All right, so we're going to start with a talk on consolidation and ground glass capacity. And then here's the other three topics. It'll lead directly into a mosaic lung attenuation talk by Brett Elicker. Then we have cystic and micronodular disease. And we'll have each speaker introduce the next speaker. So as I was thinking about consolidation and ground glass, I sort of had two observations I thought were good orienting thoughts. The first is this is actually the least high-res specific sort of diagnosis, I think. When you think about consolidation, it looks pretty similar between thin section and the high-resolution algorithm. Ground glass looks pretty similar too. So it's not super high-res necessary for us to think about this. The second is consolidation and ground glass capacity has sort of the largest and broadest and most overlapping differentials. So here are four consolidations. Which one's which? We have an answer? No, because each one's a huge, gigantic differential. No way to know if it's a radiation, organizing pneumonia, here's a cancer, an infection. So being super specific is hard. So how can we be useful here? I think the first is no matter how much overlap there is, we want to use the most precise language that we can so at least we're not adding even more uncertainty. If there are specific findings, we want to make them so we can try to hone the diagnosis. And then the last way that we add some specificity is actually looking at the broader patient context. So we'll go through each of these and we'll start with precise language. And the issue here is really twofold. One is if we look at it and we're not totally narrowing it down to what we think it is. So we start saying a gobbledygook like this, it means absolutely nothing. Might as well say nothing there. So we have to make up our mind what we're seeing, then we have to use the most narrow, precise words possible. So nodule actually means something once you throw a line in front of it and a vagueness after it, now it means nothing. All right, so let's talk about the vocab words we can use. A consolidation is whiteness or opacity where you cannot make out the vessels in between. Or ground glass opacity, you can make out the vessels or some underlying architecture. Now these can exist together, so it is fine to say this is both consolidation and neighboring ground glass. This has more consolidation, this is mostly ground glass. It's okay that it's both, that is still precise. So what are the sort of other vocab words we want to distinguish it from? So consolidation, airspace filling is different than a mass, right? That is taking up space, destroying architecture or displacing. And this is for a large such structure and a nodule is the same thing but smaller, less than three centimeters. Now in real life, it can get complicated, right? So here, two little squares, I think those look like two small nodules, very precise. But if you look at the whole slice, then probably it's actually part of a broader patchy consolidation. So I wouldn't call those discrete nodules. So what are some other ones we have to make a distinction from? Consolidation is different from atelectasis. So this is volume loss, we have crowding of the airways and the vessels. Or a more linear or fibrotic type appearance, this is fibrosis, this is also different. All right, ground glass opacity, what are the vocab words that we need to be precise over? Ground glass hazy opacity is different than many nodules, right, that is actually a different differential. Or from a single nodule, also a different differential. So we wanna be specific between these three. And then what do we wanna do in distinction from? Ground glass, we should be able to differentiate from atelectasis. So here a little bit of hazy opacity along with gravity, we just say it's atelectasis. We hear almost the whole lobe is collapsed from effusion, pretty comfortable saying it's probably all or mostly atelectasis. All right, since we're talking about vocab words, every time since I started I hear ground glass, I've always wondered, is it the thing with a dash, or does it have a space, or is there no space? Who know, or even I just did it without the word opacity. So how do we get the answer? By Dr. Google, of course. We have millions and millions of the first ones, probably can't tell the difference with a dash or not, but fewer when there's no space there, and you can't search the last. So it looks like the top two win. We could also not use Google. Fleischer Society Glossary recommends the one with the dash. So I guess that's the right answer. And what is it? It is glass that has a sort of hazier matte appearance, either ground mechanically or etched with acid, and now it's a sort of matte, not shiny. It's ground glass. All right, so precise language, we want to decide what we're looking at and describe it precisely, and that gives our differential. Next is we wanna look for specific findings, and the goal here, of course, is that this narrows the differential so it's not such a big, overwhelming differential. So in the world of consolidations, there are five things, I find, that help us sort of add some specificity, and we'll go through each. The first is, if you have multiple rounds of imaging, is how acute or not is this opacity? Because certain dense consolidations, or just consolidations, have different sort of time courses, and then that will help narrow it for us. So I'm gonna pull out of sort of all of our options just three to give you some examples. So pneumonias, the classic, most common ones, are acute, right? You're fine, you get sick, then you get better. So all these viral pneumonias, particularly right now, are on the more acute side. Now, there are occasional infections, so here's fungus or TB, that would be sort of more subacute, or even potentially chronic, but many of our infections are on the faster side. Radiation is an example where it sort of runs a gamut of chronicity, mostly sort of subacute to chronic. So in the earlier phase, for months, you actually don't have sometimes super acute, but then it starts showing up, and progresses over months and months and months, this inflammation, and so that leads to hypermetabolism and opacity. Eventually, it starts evolving to fibrosis. So then, at some period of time, it gets smaller, denser, sort of more speculated because there's fibrosis, and it can get confusing. Even there, it can slightly progress, that has messed some people up, or even have a little bit of metabolism. So for example, here's a treated right hilum, no opacity in the lungs at that one month, but then it really starts showing up, starts consolidating down, and here, even before six months, it's sort of looking like more fibrosis is forming. Or here's a ground glass nodule that got treated, and after it got to that fibrotic stage, smaller, denser, has the more angular fibrotic look. And again, PETCT is tough during this sort of subacute to chronic phase. Here's a really nice example, a non-con chest and opacity that was fibrosis. PETCT really shows us the recurrence, that's only because we're not in the window where it's hypermetabolic from the treatment. And then the chronic examples are cancers. So here's a lymphoma, and even after treatment, this lymphoma has lots of residual fibrosis, the opacity is chronic before and after treatment. Or here's a lung cancer, specific type that leads to these consolidations, very slow growing, so these are chronic opacities. All right, next specific thing to add. If you can see the reverse halo, or atol sign, where we have increased density in the outside of a lesion, little bit less dense on the inside. We think of this classically as organizing pneumonia. Turns out the sensitivity is not that great. So this is not a required finding for you to suggest that it's organizing pneumonia. How about the reverse, though? Can you get faked out by this? Absolutely, so here's an infection, a little bit less dense in the center, infarct less dense, and a really nice infectious lesion, very not dense in the center. So not 100% specific or sensitive. The sort of specificity then depends on who we're imaging. So in immunocompetent patients, it's pretty reasonable to suggest organizing pneumonia. In immunocompromised, we're probably an infection most likely. And that brings me to my last point, which was how important it is to add these sort of clinical factors. The next sign is a halo, so the less dense is on the outside, classically described in this invasive aspergillosis, but actually pretty, not terribly specific. I mean, maybe it'll help you there, but we do see it in other things. Here's some mets and a lung cancer. The next few ones that add specificity, so one is some nodules around consolidation. Here's a really nice galaxy sign, perilymphatic nodules. So the overall pattern is pretty helpful towards sarcoidosis. Next is pleural abnormality, some abnormal pleura with the lesion next to it, volume loss, the sort of swirling of that comet tail sign, classic rounded atelectasis. And last one for consolidation is if the consolidation has an abnormal density. Very dense here for amiodarone, very not dense here for lipoid pneumonia. We don't see this much, but this helps. All right, so that was consolidation. Ground glass, what adds specificity? Again, we have a couple options, or items, and I'll go through each one of them. First is if it is focal, patchy, or very diffuse, because each of these actually have a differential that can help narrow it. Now, there's still many options, but I think that's an important, helpful, narrowing description. Next is chronicity, and I sort of think in ground glass, more in just two categories instead of three, and there's the sort of acute ICU or hospital admission level diagnoses, and everything else. Next specific sort of helpful addition is if you have ground glass opacity and interlobular septal thickening. So this is where it leads to that very sort of beautiful paved pattern, so-called crazy paving. Now, people love to attach this to alveolar proteinosis, right, but is that really that specific? That's absolutely not. So here we have just lung damage. Here's the ever-common Niemann-Pick disease, a storage abnormality. We see this sort of pattern with other diseases, certainly. And really, it's the same differential as a ground glass. And one more example, again, not from proteinosis. So most commonly in our practice, this is probably from sort of an acute lung injury situation, maybe pneumonias, and it is true in proteinosis. It does show up fairly frequently, but if we're looking at acute disease, it's probably not proteinosis. The next specific finding is air trapping, and that's actually the whole next talk, so I will not add into that, but add some specificity. And the last one is if you ground glass opacity with other findings that make it more of a classic ILD case where we go down which ILD is it. And that, of course, was covered in other courses related to this one, so I won't go into depth, but that's the last specific finding. And so then my last comment about how do we be helpful here? We use precise language, we look for specific findings, and then we try to tie it together with clinical factors. And this is a very broad differential of categories, specific diseases, even our super common Niemann-Pick disease. So how do we narrow it? It's this clinical context. Even if you don't have the imaging, what is the clinical time course? Risk factors of the patient, patient demographic, et cetera, et cetera. We help integrate these factors into the imaging. And it's sort of fun. This is what's nice about chest imaging. This is the being a doctor part of radiology, integrating together, and actually one day if we actually get these data properly into a computer system, I think it's great for AI as well. So when we see dense consolidations in ground glass opacity, big differentials, but our goals here are to use precise language, add specific findings if we see them, and then integrate the clinical picture. And that's it, thank you very much. Because mosaic attenuation. And mosaic attenuation is essentially a compilation of two different findings. It's when there are two densities of lung, which are usually fairly geographically separated from one another. And in essence, as I said, it's a compilation of two different findings. And the terminology's a little bit controversial here. Some people like the term mosaic perfusion, others do not. I do because it makes a commitment as to the fact that the lucent lung is abnormal. So mosaic attenuation is a summation of mosaic perfusion plus ground glass opacity. Mosaic perfusion represents areas of lucent lung. You're making a judgment that the lucent lung is abnormal. It's most commonly due to some kind of airways disease. And it results in decreased perfusion, even when it's due to airways disease. And those mechanisms are, in airways disease, you get hypoxia and reflex vasoconstriction, which reduces blood flow. That's why we call it mosaic perfusion. It also may be seen with vascular disease, mainly chronic thromboembolic disease, in which case it's specific narrowing or stenosis of vessels. So this is lucent lung abnormal. The other finding which is in that category of mosaic attenuation is ground glass opacity, which is one of the most nonspecific chest CT findings we see, because it represents changes below the resolution of CT. Those changes may be alveolar, interstitial, or both. And the differential diagnosis, as Dave mentioned, is mainly based on symptom duration, distribution, and associated findings. And in this talk, I'm really gonna focus mainly on mosaic perfusion, lucent lung is abnormal. Let's first talk about how to make that distinction between mosaic perfusion and ground glass opacity. These are all features of mosaic perfusion, lucent lung abnormal. It tends to be very geographic, very sharp borders between the abnormally lucent lung and the more normal opaque lung. And narrow windows can be very helpful here. This is why I think using a very narrow window can accentuate these differences and uncover areas of geographic lung attenuation. And you can see, this may be kind of patchy, lobular regions in airways disease. It can become larger, but even here in airways disease, it's pretty patchy. And as I said before, it can also be seen with vascular disease, where it tends to be much larger, usually peripheral swaths of lung, which are affected, and it's less lobular and less patchy in nature. So that's one feature. Another feature is vessel size differential, which is helpful when you see it, but you don't see it in all cases of mosaic perfusion. So when you see, and this is, again, case of airways disease and vascular disease, the vessels are much larger in the normal, more opaque lung, and are smaller in the abnormal, lucent lung. And when you see that vessel size differential, it implies the lucent lung is abnormal and the pattern is mosaic perfusion. Lastly, we routinely do expiratory images in almost all of our high resolution CTs, at least the first one. And as most cases of mosaic perfusion are due to airways disease, you would expect, as the patient breathes out, those areas which are lucent on inspiration will stay lucent on expiration and not change in density, right? That's abnormal. You can see that the lucent areas of lung are not changing in density, whereas the opaque areas of lung are getting much, much denser. So about 20% of patients with air trapping have a normal CT on inspiration, so this increases your sensitivity for detecting disease in general. It increases the diagnostic accuracy in terms of what your differential diagnosis is. And the most common things that give you air trapping in isolation, and we'll talk more about this later, mosaic perfusion and air trapping are helpful when they are seen as the main finding or an isolated finding. And these tend to be common things like chronic bronchitis, asthma, and we'll talk more about constrictor bronchiolitis in a minute. Other features that can help us distinguish is the pattern mosaic perfusion or ground glass. Well, associated findings. If you see other findings of airways disease, like in this patient with cystic fibrosis, we see bronchiexus in airways inflammation. This implies that the lucent lung is abnormal from airway obstruction. We also see this in vascular disease. Two lung densities, more normal lung is central. All this lucent lung peripherally is abnormal. We can see findings of chronic pulmonary hypertension, maybe chronic thromboembolic disease. So, associated features may also be helpful. Those are all cases of mosaic perfusion, lucent lung abnormal. I'm not gonna talk about ground glass, as Dave talked about that already, but just to contrast the features of ground glass opacity compared to mosaic perfusion, right? These are usually a greater difference in density between the, in this case, abnormally opaque lung, and the normal, more lucent lung. Ill-defined margins, they don't have that nice sharp border. No vessel size discrepancy, and then if we got expiratory CT, we would expect both the opaque and the more lucent lung to increase in attenuation as the patient breathes out. Okay, that's just to contrast that. This is a case of pneumocystis. How good are we at making this distinction between ground glass and mosaic perfusion, either due to airways or vascular disease? It turns out that we're pretty good. 80 to 90% of the time, we can make that distinction. With the exception of mosaic perfusion seen in the context of vascular disease, in which our accuracy goes down quite a bit, and I'll show you a few cases of that, and accentuate why that is very difficult. But let's transition now to differential diagnosis and talk about airways disease specifically, which accounts for the vast majority of cases of mosaic perfusion. So we're gonna talk about differential diagnosis of airways disease. Now, small airways diseases affect the middle of the pulmonary lobule. That's where the airway lies, and they affect it in different ways. And there are classification schemes for airways disease that breaks it up into mainly two categories. Nodules, whether those nodules are ground glass or soft tissue, or there's tree and bud associated with those nodules. And then focal areas of lucency, which is mosaic perfusion. Now, I'd say mosaic perfusion can be seen in any one of these categories, but it's most helpful when it's seen in isolation without nodules, seen as an isolated finding. And that's what we really wanna talk about, mosaic perfusion as an isolated finding. This is a study looking at different etiologies of isolated mosaic perfusion. And there's a lot of things to remember here. It's not actually all that important to remember the specific causes. I would point out some trends. Everything I've highlighted here in red is associated with a specific pattern of injury. In chess, we like to talk about patterns of injury. Nonspecific interstitial pneumonia, usual interstitial pneumonia, lymphoid interstitial pneumonia, organizing pneumonia, blah, blah, blah, blah, blah. And then you get very confusing. All of these I've highlighted in red end up with the same pattern of injury. And that pattern of injury is called constrictive bronchiolitis. In our practice, constrictive bronchiolitis probably accounts for a very high percentage of patients who present with isolated mosaic perfusion or air trapping. So it's an important pattern to be aware of. It has several synonyms. It basically is a histologic pattern where you get scarring forming around the small airways, very distal. So you actually don't see that scarring on imaging. You see it only on histology. What you see is the fact that the error is obstructed at that spot and doesn't get past it. And so you see the secondary effects of hypoxia and reflex vasoconstriction. So the main finding is actually mosaic perfusion and air trapping. The lung is otherwise normal. There's a variety of different causes. If it's post-viral and occurs in childhood and is asymmetric, we call that Swire-James, right? Swire-James is a post-infectious bronchiolitis or constrictor bronchiolitis. We see it with connective tissue disease, with drugs, with various toxic inhalations. Popcorn lung's a classic one where you get diacetyl. This is the main pattern we see in graft-versus-host disease. So that's the main finding you're gonna be looking for in that context. And this is the most common manifestation of chronic rejection in lung transplant. All of these end up in the same place and look the same on imaging. The treatment actually for constrictor bronchiolitis is different from any other ILD. It's probably not very effective, but it does have a unique treatment. So it actually is important to make this diagnosis of this entity. There is a large spectrum of different imaging findings that may be seen with constrictor bronchiolitis. And I've shown the extremes here. Sometimes all you have is air trapping. And you don't even have anything on inspiratory CT. Sometimes you have isolated mosaic perfusion. When constrictor bronchiolitis is severe and it's been there for many, many years, eventually the large airways start to dilate, probably because it takes a lot of pressure to ventilate through that area of obstruction, so the large airways start to balloon out. So if it's been there for, this is a patient with a post-viral childhood viral infection and then 60 years later, this is what they look like. So eventually you start to get bronchiexis. But most findings, most cases are gonna be isolated mosaic perfusion or air trapping. So this is an important pattern to be aware of. What about, as I said, vascular causes of mosaic perfusion are challenging. The most common one to do this is gonna be chronic thromboembolic disease. So if you have a patient coming with pulmonary hypertension, you get a high-resolution CT, and you see these geographic areas of lucid lung with sharp borders, particularly when they're peripheral and large, really, really think about chronic thromboembolic disease. I think, I'll talk about the challenges in a second. This is a study looking at the different causes of pulmonary hypertension presenting with mosaic perfusion, and the majority were vascular diseases, and in our experience, chronic thromboembolic disease is very common. Okay, another two more minutes or so. Talk about some of the challenges. I've talked about some of the ways we make that distinction between mosaic perfusion and ground glass. None of them are perfect. And you're gonna see cases that, you know, sometimes overlap with each other. One situation is sometimes, particularly with viral infection, we see this a lot, ground glass can be very geographic and very sharply demarcated. These are all cases of ground glass presenting with nice, sharp, demarcated borders. So that's, again, not a perfect finding. No finding is 100% specific. In most cases, the difference in density between the abnormally opaque lung here, the ground glass, and the more normal lucid lung is a lot bigger than with mosaic perfusion. In most cases, mosaic perfusion, the density differences are usually quite mild. So that may be one thing to help. But this is sometimes challenging. Okay, so those are all cases of ground glass. These are all cases of mosaic perfusion. As the extent of mosaic perfusion gets more and more severe, it gets more difficult. This is a pretty easy case. Lucid lung, sharp borders. This is a little bit more challenging. There's quite a bit of difference in density between these two. The borders are very sharp. We call this ground glass in the first scan, and then it was unchanged six months later. And then we did expiratory CT, and the lucid lung didn't change in density. So actually, this is a case of a very extensive mosaic perfusion where all that blood that used to go out to that abnormal lung is probably being shunted back to this lung, which explains the really marked difference in density between the normal central lung, which is probably hyperperfused, and the abnormal peripheral lung. And then when, quote, unquote, mosaic perfusion, it's not really mosaic in this case, when you have airways disease that's very diffuse, you don't have a normal lung to compare it with. And so it can be really difficult. This case looks normal to me. This patient had horrible pulmonary function tests. They were very obstructed, and that was really helpful. So this is actually, quote, unquote, diffuse mosaic perfusion, or at least diffuse airways disease. These can be really difficult. Okay, so in summary, we've talked about the ways that you can distinguish the two different causes of mosaic attenuation, which are ground glass and mosaic perfusion. And the typical features that favor mosaic perfusion are geographic borders, bigger vessels in the normal lung, smaller vessels in the abnormally lucent lung, and air trapping on expiratory CT in patients with mosaic perfusion from airways disease. Obviously you won't get that with vascular disease. Talk about some of the challenges, and the differential diagnosis, which is mainly, if you have isolated mosaic perfusion or air trapping, the main differential diagnosis is constrictor bronchiolitis, asthma, and hypersensitive pneumonitis. Most cases will be constrictor bronchiolitis, and that has multiple different causes. Okay, thanks very much. First we'll start by taking a glimpse at processes that cause cysts, but aren't really diffuse cystic lung diseases. And then we'll delve into a more in-depth case-based look at the major cystic lung diseases that I think are the hardest to distinguish, highlighting helpful unique features. And finally at the end, I'll provide an algorithmic approach to help you narrow the diagnosis when looking at a case. Of course, multiple things can cause focal cystic lesions. Here are some examples, congenital pulmonary airway malformation, post-traumatic pneumatoceles with surrounding contusion. Here's a case of pneumatoceles from barotrauma in a COVID ARDS patient. Of course a bleb with an associated pneumothorax, and your cystic adenocarcinomas. ILDs can also have secondary cystic change, but in this context it's not the primary abnormality. For example, here are three cases of DIP. They show diffuse lower lung predominant ground glass capacities, but just a few scattered cysts associated. And of course, infection can also cause cystic change. The classic example being pneumocystis pneumonia, which usually presents a central ground glass capacity, sometimes progressing to upper lobe predominant pneumatoceles. Tracheobronchial papillomatosis due to HPV infection causes tracheal nodules, seeding the posterior lungs, where there are cavitary nodules. And as a final example, here's a large echinococcal cyst in the left lung with a beautiful water lily sign from the ruptured membrane's layering. Of course, you have to watch out for mimics. Cystic bronchiectasis, you'd have to follow these back and see that they communicate with the airways. And of course, emphysema, unlike cysts, does not have walls and you might see the central dot sign from the pulmonary artery. Now we'll delve into cystic lung disease. I've divided the cases by etiology, starting with case number one. This is a 44-year-old female presenting with two weeks of progressive dyspnea and nonproductive cough, COVID-negative. She has a large left pleural effusion and left retroperitoneal mass and she was transferred to Cornell for biopsy. Focus on her lungs here. I know it's blind-eyed, but. Here you can see on these still images there's a few small cysts throughout the lungs and a large left pleural effusion. Keep in mind, this was a 44-year-old female, so this is really a slam-dunk case for lymphangiomyomatosis, or LAM. Her retroperitoneal biopsy showed a lymphangioma and her pleural fluid was a chylophorex, which was subcutaneous. Her retroperitoneal biopsy showed a lymphangioma and her pleural fluid was a chylophorex, which was subsequently treated. So let's break down this really big word. The lymphatics, lymph, blood vessels, angio, and airways are surrounded by smooth muscle leiomyoma proliferation. The abnormal proliferation of smooth muscle cells, LAM cells, occurs in the lungs and the thoracic and retroperitoneal lymphatics. It actually belongs to the PECOMA family of neoplasms. Now, where the LAM cells proliferate explains everything that we see on CT. They obliterate and dilate the lymphatic vessels, leading to chylus effusions, ascites, and lymphangiomas. They can even have arterial involvement, leading to hemocytosis and hemoptysis, as you see in this rare case of LAM presenting as diffusivular hemorrhage. They, of course, can narrow the airways, leading to air-trapping cysts and pneumothoraces. In fact, 81% of patients will have a spontaneous pneumothorax over the course of their disease. It is a rare disease, usually in women of reproductive age. The etiology isn't known, but hormonal factors likely play a role in pathogenesis and clinical course. It can be progressive and usually presents with chronic dyspnea and cough. Now, what do you expect to see? Small, round cysts with minimal variation in size that are diffused throughout the lungs, only sparing the lung apices in some patients. Usually, the surrounding parenchyma is normal. Here's just another example. Again, diffused cysts of similar size, only sparing the lung apices. On pathology, this is gonna be proliferation of atypical smooth muscle cells, and they tend to stain positive with HMB45. There are some things on your histologic differential diagnosis, one being benign metastasizing leiomyomas. These are hematogenous metastases from benign uterine tumors, usually after a hysterectomy. But these patients are gonna have nodules, which we don't see in LAM. They usually don't have cysts, despite this interesting case I'm showing you here. And they won't be HMB45 positive. Metastatic leiomyosarcoma, the cells will have more atypia, cuz it is obviously a malignancy. These will also have nodules, although they can cavitate and cause pneumothoraces like other sarcomas. Diffused pulmonary lymphangiomatosis may sound familiar, but it's a totally separate process where there's proliferation of the lymphatic spaces within the pleura, mediastinum, and lungs. And it's usually in young adults or children without any cysts. Our next case is a 76-year-old with MGUS and hypergammaglobulinemia. You see, fewer cysts than we saw in our last case, and they're more variable in size. With that history, think of lymphocytic interstitial pneumonia, LIP. Follicular bronchiolitis and LIP are on a continuum of reactive polyclonal lymphoproliferative disorders. With follicular bronchiolitis surrounding the airways, leading to central lobular nodules and air trapping. Whereas LIP has diffuse infiltration of lymphocytes with extensive alveolar septal involvement. What do you expect to see on LIP? As we said, fewer cysts of variable size, often perivascular, and there might be superimposed ground glass. LIP has a huge amount of associations, but just keep in mind the main ones. Autoimmune disorders like Sjogren's, dysgammaglobulinemia, like in our index case, and infections. Our next case is a 53-year-old with Sjogren's, found to have nodules when being treated for pneumonia. In fact, those nodules are actually coarsely calcified, didn't play. So in a patient with Sjogren's and cysts and coarsely calcified nodules, you should think of one entity, and that's amyloidosis. There are actually two main forms of amyloidosis, nodular parenchymal and diffuse alveolar septal. Nodular parenchymal amyloidosis is typically in adults in their 50s and 60s, an asymptomatic, benign, coarse. And cysts are actually rare, except in patients that have Sjogren's. Diffuse alveolar septal amyloidosis is less common, rarely has cysts, and has a far worse prognosis. And the abnormality is sometimes hard to distinguish from other things like lymphogenic carcinomatosis. Sjogren's is associated with AL amyloidosis and reactive lymphoflipid disorders like LIP and malt lymphoma. Patients typically have a localized pulmonary AL amyloidosis, and like we said, the cysts and amyloidosis are usually in patients with Sjogren's. In fact, here's a patient with LIP and amyloidosis. We can guess that the ground glass is probably due to the LIP, nodules from the amyloidosis, and the cysts could be from either. Case four is a 68-year-old with a history of myeloma, presenting with recurrent left pneumothorax. There are multiple cysts, and these nodules with nodulars and masses with actually some calcification. If you're thinking amyloidosis, so am I, and we'd be on the right track, because this patient actually ended up having light chain deposition disease. It's really similar to amyloidosis, but the abnormally folded proteins don't form amyloid fibrils, and they don't stain for Congo Red. It's usually multisystemic, but can be localized. And importantly, 75% of patients either have MGUS or multiple myeloma. So if you see that in the history, that should definitely raise a flag. It, like amyloidosis, can be divided into diffuse and nodular. With diffuse having a much worse prognosis, some people even say it's not important to distinguish it from diffuse amyloidosis, because the course and prognosis is similar. Nodular light chain is more often localized to the lung, and it is important to distinguish this from nodular amyloidosis, because there's a higher incidence of lymphoproliferative diseases, like Sjogren's, and plasma cell disgraces with renal failure. You expect to see nodules in most patients. They're usually not calcified, unlike amyloidosis. Patients usually have cysts. The cysts often have vessels in their walls, but we saw that in LIP. One distinguishing feature that has been described is having at least one cyst with a traversing vessel. Now, I've presented you, admittedly, with a diagnostic dilemma. I told you Sjogren's is associated with amyloidosis, and LIP, and light chain deposition disease. All of them can cause cysts, and sometimes nodules, and it's associated with malt lymphoma, which is usually a nodule. And on top of that, light chain deposition disease and amyloidosis are both associated with plasma cell disgraces. So if you're taking the boards, and it says Sjogren's, and there are cysts, just pick LIP. But in real life, think of all these other associations, and if you have someone with Sjogren's, an enlarging nodule, think of lymphoma. Case five is a 48-year-old found to have bilateral renal masses and hematomas. So what's different about this case is our cysts aren't round anymore. They're kind of elliptical and floppy, and a lot of them are paramediastinal. This is a case of bird hog duvet syndrome. It's a rare otosomal dominant disorder with skin lesions and benign or malignant renal tumors. They have spontaneous pneumothoraces, and like I said, their cysts are usually floppy or elliptical and paramediastinal. Here's just a couple other cases. Again, you're seeing those floppy paramediastinal cysts. And we actually looked at patients with bird hog duvet, LAM, and LIP to try to see if we could distinguish which one they had based on the cyst distribution. And we found that if they had diffuse cysts and the patient was female, it was LAM. If they had lower lung predominant cysts with a disproportionate number of paramediastinal cysts, it was usually bird hog duvet. So why are they elliptical? In bird hog duvet, there's little communication with the airways, so the elastic recoil forces act asymmetrically on the cyst, causing it to be lenticular. Whereas in the other entities I described, it does communicate with the airways, so the elastic recoil forces are symmetric, causing the cyst to be circular. Just to throw in one other genetic disease that can cause cysts, here's a case of neurofibromatosis. And you see the neurofibroma and a couple cysts in the lungs. For our final case, we have a 51-year-old smoker with dyspnea and exertion and airflow obstruction on PFTs. In 2012, she had nodules with bizarre-shaped cysts that eventually kind of coalesced in 2014 to pretty much just look like emphysema. This is Langerhans cell histiocytosis. This has a predictable progression, where you start with ill-defined central lobular nodules that then develop into upper lung predominant bizarre-shaped cysts, sparing the costophrenic angles. And at end stage disease, it can be hard to distinguish from severe emphysema. Here are two companion cases, again showing that spectrum, with patient one really being that beautiful, classic, bizarre-shaped cysts. So in summary, we've looked at a couple things that can cause cystic lesions and did a little bit of a deeper dive into diffuse cystic lung diseases. Now, when I'm looking at a case, first I figure out, are there cysts? So do they have thin walls? And once I've determined cysts are present, are they diffuse on the patient female? I'd favor LAM. Are there a disproportionate number of elliptical paramediosonosis? Go with bird hog duvet and check for renal tumors. Is there a vessel traversing a cyst? Maybe think of light chain deposition disease. Is there upper lobe predominance? Perhaps Langerhans cell histocytosis and check for smoking history, emphysema, those nodules. Is there a posterior predominance? Think of tracheobronchial papillomatosis and check the trachea for nodules. Are there nodules with the cysts? Are they big and coarsely calcified? Think of amyloidosis, but of course include meths. Are they big and non-calcified? Can you add in light chain deposition disease in your differential? Are there central lobular nodules? Perhaps LIP and follicular bronchiolitis should move up on your differential. And finally, are there ground glass capacities? Think of pneumocystis pneumonia, DIP, and LIP. Purposes of this talk, micronodules are defined as nodules that are less than five millimeters in size. And when I first see tiny nodules on a chest CT, first thing I wanna do is figure out the distribution of those nodules relative to the secondary pulmonary lobule. Because if I can figure out what portions of the secondary pulmonary lobule are affected by a disease process, I can really tailor my differential. So that means we need to start by reviewing the anatomy of the secondary pulmonary lobule. So the secondary pulmonary lobule is the smallest unit of lung completely surrounded by interstitium. It's about one or two centimeters in diameter and polygonal in shape, and is regularly or irregularly spaced throughout the lungs. The borders of the secondary pulmonary lobule are the interlobular septa. And it's normal to be able to see one or two interlobular septa on a chest CT. But if you're able to start putting them together as polygons, they're abnormally thickened. In the middle of the secondary pulmonary lobule, the centrolobular region, there's a tiny arteriole and a small terminal bronchiole. Now the other portion of anatomy I wanna review that comes into play with micronodules are the lymphatics. Where are the pulmonary lymphatics? Well, they're in the interstitium. So that means they're in the interlobular septa, those borders of the secondary pulmonary lobule. They're in the subplural interstitium, so immediately deep to the pleura. And they're in the peribronchovascular interstitium, extending from the hila out along the bronchovascular bundles into the lungs. So I keep all that anatomy in mind when I look at tiny nodules on the chest CT. And I say to myself, self, do those nodules look like they're located in the middle of the secondary pulmonary lobule? Are those nodules distributed along the regions where I'll find lymphatics and therefore are likely perilymphatic in distribution? Or are those nodules all over the place relative to the secondary pulmonary lobule? There's some that look centrolobular, there's some that look perilymphatic, and there's no dominant pattern. Therefore, they're randomly distributed. And we'll talk about these three different patterns, what they look like on CT and what they mean in terms of the differential diagnosis. So here we have two patients who have small nodules within the right lung. And notice that these nodules all seem relatively regularly spaced. Notice also that there's a thin rim of spared lung between the nodules and the pleura, between the nodules and the costal margin, between the nodules and the fissures. So that distribution, regularly spaced, little bit of spared lung between nodules and pleura, tells me that these are likely centrolobular in distribution. Because the secondary pulmonary lobule is regularly spaced throughout the lungs, and there's gonna be a little bit of space, little few millimeters of lung, between the centrolobular region and the periphery of the lung. So we already said that centrolobular region contains a small bronchial and a small vessel. So what diseases spread via the small airways or via the small vessels? And there are a lot of them. And the differential for centrolobular nodules is quite broad. The vast majority of nodules, centrolobular nodules, are airway disseminated, or spread via the small airways. And you can have airway disseminated infections of all different varieties. You can have aspiration filling the small airways. Inhalational processes, either particulate matter, silica, co-workers, pneumoconiosis, or something that one has inhaled that incites an immune response. And then I keep in mind that any time that there is dilation of small airways, they're not gonna clear very well. And so you can get impacted mucus, and that can give you centrolobular nodules as well. I also wanna keep in mind the fact that you can have arteriolar dissemination of the process. And there are certain entities that'll give you centrolobular nodules as a result of small vessel involvement. But the vast majority are due to small airway dissemination. Now there are a couple of subsets of centrolobular nodules that I look for because it really helps me focus my differential further. One is when you have branching centrolobular nodules that look like little Xs and Ys. Those are referred to as a tree-in-bud configuration of the centrolobular nodules. And huge percentage of those, 95%, are due to infection or aspiration. And when you have an airway-disseminated infection or aspiration, it makes sense that those nodules are often clustered. They're not absolutely everywhere evenly distributed in the lungs. They're in a clustered distribution. And that comes into play as we'll see in a few minutes in distinguishing them from a random pattern. Another subset of centrolobular nodules that I think is helpful is when you're dealing with ground-glass centrolobular nodules. So here we have two patients in whom there are extensive centrolobular nodules, but they're really subtle. They're ground-glass in attenuation. When I see ground-glass centrolobular nodules, that tells me that likely there is some sort of immune response going on in the small airways. So the patient has likely inhaled something that has incited an immune response, and that tends to give ground-glass within the small airways. That immune response can be triggered by cigarette smoke, which is respiratory bronchiolitis, or it can be triggered by some sort of organic antigen, which is hypersensitivity pneumonitis. I do want to put in a plug to look at MIP images when you're dealing with nodules, particularly these ground-glass centrolobular nodules. Sometimes a ground-glass is so subtle that you're wondering whether you imagined them. And if you look on MIP images, it really can accentuate the nodularity and confirm that you are indeed dealing with centrolobular ground-glass nodules. All right, so that was centrolobular nodules that my slides no longer advance. Let me try it this way. Here we go. So let's move on to perilymphatic nodules. Remember, the lymphatics are located in the interlobular septa, in the subplural interstitium, and in the peribronchovascular interstitium from the hila on along the bronchovascular bundles. And here, as a result, when you have a perilymphatic distribution of nodules, when nodules are depositing along the lymphatics, you get beading and nodularity in all of those different areas of interstitium. So here we have two patients who have peribronchovascular nodularity. The one on the left has really marked thickening and nodularity in the axial perihilar interstitium. And then the one in the middle, the nodularity and beading extends more peripherally along the bronchovascular bundles. That tells me that I'm likely dealing with a perilymphatic process. Here's another person who's got perilymphatic nodules. Notice that the nodules here are butt up against the costal margin. They're immediately abutting the pleura, both along the costal margin and along the fissures. There's no spared lung between the nodule and the pleura. So that tells me that these are not centrolobular, but in fact are perilymphatic in distribution. And then sometimes you can see thickened interlobular septa that form polygons throughout the lungs. And often when you look closely, you'll see that the interlobular septa are just a little bit beaded and nodular and not entirely smooth. That again tells me that I'm dealing with a perilymphatic process. Luckily, there's often nodularity in all three of those areas of interstitium that really sort of clues me in that I'm dealing with a perilymphatic process. So in the patient that has interlobular septal beading, they also have subpleural nodules along the costal margin. They also have some beading of the more central peribronchovascular bundles. All of which tells me I'm dealing with a process of deposits along the lymphatics. Not surprisingly, process of deposits along the lymphatics is often clustered in distribution. So it's not completely symmetric absolutely everywhere. And that'll come into play again in a few minutes to contrast these from randomly disseminated nodules. When I see perilymphatic nodules, I think primarily of sarcoid and lymphogenic carcinomatosis. Those are by far the two most common entities that give nodularity along the lymphatics. I mentioned, and I had silicosis and co-workers pneumoconiosis in the centrolobular differential. Keep in mind that when one breathes in the silica particles, initially they're centrolobular in distribution, but eventually they're picked up by macrophages and move into the lymphatics. And so you can get a perilymphatic distribution of disease. Here are those cases that I showed earlier of peribronchovascular beading and nodularity, of subplural beading and nodularity. Those were all cases of sarcoid. It is helpful to keep in mind, I mentioned that perilymphatic nodules are often clustered. Sarcoid, because of issues regarding lymphatic clearance, often lies within the upper lobes, particularly the posterior upper lobes. Sometimes that helps discriminate, that helps you know that you're dealing with sarcoid. Silicosis, co-workers pneumoconiosis is often upper lobe predominant as well, because that's where the particles get inhaled. There tend to be a little bit more calcifications within the parenchymal nodules than in sarcoid, which is also upper lobe predominant. And obviously, you'll have a exposure history in patients with silicosis. And lymphogenic carcinomatosis, for whatever reason, tends to really involve the lymphatics in the interlobular septa, as opposed to sarcoid, which tends, for whatever reason, to spare the lymphatics in the interlobular septa. So when I see really striking interlobular septal thickening to be able to form these polygons that are slightly beaded and nodular in configuration, I think first and foremost about lymphogenic carcinomatosis. Okay, and then I just wanna finish up with talking about a random pattern of nodules. With a random pattern of nodules, you will see tiny nodules essentially in all parts of the secondary pulmonary lobule. There'll be some that look centrolobular, there are gonna be some that look perilymphatic, and there's no dominant pattern. So here is somebody who has tiny nodules all over the place throughout their left lung. And some of these nodules, when you look closely, look centrolobular. There's a little bit of spared lung between the nodules and the costal pleura. There are some that are very regularly spaced. There are other nodules that seem to be beating the bronchovascular bundles and are along the fissures and against the costal margin, so those look perilymphatic. There are just as many that look perilymphatic as there are those that look centrolobular. This is a random pattern of nodules. Here's another case, same thing. Tiny nodules absolutely everywhere, both lungs, and there's some that look centrolobular, there's some that look perilymphatic. I can't tell you a dominant pattern here. When I see this, randomly distributed nodules, I think about systemically disseminated processes because these tend to be distributed both via the lymphatics and hematogenously. I think about widely disseminated infection, miliary tuberculosis and disseminated fungal infection, histo, blasto, and I think about metastatic disease from a variety of cell primaries, breast, thyroid, renal cell, and melanoma being sort of the top ones. So again, and these, because these are systemically disseminated, they are not clustered. They are absolutely all over the place. Here's this patient on the left had miliary tuberculosis, patient on the right had thyroid cancer mets, and you can see on coronal reformats that patient with thyroid cancer metastases, evenly distributed throughout the lungs bilaterally, absolutely everywhere, don't look clustered. Okay, so that is pulmonary nodules. When I first see them, first thing I do is look to see whether they touch the fissures because if they don't, if there is subplural sparing, then I think I'm dealing with central lobular nodules. And if they are branching and tree and bud in configuration, likely due to infection or aspiration. If they are ground glass in attenuation, likely due to something invoking an immune response, hypersensitivity, pneumonitis, or respiratory bronchiolitis. If the nodules do touch the pleura, then they're perilymphatic in distribution. And if they are predominantly perilymphatic, if that is the dominant pattern, tend to be clustered, I'm thinking about sarcoid or lymphogenic carcinomatosis is my number one and two differentials. If they are not predominantly perilymphatic, there's some that are perilymphatic, but there are others that are central lobular and they are all over the place that aren't clustered, then I'm likely dealing with widespread disseminated infection or metastatic disease. And with that, I conclude.

lung opacities

consolidation

ground glass opacity

mosaic lung attenuation

cystic lung diseases

micronodules

lung imaging

differential diagnoses

lung diseases