Thank you, and good afternoon, everyone. It's really great to be here, and I'd like to thank the ACCP again for inviting me. Hawaii is just such a lovely place. And so, I guess we'll get started with this discussion on ILD, and no relevant disclosures. I'd like to thank my pathologists. We have some great pulmonary pathologists at NYU, and all the pulmonologists and the trainees that has made it so rewarding to be at NYU. So our objectives are to review some HRCT anatomy, and we'll touch on an approach to interpreting ILD. And there are four basic categories that we'll touch on, and each of them has a specific approach. So I think you may have seen this QRS code to enter that for your audience response. Okay. So we'll move on to the anatomy. And the anatomy of the lung interstitium is very helpful for interpreting HRCT. There are two different aspects of it, the axial system and the peripheral system. The axial is along the bronchovascular regions and extends to the secondary pulmonary lobule. The peripheral is the subpleural interstitium that also connects with the interlobular septum. And the interlobular interstitium is also present within the secondary pulmonary lobule. So let's look more carefully at the secondary pulmonary lobule. It's a 1 by 2 centimeter polyhedral structure. And it's fed by a bronchus and artery. And we can see these lobular arteries more in detail in this schematic. And it enters and forms the central lobular core. There's interstitium running along these structures, in addition to present along the entire circumference of the interlobular septum and the secondary pulmonary lobule. In the secondary pulmonary lobule, there are acini up to nine in number. And in the interlobular septum, there run the pulmonary veins and also the lymphatics in addition to in the bronchovascular region. Now in terms of high-resolution CT, we can resolve about 0.5 to 0.3 millimeter structures. And that's why we don't see the interlobular septum in the normal setting. We can only propose where the secondary pulmonary lobule is by the location of an artery approximating and becoming close to the pleura. And this is presumably the lobular artery. And this is the central lobular core. And there are interlobular arteries that separate and divide. We don't see the bronchioles, usually within a couple centimeters of the pleura. And usually the vessels, the arteries don't approach the pleura at all up to one centimeter after one centimeter. And notice how we don't see the interlobular septi, but that's the normal kind of size that we expect it to be. And this is helpful. We also look at the predominant abnormality. And sometimes severe disease can look very similar. So we want to look at the edges of disease. We look for other associated findings, central, peripheral, upper, lower. Lung volumes are really key. Effusions and adenopathy are helpful. And whether the process is acute or chronic helps us form a differential diagnosis. And clinical history, of course, is key. So these are just examples. This is a high resolution CT in which we can do some additional imaging. I think you well are aware of prone imaging. And we do perform it at times. We don't always. Because when we see non-dependent abnormalities, we usually assume that this is pathologic. It's not dependent atelectasis. But prone imaging can be performed. As you can see, the opacities are improved a little bit because there was some atelectasis contributing. But there is persistent abnormality. But in general, we knew the patient already had a little fibrotic abnormality. Here is similarly a patient with a mild dependent only finding. And this is a subpleural curvilinear opacity. And this was shown to resolve on prone imaging. And we perform our prone imaging with low dose CT to minimize radiation exposure so the patient doesn't receive more than one full CT scan. They get a portion of the chest imaged through the lung bases. Okay. And so these are the four major categories that we'll cover in terms of patterns, nodules, signs or reticulation, increased lung opacity or decreased lung opacity, which can be caused by a variety of abnormalities. So now I'll turn this over to my colleagues, Karina. So hello, everyone. So our first case is a 61-year-old man from Senegal presented with subacute weight loss and nonproductive cough. Now his past medical history and past surgical history, there was none. His family history as a brother with hyperlipidemia. He's not taking any medications. He's a never smoker, denies IV drug use. He had moved from Senegal eight years prior to New York City and works in a refugee resettlement program. He notes mold in his apartment and he has no other exposures. In terms of data, his labs were significant for a white blood cell count of 2.9. His infectious workup was negative, notably his HIV was negative and sputum, AFB, stain and cultures were negative along with PCR. Give me one moment. Sorry, multiple screens, so we're just going to scan through here. There's our axial. Moving on to our coronal images. So as you can see, it was pretty significant for diffuse nodules. This is the mediastinal view. So, this is our question. You don't answer during this slide. The next slide will be the polling. But how would you characterize the distribution of this pulmonary nodules? A, central lobular, ground glass. B, tree and bud. C, perilymphatic. D, random. Okay. So please answer. All right. A couple more. Okay. We're at 40. Peribronchovascular, which is correct. Okay. So, just a little thing on this patient. So, he was actually admitted for evaluation for Miller ATB. As we saw before, all those cultures were negative. He actually underwent a transbronchial biopsy, and it was consistent with sarcoid, which is classic with this type of distribution. So, Dr. Koh will now be going through the imaging. Okay. So, thank you. And these are the images. And the answer, peribronchovascular, is probably better to be termed perilymphatic. And I'll go through that on subsequent slides. The main aspect is that the nodules are along the fissures, and the fissures are where the lymphatics are. And also, they're along the bronchovascular bundles. They are peribronchovascular. You can see the nodular study. And notice how there are areas of sparing. It's patchy. It's not a diffused process in that every area, every secondary pulmonary lobule is involved by a nodule in a random distribution. So, we'll go through this more on subsequent imaging. And this patient did have some adenopathy. Some patients don't have adenopathy, but the presence of adenopathy, of course, I think you well are aware, is consistent with sarcoid. So, I'd like to go through this algorithm. And the algorithm is one in which we first ask the question, are there nodules adjacent to the pleural fissures? So, that's the first thing we always do. And if so, we know that nodules are either associated with areas where the blood goes through, which would be random distribution, or primarily lymphatic or lymphatic only. And with random distribution, both the lymphatic areas and every other area in the secondary pulmonary lobules affected here, it's really the perilymphatic disease. So, this is that first step. You want to look for the nodules along the fissures, which can be in the horizontal plane. This one's the minor fissures. This is a major. And we can see it's in the subpleural region because that's where the lymphatics are. So, once we decide, again, that there are lymphohematogenous nodules, we can decide if it's perilymphatic disease. Are the nodules only associated with the areas where the lymphatics are? And in that case, it would be along the peribronchovascular bundles. And that's why we determined in that last case that it was a perilymphatic disease or peribronchovascular disease distribution. And here is just an example of an additional patient with the fissural nodules, the nodules that are emanating around the vessels. You might not necessarily see the bronchi. And if I just saw these nodules alone, I would wonder if they were centrilobular because they look like they're only in the peribronchovascular region, but the clue really is the fissures in the pleura. So, that's why the first step is always key. And then if you see it all in the areas where you expect the lymphatics to be, then it's perilymphatic disease. And, of course, the most common is sarcoid. And the main teaching point is that once you get really severe disease, it's very hard to know what is the cause, but look at the edges or the less affected areas. And you can see the studding along the bronchovascular region, along the fissures. So, this is purely perilymphatic. Sarcoidosis, I think you're well aware, has a symmetrical adenopathy that can calcify or not. If it calcifies, it can be just solid calcifications, but this is termed eggshell, which has been described classically with silicosis. But we see sarcoid much more frequently, so eggshell is what we think. There's usually a lack of effusions, and when we see pleural effusions, we wonder if there's some other coexisting abnormality, or it's not sarcoid. And, of course, with advanced sarcoid, you get fibrosis. So, this is the other differential diagnosis for perilymphatic nodules. It's very nice because there's pretty brief differential diagnosis that we think of. And the second is lymphogenic carcinomatosis. It looks very different in a sense, but notice there are nodules along the fissure. The next step is, are the nodules primarily associated where the lymphatics are? Then it's perilymphatic disease. And I'd say, yes, this is interlobar septi. This is actually beating of the interlobar septi. There are probably some hematogenous metastatic nodules, too, but the primary pathology is lymphatic abnormality centered along the fissures and some peribronchovascular regions. So these are the big two, and it's been described sometimes that silicosis can have a perilymphatic distribution and also can have a centrilobular. So lymphogenic carcinomatosis is a hematogenous disease that spreads to the lungs and then invades the lymphatics, and that's why it's basilar predominant. As you noticed, I showed a basilar image because that's where the disease was most prevalent or severe. Septal thickening is a primary finding, and that helps differentiate from other entities, although you can get a few interlobar septi that are thickened with sarcoid. This is a case of silicosis, just showing how you get the eggshell calcification. This is a classic pneumoconiosis appearance. Notice how there is a calcified nodule in this patient in silicosis. It's pretty rare to get calcified lung nodules with sarcoid, so I find that helpful. So in the lymphohematogenous group, we touched on the perilymphatic disease, but now we'll go to random, which is basically a hematogenous disease that affects every area within the secondary pulmonary lobule. It's not just the bronchovascular region. It's not just the sublural region. It's all the areas in between, and we always think of TB and endemic fungi first, hematogenous mets, papoid thyroid, numerous other causes, and silicosis. And here's just an example. You can see the nice, studding lung, the fissures. You can see the studding here, but it's not only in these areas. It's all over, and notice how it's worse at the bases. There's more blood flow to the bases. That's why this is miliary disease, and the name comes from the millet seed, which is only two millimeters in size. So I don't know if any of you recognize this sign, but I like to show it because it has such an unusual appearance, and this is actually the galaxy sign. The galaxy sign is seen in sarcoid and represents multiple small granulomas that coalesce and is seen in alveolar sarcoid but doesn't necessarily have to be present. Patients with alveolar sarcoid, interestingly, do not oftentimes have adenopathy. So this is another category. We're now into the centrilobular, so the other arm of the nodular patterns in which you don't see any visual nodules. Notice how they're sparing, and that's because the diseases are filling the air spaces in the centers of the terminal respiratory bronchioles that are within the secondary pulmonary lobules. So you can imagine the secondary pulmonary lobules, there are about, you know, multiple terminal respiratory bronchioles that are filled, and this is termed mixed bronchiole or peribronchial disease. It's ill-defined. It's ground glass attenuation. The air spaces around the respiratory bronchiole are affected, and this is hypersensitive pneumonitis, and this was non-fibrotic because there wasn't the findings we see in fibrosis, and we'll talk about that later. Respiratory bronchiolitis can always present with centrilobular ground glass nodules, but usually not that severe. They're usually very mild, upper-low predominant, and I always think of flicker bronchiolitis, uncommon entity, and, of course, Langerhans cells histiocytosis that can be solid nodules, but they're centrilobular, and there is no branching that would suggest tree and bud, and there are some less common entities, and never to forget about adenocarcinoma. Okay, so with that, turn it over to Matt. Okay, so the next case was a woman who was 54 years old. She had a recent onset of polyarthralgias that had been treated with prednisone and methotrexate, and she presented with subsequent fatigue, low-grade fevers, and respiratory complaints, specifically cough and exertional dyspnea. Her family history was notable for rheumatoid arthritis, and her medications are shown. She otherwise had no known exposure. She worked as a therapist in New York. Her labs are shown here, notable for a mild leukopenia and a lymphopenia with an absolute lymphocyte count of 700. Her inflammatory markers were elevated with a CRP of 15 and an ESR of 89, and her serologies are shown. You can see that her ANA was 1 to 640 in a nuclear pattern. Anti-DNA was 44, and rheumatoid factor was 38. The rest of her serologic workup was negative. Her imaging is shown here. And you can see there's basilar predominance of disease, maybe better shown in the coronal. And again, basilar predominance, consolidative changes bilaterally, and also potentially peribronchovascular. I'll show, let's see, here are some selected slices. And the question that will appear on the screen next will be, what radiographic sign was thought to be evident on this CAT scan? Was it A, the halo sign, B, the reversed halo sign, C, the head cheese sign, or D, the galaxy sign? Okay, good, and so the group consensus is correct. The answer was reversed halo sign, which is often thought to be consistent with a diagnosis of organizing pneumonia, obviously nonspecific, as Dr. Kowal will go on to explain, but shown here, sort of the dense consolidation with a lucent interior. And in this case, the patient was given a diagnosis of autoimmune, interstitial pneumonia with autoimmune features. She later has been given a diagnosis of lupus, has been treated with prednisone and hydroxychloroquine with improvement in symptoms and radiology, and is doing well. Thank you. Yeah, so of the other fours, I thought you all got the correct sign, and the galaxy sign against it is the coalescent granulomas of sarcoid that's, they call it the galaxy because it looks like the Milky Way of multiple planets. And then the halo is with angioinvasive aspergillosis. And the HEDGs is with air trapping with ground glass associated with HP usually. Yeah, so in this finding, there is a less dense area surrounded by more dense peripheral soft tissue. And the halos can actually be quite sizable and it's always helpful to look in different planes because it might not be as apparent on one section as opposed to another, one projection as opposed to another. So we are in the category of increased lung opacity, which is very large and broad. And basically it's anything that increases the lung opacity that in the definition of consolidation obscures the vessels, so you don't see the vessels with possible air bronchograms. So you can see here. And this is ground glass. Ground glass attenuation is any increased lung opacity is a descriptive term that is higher attenuation than the surrounding parenchyma but does not obscure the vessel. So this person has a mixed pattern of ground glass and this more peripheral soft tissue. And these areas are almost reverse halo. Some people also call this perilobular and you can see how the secondary pulmonary lobule is here. It's like perilobular density. There's clearance in the center of the lobule. And that's characteristic of organizing pneumonia. So there's a very broad differential, as you well are aware. And we as radiologists are simplistic and just think, well, it could be fluid, cells, tumor or blood. And it helps, particularly in the acute setting and extensive confluent institutional abnormalities such as sarcoid and even UIP, you know, fibrotic change. We always keep in the back of our mind. And tumor is like the big thing that can be the mimicker of any inflammatory disease. And we try and approach it with acute and whether it's some acute chronic with acute, of course, including edema, infection, blood. And we always try and think about acute vasculitis. And then acute lung injury and diffuse alveolar damage, acute eosinophilic pneumonia, particularly in young patients, and HP and drug reactions. These are like kind of mental checklists and we look for clues that might help on the CT. And subacute chronic is pretty broad, but chronic infection, hemorrhage, vasculitis is there. And then we always keep in the back of our mind lymphoproliferative disorder and mucous adenocarcinoma. So that's the way we kind of approach it. And there's kind of more like ground glass from disquieting of interstitial pneumonia. There's some interstitial pneumonias we think of that are purely ground glass. So in terms of histology, the histology is that of connective tissue filling air spaces. The lung is architecturally preserved and that's why we don't see any fibrosis early on, although we've seen cases of organizing pneumonia that never resolve and ultimately progress to a fibrotic appearance. And there's mild interstitial inflammation. So organizing pneumonia, as you well are aware, it can be idiopathic or secondary, but I'd like to mention the halo sign. The findings can be oftentimes waxing and waning. We think of organized pneumonia, we see that, oftentimes peripheral and the differential as we mentioned. And this is just more the classic organizing pneumonia, it looks like a pneumonia. It doesn't have the reverse halo, but it fails to resolve. There's no cultures that come back and so we always think of that. In terms of the reverse halo sign, I think we touched on that, but these are just examples of how varied the reverse halo sign can appear. There's ground glass centrally, there's dense consolidation, it should be about around three quarters of the circumference. And in terms of reverse halo sign, the differential diagnosis is exceedingly broad, but we always think of organizing pneumonia, but fungi, including mucor, paracoxia, mycoses, blastoma, mycoses, infarct, that's something we always keep in the back of our mind. Someone in the acute setting, someone comes to the ER with chest pain. If we see a couple of focal, only peripheral wedge-shaped abnormalities, we think of infarct. And sarcoid and TB can have been described, there are more nodular areas in the center. There are some papers that nicely went through and characterized some of the imaging findings of the reverse halo sign and these additional findings, including cancer. And here's just another example of a patient who had organizing pneumonia and this resolved five months later, completely cleared, but this is a reverse halo sign and you can see how it's much more apparent, I say, like in terms of the degree of consolidation on this coronal image. And this is just a case that I've mentioned before, but there are some findings that can lead to one wondering about malignancy. This is actually what we would term maybe consolidation, maybe. You can see there's some air bronchograms, but what's worrisome is there's just a relative lack of air bronchograms in the middle of it. But what's interesting is you see this perfectly preserved vessel. And so this is something that's like filling the air spaces. It's not necessarily invading that portion, but notice here how there's chest wall invasion. And actually, I'm not showing you, but this patient actually had a pancreatic mass that was only imaged like a little bit of it, but putting the two together, we suggest abdominal CT bits. This was metastatic pancreas, mucinous adenocarcinoma, teaching point, mucinous tumors such as mucinous adenocarcinoma of the lung or metastatic disease, gallbladder, pancreas, GI, we've seen present as chronic infections. But always look for aggressive features and that right chest wall invasion here is really what points out. So I guess I'll turn it over to Karina. Thank you. All right, moving on to our third case. So we have a 66-year-old woman with a diagnosis of asthma and chronic dyspnea referred to the pulmonary clinic for evaluation of an abnormal chest imaging. In terms of past medical history, again, she carries a diagnosis of asthma, type 2 diabetes, CKD, medication, she's on fluticasone, salmethyl, insulin, social history, she's a never smoker, lives in New Jersey her whole life, she's a retired secretary and her pet is a poodle. Thank you for watching. So as you can see, there are cysts throughout both lungs. Alright, so the question is, what is the most likely diagnosis? A, central lobular emphysema? B, LIP? C, LAM? Or D, pulmonary Langerhans histiocytosis? All right, a couple more. We got to 45 last time. One more. There we go. Thank you. Okay. So that is correct. So the answer is Lamb. So this patient was actually started on serolimus. She is on chronic oxygen therapy and is currently undergoing evaluation for a transplant at our center. Okay. So this is a nice example of cystic disease. And there are these round lucencies that have a perceptible wall. And you can draw a pencil around it. So that helps differentiate from emphysema. And we look at the pattern of cysts in terms of upper and lower. And we can see that it's pretty evenly distributed, perhaps a little bit more lower, but I would say even. And both central and peripheral areas are affected. There's no pleural fusion or pneumothorax, which is what we typically look for. And this was consistent with lyomyomatosis given the equal distribution, the evenly distributed cysts in this patient. So we are into the area of decreased lung opacity. That actually is huge. And I'm going to focus really on the cysts and emphysema. There's actually mosaic perfusion and mosaic oligemia. That's a whole lecture discussion in itself because it's oftentimes a point of confusion, even for all of us still. But this is a category of decreased lung opacity. So just to go through some definitions, there's a glossary of terms that we utilize. And it was published by the Fleischner. And so what is emphysema? Well, it's usually, once you have Ebola, we usually call it Ebola when it's over 1 centimeter. And emphysema, if it's less than 1 centimeter and it's abutting the periphery of the lung, that's when we call it paraceptal emphysema. Paraceptal emphysema affects the distal portion of the secondary pulmonary lobule. Within the center of the lung, it's hard to really know whether emphysema is truly in the distal portion of the secondary pulmonary lobule or central portion. But if we're at the periphery of the lung and it's in the periphery of the secondary pulmonary lobule, it's much easier to ascertain that. The wall should be less than 1 millimeter. So we should be straining to see if it is existent. So you'll see there's an overlap between that and the cyst, in which there is a defined wall. It's usually between 1 and 2 millimeters. And it's isolated and not associated or clustered with a bunch of other cysts, although there are exceptions. Honeycombing. Honeycombing is typically less than 1 centimeter. The walls are thicker. And the main point of honeycombing is they're stacked upon each other. So you can call honeycombing if it's only one row. There's debate about whether one row is good enough. But the cysts stacked upon each other doesn't have any normal intervening parenchyma as opposed to traction bronchielectasis. And we'll go through that in the fibrotic category when we have our next case. Central labyrinth emphysema has an imperceptible wall. Sometimes you see that central core or the vessel surrounded by the leucine C. So there's a basic four that we think of in terms of cystic disease. There's some less common entities. And these lists aren't to be exhaustive. But we usually think of LAM and LCH, pneumocystis, pneumonia, or PJP, Sjogren's, and LIP. And the main thing is, is that we want to exclude these mimickers, emphysema bronchiectasis honeycombing, which is fibrotic disease and cavitary disease. And sometimes it's hard, particularly when multiple processes are superimposed. We look at, for these features that we kind of went through, and in that case I didn't mention there was a lack of nodules, although we'll touch on that. So in this case, we can see that these cysts are all kind of evenly distributed, monotonous. They have a perceptible wall. But notice how some of these are a little bit challenging. That's why clinical history is really helpful. This is when the radiologist will look in the chart to see if there was tobacco history to be confident. But this is like a nice wall. So we say that at least there's some form of cystic disease that's present. And this was lymphangioliomyotosis. There's no nodules. And this is another example of lymphangioliomyotosis, or LAM, we'll call it. You can see how the patient progressed in seven years. This patient had tuberous sclerosis, and you can see these small interspersed lucencies. And it almost looks like emphysema. And the distribution is a little helpful because it is monotonous in the coronal plane. It's upper and lower equal. If this was mild emphysema, you would expect just the upper lobes to have a little emphysema. So that's the clue. This patient had a pneumothorax. And this was the patient who progressed. And notice all these ground glass. Now this ground glass could well be inflammatory. This was seven years prior. But these areas were persistent and was related to her tuberous sclerosis that she had. So what we did is to try and look at the kidneys for these large fatty densities. This is extreme. But this is a mega angiomyolipoma comprised of macroscopic fat. So it almost looks like retroperitoneal fat. There's subappendable nodules. And they're quite frequent, even in sporadic LAM. And you can see sclerotic abnormalities with tuberous sclerosis. So these three are particularly associated with tuberous sclerosis, but the angiomyolipomas with sporadic LAM. This is multifocal pneumocyte hyperplasia, which those ground glass opacities were actually related to on that patient's CT presumably. And this was actually a nodule that was not present, developed and grew. And this individual, it was just one finding. And she had some small cysts. So she well had a history of LAM, this individual from tuberous sclerosis. She had tuberous sclerosis. But this nodule was growing and this was biopsy and path-proven multifocal pneumocyte hyperplasia. And so our rule of thumb is that for women, they tend to get cysts. Men with LAM don't tend to get cysts. And even tuberous sclerosis, they can with tuberous sclerosis men. But the nodules are present in men with tuberous sclerosis more frequently. But you can see nodules in both. So that's the teaching point that I've learned is that the more cysts you have, the more likely you'll have nodules and the more number of nodules. So that's multifocal pneumocyte hyperplasia. It's type 2 pneumocyte proliferation that's related to the tuberous sclerosis gene. So this is just a case that I like to bring up because, you know, this brings to the question, is this emphysema or LAM? This is kind of a challenge in a sense. But this actually was LAM. And it almost looks like emphysema because it's upper-low predominant. And I would wonder if this is LCH, which is smoking-related upper-low predominant. And I guess the take-home points that I learned from this case was that once you get pretty severe cystic disease, then everything starts looking alike, even with cystic disease. This actually was LAM, but just happened to have upper-low predisposition and was very severe. This patient was being considered for lung transplant. This is another patient who was pre-transplanted. You can take a look and kind of think about what you think it is. This finding is the abnormalities are lucencies. And it's hard to know, is this emphysema? But it seems like there's areas of confluence. And there's a lot of ground glass. Could this be emphysema with some respiratory bronchiolitis, superimposed? Well, are all of these emphysema? But notice how they're kind of irregular. And this is the clue. When we see irregular-shaped cysts that are very confluent, we start thinking of Langerhans cell histiocytosis, which this was. And this is pretty end-stage. This patient also was on the transplant list. And notice how the abnormality is so confluent. And the lucencies almost look like confluent emphysema. But there's just too much opacity, so there must be another process superimposed. Regardless, notice how clean the lung bases were. So this is an earlier phase of Langerhans cell histiocytosis. And I mentioned that it can present with central artery nodules. And this case is just marked because you can see how the nodules have cavitated. And these nodules then become coalescent and burnt out. And so that was end-stage LCH. This is earlier phase. And those nodules can resolve with smoking sensation. We've had cases who had biopsy-proven LCH, and then they stopped smoking and it went away. And so this is an entity that we encounter in patients with Sjogren's. And this is LIP. And the main aspect about this is that the cysts are very varied in size. And you notice how those prior cysts were all kind of monotonous and equal in size. Now, whenever I see varying cysts, I always think of enumerative seals, but I also think of lymphocytic institutional pneumonia. And also the presence of nodules is something that accompanies LIP in addition to LCH, and I guess potentially LAM, but it's not as common. So this is LIP, which we see in Sjogren's, but we also have seen in other immunocompromised patients and RA patients. So this is a case of a patient who presents with a large amount of ground glass, and there are cysts. And this patient was acutely ill, so I think that's probably the most helpful, but this was cystic PCP or PJP. And the cysts eventually develop, they can lead to pneumothoraces, as you well are aware, and can resolve or leave a significance quo. And this is just an extreme example of cystic PJP, someone who just had such severe disease, but they can become very thick-walled. I don't know if it's cystic right now or cavitary at this point. And this is just a less common entity for cysts. It's not on the list because it's so rare, but I think there is a telltale finding or clue is that all these cysts almost look like paraceptal emphysema. And there's no central lobular emphysema, that's the perplexing part. So that's when we start thinking, well, maybe this isn't just emphysema. And notice how they're all medial, they're peripheral, and they're along the fissures. And this is Burt-Hogg-Dubay. And this is actually a renal oncocytoma that occurs with Burt-Hogg-Dubay. It's an autosomal dominant, it's very rare. And there's a germline mutation in this folliculin tumor suppressor gene. And there's always some skin manifestation, cystic lung disease, a high proportion. And unfortunately, these patients do sometimes get chromophobe renal cell cancers, but the oncocytomas are benign. And there are some other tumors. So this is an approach that's been described nicely. And the first part to approaching what we think is cystic disease is to figure out if you really think it's truly cystic, and whether it's emphysema, with some superimposed process rather, honeycombing from fibrotic lung disease, or bronchiectasis, or whether it's just a cavity for infection. If it's a true cyst, then if it's solitary, I usually think of pneumatocele, trauma, lung acerations, and bronchogenic cyst, intraparenchymal. And if it's diffuse, if it's ground glass, you, of course, think of these two. You can get airspace enlargement with DIP. With nodules, of course, these are the two we think of. You can get with LAM, as I showed you, with tuberous sclerosis, amyloid I didn't touch on. It's not very common. It's so rare. And late chain deposition disease. Papillomatosis is something that can become very cystic, but oftentimes there's a thicker wall. And then, lastly, if there's nothing, we think of LAM, BirdHog, DuBay, LIP, which can lack cysts, lack nodules, but LIP also, most importantly, presents with ground glass. So the cysts are only just one manifestation or presentation of LIP, and then LCH. Okay. So our final case was a patient who was presented at our multidisciplinary discussion group a couple of times, actually. It was a 73-year-old woman. She had a 25-pack-a-year smoking history, had quit 20 years prior to presentation, and had carried a diagnosis of COPD, who developed acute onchronic exertional dyspnea and was referred for a second opinion based on abnormal imaging. As stated, she had only a medical history of COPD. She was controlled on inhalers. And she had a sort of recent interesting exposure history in which, during the course of a move to a new home and a renovation of that kitchen, she had been exposed to a large amount of dust. She had a dog, which was a Maltese, and some mold exposure in the course of the move as well. Otherwise, no identified exposures. Her labs are shown here, notable for a mildly elevated ESR at 42. She ended up undergoing a broad serologic workup that was unrevealing. Her hypersensitivity to pneumonitis panel was negative. She underwent a bronchoscopy, and the BAL fluid was notable for lymphocytosis of 32%. The CD4 to CD8 ratio was 2.8. And the transbronchial biopsy revealed nonspecific focal fibrosis. There were no granuloma seen, and special signs were negative. Our imaging is shown. As you can see, there's sort of areas of fibrosis. It's peripheral and axillary distribution. Associated with that, there's areas of ground glass and consolidation. And again, in the coronals, we're appreciating much the same sort of upper and lower distribution of reticular changes with associated ground glass and axillary distribution. So the question is, what is the best characterization of this imaging? Is it typical for usual interstitial pneumonia, probable UIP, indeterminate for UIP, or is it an alternate diagnosis? All right, and it was, you can see the group is mixed and probably as were we, the correct answer that was decided on after sort of extensive discussion in the multidisciplinary discussion group was that this was most consistent with indeterminate for UIP rather than an alternative diagnosis, although there probably is an equally valid case to be made for both. There's obviously significant ground glass, which is not consistent with UIP. It's upper and lower predominant, which is not consistent with UIP. There was some evidence of bronchiectasis, no honeycomb cysts. However, it wasn't felt that the imaging was classic for any definitive other alternative diagnosis and that's why indeterminate was decided upon. I'll let Jane take over. Okay, yeah, so just to go over, I'd like to point out these small lucencies that probably indicate traction bronchielectasis and we'll show some other examples but they're basically like dilatation of the terminal first-group bronchioles, traction open. And notice how these lucencies are in the area of reticulation ground glass. So that's helpful because sometimes you can run into kind of emphysema and to try and figure out whether it's emphysema or traction bronchielectasis, it's just helpful to see the reticulation ground glass in that area. And again, it is upper and lower and central and peripheral, which really doesn't fit at least the probable or the consistent with their typical UIP. So we are in this reticular category and there's a broad differential, but we're going to focus on the fibrosing entities. And so this was a case of fibrosing HP, which you well are aware is caused by organic agents or small particles. And if we look for the bronchielectasis and honeycombing, then we call it fibrotic and then otherwise it's non-fibrotic. In the case that I showed of the central upper ground glass nodules, that was a case of HP. Sometimes patients, however, may just present with patchy ground glass opacity such as, and it becomes very difficult because it almost looks like heterogeneous lung attenuation, mosaic attenuation with no clues that there's ground glass nodules. And the pathology is just inflammatory and there are granulomas that are present. And this is just an example of the challenge of hypersensory pneumonitis. This person was a bird fancier and this is what we would call heterogeneous lung attenuation or mosaic attenuation. And if you notice that it's a patchwork of quilt-like pattern, you can imagine the secondary pulmonary lobules here, colo-west. And so when we see that, we call it a heterogeneous lung. And then we have to decide, well, is this the pathology or is this the pathology or this the pathology? So then you first look at the vessels in the dark areas. So I find the darkest area, I look at the vessel, if it's attenuated, then I know there's at least a component of decreased blood supply to this area and it could be from air trapping or from chronic PE. So we know there's a component of that. And then I look to see the number of attenuations within the lung. Is it just air trapping and then normal lung? If it was normal lung, all of this should be homogenous. If it's more than one attenuation, then you know there's some ground glass. So this is what's called mixed attenuation, mixed air trapping and ground glass. And this is what's been classically called the HEDGES sign, which I think you may have heard. It's also been called the three sign because of the three attenuation signs. So this was actually accentuated on expiratory imaging. And this is nice because you can see how this failed to increase attenuation, this increased attenuation. This is ground glass. And this was area of air trapping. But look here, you can see that this area was actually normal lung. It increased in attenuation. So this was air trapping, which has been described with fibrosing and subacute. This person didn't have traction bronchoelectasis. So this is probably severe, I guess, intermediate going to fibrotic HP. Because I don't see bronchoelectasis, but you see all this reticulation. Maybe there's a small area here. So upper lobe involvement is helpful if present, if it's ground glass. Oftentimes the fibrosis with fibrosing HP is basilar. So it mimics IPF. So the first step, as I mentioned, is to identify if there's architectural distortion. And the best way is to look at the characteristics of the reticulation in terms of whether it's coarse but also, most importantly, if you see traction bronchoelectasis, as you can see here. Here you can see the bronchus that looks dilated and it's not a straight tube anymore. So we know that there is fibrosis. So that's the first thing is to establish because other things, such as acute infection, can mimic fibrosis if we don't pay attention to these areas of distortion. So here is traction bronchoelectasis and bronchoelectasis here. There's some normal intervening lung. You know, the cysts are kind of spaced evenly. They're not all clustered. Whereas here, I would say they're getting clustered and this is an area of honeycombing. So hopefully that shows a little bit of a difference. They're just kind of more extensive and less extensive disease. So we establish that and this is just the honeycombing to show how it's multiple cysts stacked upon each other and it's multiple rows in general. In terms of UIP, we look for honeycombing, but there should be reticulation in ground glass that's a gradient from upper to lower. If you only see lower lobe, that's unusual for UIP. The upper lobe should be involved, although there's a gradient and this was shown to be a good sign. In histopathology, the appearance on CT kind of parallels the pathology. There's temporal and spatial heterogeneity. That's why there are areas of lung that are relatively spared. And then there are bass or fibrotic changes, whereas the upper lobes may have no honeycombing. And this is just a fibroblastic foci that you probably are well aware of. It's part of the diagnosis of UIP and this is a subplural fibrosis that is one of the criteria. So in terms of the categories, and this goes back to our audience response, the four categories are whether it's fibrosis. So that's why we went through that traction bronchiectasis and bronchiolectasis. And then we look for honeycombing. Is it basal, subplural? And there's no features to suggest alternative diagnoses. And that would be extensive ground glass that would suggest an acute exacerbation of UIP, DIP would be here, consolidation would be organizing pneumonia, extensive mosaic attenuation, a lot of air trapping, that would suggest HP, and nodules and cysts, of course, would be another, like cystic lung disease, or SOCA. So if these are satisfied, it's typical UIP, probable UIP, and I think all of you knew that it wasn't typical or probable. It's the same, but there's no honeycombing. Indeterminate and non-IPF diagnosis or alternative diagnosis is kind of a little bit of a gray zone and there's inter-observer kind of variability. But I tend to think about it as it looks kind of like UIP, although there are findings that are a little odd. It's central peripheral is involved. You don't only have that peripheral involvement. It's a little bit too much ground glass. And that's when I think, oh, maybe the patient had NSIP or maybe a connective tissue-related UIP. I usually reserve this for something that's like really kind of suggestive of something else. But I think there's a gray zone. So I think, you know, the fact that it was kind of almost equally divided is not surprising. So this is someone who has traction bronchiectasis, bronchioectasis. Notice the peripheral nature of the ground glass. There's reticulation that can be visible in addition to the ground glass, which is probably microfibrosis. And notice how it's basilar predominant. So this all is consistent with probably UIP pattern because we don't see honeycombing. Now this patient had no honeycombing and notice how peribronchovascular was. There's no upper lobe involvement. And this would be something that is kind of weird, it's for UIP from IPF. This is something that needed to have multidisciplinary determination. It's on pathology. The patient went to biopsy because this was an indeterminate or potentially alternative diagnosis. It wasn't UIP on CT and it was UIP. So this is just unusual kind of presentation. The patient didn't have connective tissue-related UIP. And it was shown in paper that for each of these categories, you all could have IPF. Of course, going down, percentages get lower that the biopsy will be UIP from IPF and UIP. So the alternative diagnosis still had a small proportion of patients that proved to be UIP. So this is just upper lobe fibrosis. Of course, we think of sarcoid and pleuro-parenchymal fibroelastosis. And I like to just point out how PPFE is not so common. But it has this really dense apical cap that actually has such a sharp transition to normal lung. There are cases in which there's associated UIP. So you can have PPFE and coexistent UIP. But the dense apical nature raises possibility. And the perihyal nature of the fibrosis here should always bring up to the late sarcoid or silicosis. And I like to just finish up with this last case because this is something that we've been encountering and it was described by Chung and colleagues more recently. It's what's been called the straight edge sign. And notice, remember the UIP from IPF, CT, it's a gradient that gets worse. It's almost like this is spared and this is severely affected. This is the straight edge, I call it flat table, straight edge sign. And you can see how there's a lot of honeycombing for relatively not so much ground glass and reticulation. So this can be considered maybe a form of exuberant honeycombing. I've seen much more extreme cases of exuberant honeycombing. But this is probably an example because there's such a relative lack of ground glass and reticulation. And here, this is what's called the anterior edge sign in which the rest of the lung looks pretty good, just the anterior aspect affected. And it's been associated with connective tissue-related ILD, these three signs. Just to mention, the straight edge sign has been associated with NSIP and is perhaps more frequent in NSIP than UIP, such as in patients with scleroderma. The UIP patients tend to be RA patients that we've seen. A lot of the fibrosis that has like, the RA has more likely UIP. So with that, I'd like to thank you. I think we covered the secondary pulmonary lobule anatomy. We discussed kind of how to approach different presentations and patterns on HRCT. And I guess the main thing is to just start by first categorizing the dominant pattern. You will always have maybe some ground glass opacities mixed with reticulation. And so you have to just kind of figure out which one is the major pattern to start down an algorithm. So with that, thank you. Thank you all. Thank you all for coming. Thank you.

HRCT

interstitial lung disease

nodules

reticulation

increased lung opacity

decreased lung opacity

sarcoidosis

hypersensitivity pneumonitis

CT images