that are kind of my pocket phrases. One is lung disease of prematurity if they're less than 28 days of age. And one is chronic lung disease of prematurity if I know they've been around a while and they were born prematurely and I can see that things aren't going great. This sort of encapsulates a lot of different types of terms that can fall under this umbrella category including what you would typically think about for surfactant deficiency, the new and old definitions of bronchopulmonary dysplasia. And if you're following the childhood interstitial lung disease criteria or classification system, disordered alveolar growth is sort of the preferred term that encompasses premature lung disease. So if the terms are kind of wishy-washy or very generic, what can we actually do in our reports to help the neonatologist? They're trying to titrate their ventilation and use as little barotrauma as necessary to ventilate these patients. So you want to talk about your imaging findings in terms of the general pattern that you see, the symmetry of the lungs, what are the lung volumes, what are the densities that you see and where are they located? Is there development of cardiomegaly? And trending the imaging findings over time can be really helpful to our clinical colleagues. The last category that we're gonna visit is very closely related to premature lung disease because it consists of a few different types of air leaks that develop most commonly in the setting of lung disease or prematurity. These are parenchymal diseases that are known complications of respiratory distress syndrome. This is one of the terms I had on the last category, just a quick note to say that this is a clinical diagnosis. It is most commonly associated with lung disease or prematurity, but again, not necessarily an imaging diagnosis. These air leaks can be iatrogenic and they're most commonly associated with barotrauma. The first is pulmonary interstitial emphysema, or PIE, which results as a rupture of over-descended alveoli leading to entry of air into the interstitium. Pretty classic example here in a little preemie who developed PIE on day of life two. We have tiny little scattered cystic lucencies and radiating linear lucencies coming from the hilum involving the entire left lung, which is hyperinflated. There's also involvement of the right upper lung and then a pneumothorax that develops. So tiny little cystic bubbles, lucencies throughout the lung. If we trend this baby's chest imaging over time, day of life one had moderate diffusive pacification, rapidly developed the PIE bilaterally. The pneumothorax was better demonstrated on the cubitus view. And then by day of life seven, something interesting happens. The left side PIE completely resolves and now we have PIE mostly in the right upper lobe. And by day of life 45, we've developed chronic lung changes which are fairly diffuse other than a couple areas of air trapping. So these babies tend to change quickly and move very quickly with our imaging findings. There is an entity maybe you haven't heard of before called persistent pulmonary interstitial emphysema and this is essentially a chronic form of the typical type of PIE that we think about, which is rapid and quick and changes day to day. This is something that doesn't change day to day. So this is a 25 day old premature girl who developed multiple cystic structures throughout the right lung and had a CT scan which demonstrated these tiny little punctate and linear densities surrounded by gas. This is called the line and dot pattern and is felt to be a fairly characteristic finding for CT diagnosis of persistent pulmonary interstitial emphysema. Other types of air leaks include pneumopericardium and this is a big old bubble in the chest. It occurs when gas is contained within the pericardial reflection with lucency that outlines the heart, aorta and pulmonary artery. This is important because clinically it can cause tamponade which is manifested with increasing respiratory distress, cyanosis and muffled heart sounds. Sometimes it can be difficult to differentiate pneumopericardium from other types of air leaks like pneumomediastinum or even pneumothorax but if you have a suspicion in something that looks like it's completely surrounding the cardiac silhouette, you might wanna give your clinician a heads up that this is possible and they should be on the lookout for signs of tamponade. Pneumomediastinum, hopefully this looks familiar to most of you. This is a very common air leak that happens in the setting of premature lung disease and anterior pneumomediastinum oftentimes has this very classic appearance where lucency outlines and lifts up the thymus creating the spinnaker sail sign. Anterior pneumomediastinum usually resolves without conservative treatment even when it's very large. So even though it can look kind of scary, it usually doesn't have any long-term consequences. Posterior pneumomediastinum is a separate entity that I wanted to be sure that you're aware of by the end of this talk and that is because it can cause and decrease in venous return to the heart and have pretty dire consequences. So here's an example of a baby that was born at 27 weeks gestational age who developed a large bubble in the center of the chest on the frontal view and when we looked at the lateral view, we could see this localized posterior to the heart. So this is posterior pneumomediastinum. This baby unfortunately wasn't doing very well. They underwent an echocardiogram and the representative image is down here. Now this is an echo so it's kind of flipped from our normal orientation but the right and left ventricle are down here. The right atrium is normal in size and looks pretty good the left atrium though is teeny tiny and being squished by this large collection of gas in the posterior mediastinum. So this baby was too sick and too small to really undergo any significant intervention and unfortunately expired but perhaps if we had recognized it sooner, maybe there could have been a different outcome. This is another example from the literature. It is a kind of a rare entity so I just wanted to give you an example of what this might look like on CT. Another preemie here who developed a midline cystic structure with gas dissecting into the upper abdomen. Just to show you where this localizes, the literature on this entity is mostly from like the 70s and 80s I think and they talk about gas collecting in the pulmonary ligament and infrasigus recess. I'm not sure how important that is but it is important for you to know that if you see a sizable collection in this region that the baby is at risk for impaired venous return and that deserves a phone call. So in summary, we have a variety of bubbles that can happen in the perinatal period starting in the congenital time frame. The things that are important for you to think about are the high risk features of an elevated CVR, 1.6 or higher, an enlarged placenta, diaphragmatic inversion, severe mediastinal shift and lung hypoplasia. Those are things for sure you want to include in your report. Lung disease of prematurity, think about the terminology that you're using and carefully like bronchopulmonary dysplasia and try to focus on the findings and the imaging like the pattern of lung abnormality, the degree of lung inflation, symmetry, heart size and trending imaging over time. And for air leaks, be on the lookout for posterior pneumo-mediastinum which can be really important for care of your patient. Thank you to everyone who is hanging in there late on a Wednesday afternoon. We appreciate your time. So I'm gonna talk about a pretty broad topic, pediatric chest wall. There's a lot of things that live in the chest wall. I cannot possibly cover that in 15 minutes so we're gonna go on a bit of a tour. And I've included some common everyday things and a few more unusual things to try to make this a little bit interesting and keep you guys awake at this late hour. As I said, the differential is pretty long so impedes the question of normal versus abnormal is often the first thing we deal with. So there's a lot of normal variants that happen in the chest wall, a lot of issues with structure. Trauma is always very important when you're dealing with kids, particularly babies and infants. The differential for focal lesions in the chest wall is quite long. Typical pediatric differential, a lot of different things to deal with here. There are multiple benign things that we wanna think about but the one to really pay attention to is gonna be Ewing sarcoma in the chest wall. That's a bad actor and pretty common. Infection occasionally crops up so we'll show a little bit of that. We see a lot of vascular malformations at my hospital. Mostly venous and lymphatic, though occasionally high flow. And then we'll round out the talk with a little bit about metabolic bone disease and bone dysplasias. Can everyone hear me okay in the back? See some nods, thank you. So we're gonna start off with an unknown case. This was a 16-year-old who presented with pain for an MRI of the cervical spine. The MR Tech was gracious enough to notice that we had this huge signal void on the scout and the initial coronal T2 weighted sequence so really can't see anything. We pulled the patient off the scanner. What the heck's going on here? Get an X-ray and we see this really long radiopaque foreign body despite the patient denying any history of metal or radiopaque foreign body. Get a CT and we can see that this is a sewing needle right next to the subclavian artery denoted by the red arrow. And upon further questioning, the patient did admit to being stabbed in the front of the chest a few weeks earlier. She kind of forgot about it and didn't think it was a big deal. So two teaching points there. In pediatrics, you want to keep your mind open. The differential is broad. Weird things happen. And for those of you with kids, we'll know that teenagers are liars and never tell you the truth about anything. So always keep that in mind when reading teenage imaging. So we're gonna start off with normal anatomy and a couple variants. So this is an advertisement for ultrasound, particularly in the chest wall, particularly in younger kids. So we can see at one centimeter of depth on this ultrasound of the chest wall that we can see almost all of the structures we need to see. Sub-Q fat, costal cartilage, the intercostal muscles are beautifully depicted. We can see the layers of the muscle fibers and we can see the pleural reflector. So often ultrasound can be fully diagnostic for a lot of the things that we deal with in the pediatric chest wall. To start off with, first case here is gonna be pectus excavatum. This is really common. I probably don't need to tell most of you guys about this. Some publications describe this as up to 90% of anterior chest wall malformations or diseases. So you measure your transverse diameter and your AP diameter, calculate a Haller index. Depending on your insurance provider, an index of between three and three and a half is gonna be enough for surgery. I usually try to nudge that up a little bit just to give the benefit of the doubt. And at my hospital, we treat these with a NUSPAR. So you just put the bar underneath the sternum. Over time, it'll pop the sternum out, hopefully correct the deformity. One thing to keep in mind, especially in cases of more severe pectus, is that these can coexist with sternal tilt or sternal torsion, which can make the surgery a little more complicated, may require an osteotomy to level out the sternum to allow it to be pushed back out. So that's a nice thing to mention to your surgeon. Another thing that we see a lot, especially in younger or skinnier kids, is they'll present with a hard, quote-unquote, bony lump. And despite seeing bony in the script, we really like ultrasound for these cases because you can usually make the diagnosis more easily even then on x-ray. So this is an example of a rib asymmetry, often just a slight overgrowth of the rib cartilage, usually anterior near the sternal junction. And it's a normal finding. Doesn't require further workup, doesn't require any treatment, and you can send the parents and the child home happy and say, stop touching it, gain some weight. Look like some of us slightly older people, you won't feel it anymore. Moving on to trauma. So trauma is a really big topic, particularly in younger kids, really everyone, but child abuse in the infant and toddler population comes up quite a bit, especially since COVID, unfortunately. So this was a two-month-old who presented with irritability and sort of an intermediate abuse workup in the ER, maybe some bruising, they weren't really sure. The person that read the survey said they were concerned about some possible expansion of the anterior rib ends on the left. That can be a really tough call depending on how early the injury is. So next teaching point is please consider CT for these cases. If you're concerned about a rib injury or if there's chest wall bruising, don't be afraid of CT in newborns and young infants. Our doses are really, really quite low now and especially for costochondral junction fractures, sometimes they're impossible to see on x-ray, especially if they're early. So we can see expansion of the fractured rib ends with heterogeneous mineralization. Compare that to the normal rib end which has a homogeneous width and homogeneous density. So these are costochondral junction fractures. Excuse me, they are considered metaphyseal equivalents, right, that's where the physis equivalent at the end of the rib lives. So this is a high specificity abuse fracture and should be treated like a classic metaphyseal lesion anywhere else in the body. Here's a more obvious case of a child with multiple costochondral junction fractures. We can see expansion and heterogeneous density in the fractured ribs and non-expanded normal density in the non-injured ribs. So the coronal sometimes is really helpful because you can see multiple ribs and get symmetry. MIPS are also helpful and the 3Ds sometimes help with these. So here's an axial view, same child, and we can see not only are the rib ends expanded and kind of heterogeneous in density, but we've got this weird irregular contour with these metaphyseal clefts. So this is a growth disturbance caused by trauma, caused by a fracture to the physis equivalent. The normal rib at a lower level is uniform in thickness, uniform in density. So again, don't be afraid of CT in babies. It can be really helpful to make the diagnosis and to bring to the courtroom if you don't have other definitive findings. There was a very recent, really nice paper describing a lot of these findings and similar to our index cases, the fractured rib ends often get expanded, mottled, with mottled sclerosis and these weird metaphyseal clefts. So moving on to focal lesions. Pretty broad differential. Here's another young child, nine-year-old, who presented with a hard lump, and we can see this kind of irregularly shaped but smoothly contoured osseous excrescence with a little cartilage cap on top. Yes, this is an osteochondroma. This was an oblique radiograph. You can kind of imagine if you had maybe an AP, it didn't catch that quite right. It might be hard to see. So again, ultrasound can be really nice for these. Here's another case. This was an incidental finding in an 11-year-old who presented with a cough. We have an expanded and slightly lucent rib. Little bit hard to see because there's a lot of overlap. So went on to CT. We see a very expanded rib with some ground glass mineralization internally, and which was added on bone scan. So this is fibrous dysplasia. These are usually asymptomatic. They tend to present a little bit older in childhood or teenage years, so bigger kids, not always palpable. Occasionally, we see them present with fracture. Remember, the polyostatic forms of these, if you have precocious puberty, you want to think about McEwen-Albright, and if there is a soft tissue mass anywhere else in the body, then you're thinking myxomas in the setting of mesobroad syndrome. So next case is a 10-year-old with chest pain. So I just wanted to take a moment. A lot of the time when we get these chest pain films in kids, we tend to roll our eyes. 99% of the time, it's nothing. There are some serious causes of chest pain in kids, though, and this is one. So we see an expanded and dense rib on the left. Patient went on to get a CT, and it looks, at first glance, kind of similar to that fibrous dysplasia case. The rib is expanded. There is some sort of heterogeneous mineralization internally, but if you look more closely, there's a cortical break and an adjacent soft tissue mass. This patient did not present with fever, went on for a biopsy, and this was a Ewing sarcoma. So this is something to really have on your radar in any pediatric chest radiograph, particularly with chest pain or lump or whatever, if the rib looks funny. This is the most common rib malignancy in children. The age peak varies a little depending on who you read. Eight to 13 or 15 is probably good, but you can see it in kids as young as five. It's a little bit younger than osteosarcoma, so we do see it in kids who are quite young. This is one of the real causes of pediatric chest pain, quote-unquote real, and just a reminder that there are other things now included in the Ewing classification. So there's classic osseous Ewing. There's non-osseous Ewing, which can be in the soft tissues and the mediastinum and elsewhere, Askin tumor and PNET, and they can be kind of hard to tell apart. It's not really our job. That's more of a biopsy thing. Moving on to infection. So here is a four-year-old who presented with chest pain, and this patient did have a fever. So ultrasound beautifully depicts a complex fluid collection with debris and a very clear cortical break. So right there, you've got your osteomyelitis with an adjacent soft tissue abscess. The patient did go on for cross-sectional imaging because it was pretty big, and we wanted to assess the extent of disease. So you can see the cortical break on the T2, and we can see a little bit of thickening and buckling of the extra pleural soft tissues on the post-GAD. No effusion, no invasion into the lung, so you kind of have your extent there. Chest wall infection is not something that I tend to see every day at my practice. Maybe Dr. Eklund would agree, but it's not that uncommon. Usually gonna be from direct injury with or without a foreign body, and hematogenous infection. So another note on metaphyses for hematogenous infection. So in kids without fused physes, we're looking for metaphyses and metaphyseal equivalents as a source of infection. Spontaneous soft tissue abscesses in kids are pretty rare. So if you have one, you better find the site of osteomyelitis because it's usually there, and it's gonna change treatment. And in the chest, we're talking about the rib ends, costochondral junctions, we're talking about clavicular head. There's a whole bunch around the scapula, proximal humeri, et cetera. So there's quite a few. If you do have really extensive trans-spatial extension, like through multiple chest wall layers, muscle, pleura into the lung, then you wanna start to think about unusual organisms. TB is kind of notorious for doing that, and you may wanna give a call to the clinician to do an immune-compromised workup if it's going through quite a few layers. Touch on vascular lesions briefly. So this was a six-year-old who was transferred with a remote history as a baby of a treated chylothorax, now with increasing body wall swelling. We can see on the coronal and axial T2s, there is tremendous thickening of the left chest wall tissues with several fluid locules that have complex T2 hypo-intense debris with fluid layers or blood layers. On the T2 space MIP, we can see innumerable little cystic spaces. And then this patient went on to have an MR lymphangiogram, which if you have not seen done, involves injecting gadolinium into the groin nodes most commonly. So we can see a mildly tortuous, mildly dilated thoracic duct, which becomes very tortuous and very dilated at the cephalad portion near the left venous angle, consistent with thoracic duct obstruction. So this is an example of a lymphatic disorder, lymphatic disease. It's a huge topic with a lot of different names. We're not gonna cover that today. Some are kind of static names like chylothorax, protein-losing enteropathy, plastic bronchitis, but a lot of the diffuse ones are still evolving and changing their names depending on where you read, like Kaposiform or generalized lymphatic anomaly, Gorham-Stout, et cetera, there's a lot. Big topic, but in general, we use conventional imaging to assess the extent of disease and the distribution for these, and then for treatment planning, you need advanced lymphatic imaging. So you've gotta inject the lymphatic system somewhere and either do flora or MRI. So that can be inject the groin nodes, inject the hepatic lymphatics or mesenteric lymphatics, or even old school nukes where you inject the toes, but we mostly do groin node and hepatic at my hospital. So we'll round out here with metabolic disease and dysplasias, and I think I'm gonna stop there just to respect your time. So this is a case of an 18-month-old with cough. Incidentally noted, there are markedly expanded rib ends. This is a case of rickets. This was a nutritional case. We see a lot of rickets of prematurity or ICU rickets, but there are other causes like vitamin D, et cetera. So this is an issue with proper cartilage maturation and not enough matrix depositions. You're gonna see, in addition to diffuse demineralization, particularly severe demineralization at the zones of provisional calcification, which you're gonna have in a regular contour, a little cartilage clefts, fraying, cupping, quote-unquote. We'll see expanded rib ends and then occasionally fractures. This comes into play in the differential for non-accidental trauma. More on the flip side, if you have a case with concern for abuse, and you can say there are no signs of rickets, that can be really helpful for the clinical team and the legal team. And then the last thing I'll show is skeletal dysplasia. Don't be afraid of these. They are, they can be scary, but it's really not our job to make the diagnosis. It's our job to describe the distribution, what bones are affected, metathesis, epiphysis, et cetera, and then help the geneticists to make the diagnosis. And we should be diagnosing complications like cranial cervical stenosis, tracheal stenosis, hip dysplasia, pulmonary hypoplasia, et cetera. Just got two slides left if anyone wants to leave, but this is a case of Ellis Van Creveld. We have short ribs, congenital heart disease, short curved humeri, and then look at the hands. We've got extra digits, so pretty classic case. We see a lot of these coming down from Pennsylvania at my hospital. And here's an example of thanatophoric dysplasia. I only ever read about this in textbooks in residency, but they do exist in real life. Very, very short ribs, severe pulmonary hypoplasia, so that's our job to comment on, and short curved humeri.

pediatric chest imaging

lung masses

differential diagnosis

congenital anomalies

neonatal lung disease

vascular anomalies

Ewing sarcoma

bronchopulmonary dysplasia

non-accidental injuries