Thank you guys for coming today to our talk about interstitial lung abnormalities. I'm delighted to have this presentation with Dr. Chung and Dr. Bueno. Two quick housekeeping things to mention. I may break in and out a little bit. I'm getting over a cold that my preschooler brought me, so I have gone through puberty, but my voice may squeak a little bit. And then second, what is probably a first for CHESS, one of our speaker's airplanes got struck by lightning before takeoff. She's fine, but she will be delivering her PowerPoint via video rather than being in person. So please deal with some minor technical problems. So these are our learning objectives. We'll discuss what ILAs are, why they're relevant, and what are risk factors for clinical evolution. And then we're going to go through a lot of cases in different imaging. This is our faculty. Dr. Bueno is coming from the University of Virginia. I'm Sean Callahan. I'm at the University of Utah, and Dr. Chung is at the University of Chicago. So this is me and my disclosures. All right, let's get into it. What are incidental interstitial abnormalities? If you look through the literature, there are quite a few different definitions of what ILAs are. This is the one that I've decided I like the best, I think is probably the most pragmatic, the easiest for implementing in bedside care, which is that they're subclinical, so patients not endorsing symptoms. They're high-density interstitial abnormalities, and they're being found just by coincidence, so on a CT scan looking for other things, so very often seen on CT abdomens, for example. The criterion are as below, so non-dependent, not involving something that could be atelectasis. They have to include more than 5% of any lung zone, and then they need to include cysts, honeycombing, traction bronchiectasis, reticulation, or ground glass. I made a boo-boo, and there's cysts on here twice. So I don't know if you guys are aware of this. I certainly wasn't aware of this. I wasn't taught this in med school. I feel like I went to a good med school, and I never heard of this. It's because it's new. These have formerly been recognized in the white paper from the Fleischer Society in Lancet Respiratory Medicine two years ago. They used to be called many things. I think all of us are pretty familiar with who these patients are. They get a CT scan. You see something on there that you weren't expecting. Oh, crap. What do I do now? That's what these things are. They're referred in older literature as early ILD, preclinical ILD, or subclinical ILD, and now we just call them interstitial lung abnormalities. One reason to be up-to-date on this is this is a very active research field. This is a quick snapshot from PubMed from a couple weeks ago that I took just looking at ILAs, and there's already 83 results, and you can see the slope of research articles has gone up drastically in the last couple years. One thing I think that's very important to note right at the very beginning is that this can mean a lot of things. So this CT is the one that I showed earlier where a woman who had a CT abdomen that showed some minor abnormalities here on the left, so some ground glass and reticulation. That's really pretty minor. And then this is a patient who's here on the right who's got pretty marked fibrosis. This person has an interstitial lung disease. This is pretty advanced fibrosis. So it's possible, though, that this person on the right could still be classified as having an ILA, right? They're just subclinical. I work at the VA. I have a lot of patients who drive four or five hours to see me in clinic. They could have a gangrenous foot and florid septic shock showing up to my clinic, and they may not endorse any complaints. That person could have an ILA very easily. So you may have patients like this who have really abnormal CT scans, and by definition, because it's a broad term, could technically have an ILA. One thing to be cognizant of, though, is we're getting really good at identifying these and trying to delineate what's what from now on. This is a very recent study from the Blue Journal analyzing patients with ILAs and COPD gene. So patients with COPD looked through and found that about 10% had what we would classify as ILA, using the definition I provided earlier. If you take those 10% of patients and then give them to ILD docs and some what I think are fairly conservative criteria for diagnosis of an interstitial lung disease, like half of these people ended up having what we would recognize as an interstitial lung disease significant enough that it would perk our attention, spark our attention. All right. Keep everyone awake. We're going to do some audience response questions. So first audience response question. Which of the following is not associated with higher rates of ILAs? If people could pull up the app and choose from here. So your options are advanced age, asthma, prior smoking history, and elevated BMI. Okay, we have a good mix of things, so most people got this right, so 62% asthma. This is something I would have gotten wrong prior to reviewing the literature on this. So advanced age, prior smoking history, and elevated BMI are all risk factors for developing ILAs, and the BMI is an interesting one that I think I would have picked as well. So let's just talk about epidemiology for a quick second. These are incredibly common if you go looking. They're not on the same level as finding a nodule in a patient, by example, but they're really common. One thing you guys should be aware of, though, is when you're looking at the literature on ILAs, be a bit specific in what population you're looking at, because it's going to vary based on the population. So in general, we see these at about twice the rate in smokers, so if you're looking at a low-dose cancer screening CT scan cohort, for example, you're going to see higher rates of it, usually. And then some cohorts are going to be of patients of very advanced age, and so some examples of these, the Reykjavik study that we'll cite a couple times, those patients are of very advanced age. You're going to have higher rates of ILAs. And then patients with COPD or lung cancer screening, they have higher rates, and that's opposed to studies in which patients are just all comers. So why do we care? It is common that we mention this to trainees or community pulmonologists, and people are like, you ILD docs are just coming up with new terms, like some more abbreviations that I got to remember. This is crap. Worse really matters. So if you have somebody who has an ILA, the chance that they're going to do worse is pretty significant. I've stuck in a few different studies throughout this just to denote the importance of this, and this is the Reykjavik study I mentioned a minute ago. But 10% of patients had ILA on at least one CT. Three-quarters of these people had progression of disease when they went back and re-CT'd. So they didn't just stay put, they got worse. And overall, these are associated with higher rates of mortality, too. This is a pretty neat study from Thorax earlier this year. It was somewhere in Europe. And like I was talking about in an earlier slide, a lot of these patients with ILAs, if you look really hard, it's not going to be just this incidental minor thing. A lot of times, it's actually a true ILD that you're picking up. So in about 2.5% of the patients that were getting these low-dose CT scans, they had the ILD team look at the cases, and they weren't really incidental. These patients had symptoms, like were just not great at examining or listening to patient symptoms. So the majority of these people had crackles, cough, dyspnea. The median DLCO was 64%, so borderline moderately reduced. And half were diagnosed with progressive fibrotic lung disease. Quarter of patients were started on either immunosuppressants, antifibrotics, or a combination. So if you look hard and think about it, a lot of times, these people are going to have just a true ILD, so not something incidental and minor that you should look away from. So I'll mention this briefly. This was a great study from JAMA about six years ago, in which these four groups I've already alluded to a little bit, so the Framingham Heart, COPD Gene, Eclipse, and Reykjavik studies, found that people with ILAs as opposed to patients without ILAs have higher hazard ratios for death over a time frame, and I think it was five years. One specific population to call out and pay attention to would be smokers. This is something that's repeated in nearly every study of ILAs, is that smokers do worse. This is a study looking at lung cancer screening patients. So these are our patients, even though it's in Denmark, these are our patients, like they're 50 to 70 years old, they're current or former smokers, and their mortality is higher if they have ILA. So it doesn't matter whether they're fibrotic features, inflammatory features, mortality is higher. Interestingly, it's not just cancer, and it's not just lung cancer, so it's all comers. It's all comers. All right, before we talk about specific features that you should pay a lot of attention to, let's do our second audience response question. I'll give people a little bit more time this time. Which of the following is a radiographic feature not associated with ILA imaging progression? So non-emphysematous cysts, abnormalities predominantly located in the lower lobes, traction bronchiectasis, central lobular nodules, or subplural reticulation. All right. Good job, guys. So central lobular nodules is the correct answer, so about 57%. And we have a nice mix. We'll talk about what are some good and bad features. So in general, one way to think about these is if you have ILAs that look just straight up convincingly scar, that's a bad feature. So obviously, things like honeycombing, traction bronchiectasis, or scar, those patients are going to do worst. And we'll see some examples from Dr. Bueno and a little bit of non-epizematous cysts and what those look like. Conversely, a good finding is if you have central lobular nodules. And the thinking is that those are often patients who have either an infectious component or more likely respiratory bronchiolitis. So your patient's getting low-dose cancer screening. If they have central lobular nodules, it might be because they're smokers and it's respiratory bronchiolitis. One thing to note about the data here on the right is a study from Putnam in 2019. You'll see conspicuously the odds ratio for progression is blank for ground glass and for honeycombing. Ground glass is blank because everybody had ground glass almost by definition. These people who have ILAs are going to have some form of ground glass. The more notable finding on this is the lack of data on honeycombing. And the reason why there's no data is absolutely everybody who had honeycombing had progression. So if you have somebody who has ILA and honeycombing, it's going to progress. That's pretty clear from the literature. Not only is it helpful to know that the patient has an ILA when talking to them about prognosis and whether they're going to get worse, it's helpful to look at sequential time points. So if you guys look at the guidelines for what to do on ILAs, they'll often say get a repeat CT scan. The idea is it's more information. And if a patient has sequential worsening of their ILAs, they're going to follow that track. They tend to do worse over time. And that corresponds with higher mortality, like we saw. More symptoms, dyspnea, cough, their PFTs decline, and then diminished exercise capacity. So those are the radiographic things to look out for. What are some of the clinical predictors? So before you've even looked at the CT scan, sorry guys, this is something that we get dinged on. Men have worse progression of, more likely to have progression of their ILAs, older age, and then smoking history. So not just active smoking, but previous smoking. I did not know this, and I mentioned this earlier, but having elevated BMI is a risk factor for higher rates of ILA progression. And it's not a binary thing. So the more weight, the heavier you are, the greater the risk is of ILA progression. And then like in a lot of our interstitial lung diseases, a MUC5, B promoter polymorphism, is a risk too. Okay, nearing the end of my part. I just want to see the CT scans, Callahan. Okay, we'll get there. Before we go through, just a quick reference to the Fleischner Society guidelines. With all this data in tow now, a good way I think of thinking of this is, how should I think of these packages together? I think the first step is really risk stratify the person. So risk stratify them based on their clinical features. Is this an elderly patient, are they a former smoker, are they heavy, et cetera? And then what is it about this person's CT scan that is concerning or not concerning? Concerning features being there's more scar, the worse in the lower lobes, and then particularly if they have honeycombing or traction. If you're concerned, repeat the scan. You know, we can quibble about when to repeat it and look at the guidelines for it, but repeating it provides you a lot of information. So if they're worsening, they're going to keep on worsening in all likelihood. And then take the proverbial pulse of the situation. So is this really just an ILD? So one thing when I was looking through the data for this is, if you go looking really hard and applying some conservative metrics, a lot of these patients are going to truly have an ILD. It's just subclinical. So if they have a true ILD, maybe you should send them over to see the ILD docs. Conversely, if it's a minor finding, they don't have symptoms, they don't have impaired PFTs, you have time. You can probably wait, provided there aren't these horrible risk factors. Thankfully we have some guidance. So the Fleischner Society position paper that I put up earlier does have this nice algorithm to follow that I think we can quibble with certain parts of it, but overall I think it's quite good. And it proposes steps for workup and when to repeat imaging. When in doubt, really it's fine to send these people to see an interstitial lung disease physician. That's what a lot of these studies have done, is end up referring people over for an expert evaluation. I often tell my fellows, it may not be that I do anything differently than what you guys are doing, but at least we'll all feel better about it. And then we can discuss them at ILD conference and decide are we going to initiate treatment or not. Okay. We're at 20 minutes. I think we can save questions for the end unless people have burning questions. I'm going to attempt to play this video from Dr. Bueno. Again, I apologize if the sound is off, but we'll give it a shot. Hello, everyone. I'm Juliana Bueno. I'm sorry I cannot be here in person and I appreciate the opportunity to record this session. I hope it doesn't take from the great session that I know Sean and Jonathan are providing. My part of this topic is interstitial lung abnormalities and the imaging features. I have no disclosures today. What we'll do in these 15-20 minutes is review the definition of interstitial lung abnormalities, define those imaging features that characterize each of them, and identify the imaging features that define progression of interstitial lung abnormalities. As we've mentioned before, these abnormal findings are potentially compatible with interstitial lung disease and are identified in patients without a previous suspicion for interstitial lung disease. The definition, and I would like to emphasize this, is purely radiological and based on an incidental identification of these findings, most of the time on imaging that is not necessarily dedicated to the chest, but partial imaging of the chest, that's in abdominal CTs or neck CTs. The big question is how to differentiate interstitial lung abnormalities versus subclinical ILDs. This is not a radiologic definition. It's a clinical evaluation, the one that makes that differentiation. So, based on imaging, we have two different types of interstitial lung abnormalities, and these will be defined by the description and characterization of those findings on imaging. So, they can be nonsubplural interstitial lung abnormalities or subplural ILDs. This subplural distribution is important because these are the cases where we really need to look close and identify fibrotic features of those abnormalities. So, according to the presence or absence of those, we can define those subplural ILDs as fibrotic and nonfibrotic interstitial lung abnormalities. So, there are some requirements on imaging that are actually very important. As you know, the assessment of interstitial requires very thin sections. So, ideally, we will have thin sections, less than 1.5 millimeters. And the prone and expiratory images are important to differentiate subtle findings on imaging, and we'll see some examples. So, this is the reason why an HRCT protocol is indicated after the interstitial lung abnormalities identified in a non-HRCT or a partial image of the chest. Obtaining those thin sections will allow to create an appropriate description. We will use terms to define the axial and craniocardial distribution of the abnormality, assess what's the volume of the lung that is involved, knowing that it's an ILA. The definition implies more than 5% of any lung zone involved. Again, the description and specific terminology of those imaging features will allow the characterization and the definition of those categories of ILA by imaging. So, as part of the requirements, the HRCT protocol involves prone imaging. This is an example of dependent, and I hope you can see my mouse, my indicator here, dependent supra-reticulation on the supine imaging that completely resolves on prone imaging. You can actually see now that the now-dependent portion of the lung in the right lower lobe has those dependent atelectasis. As we go down, you can see how in the posterior costoprenic angle, what looks like an interstitial lung abnormality completely resolves on prone imaging, and this is why it's important to complete the imaging and clearly find the presence or absence of supra-reticulation, in this case, being dependent atelectasis. So, when we use that terminology in the description, we will refer to the axial distribution, and this specifically refers to the presence or absence of supra-abnormalities. As we saw, this is the definition of the supleural fibrotic or non-fibrotic, so that distribution will be relevant. It's also one of those indicators of progression, as we will see later. And as for the cranial-caudal distribution, we will divide the lungs into upper, mid, and lower zones. The upper zone, the upper limit is the aortic arch, and then the inferior pulmonary vein is the one that defines the mid versus lower zone. So, in this way, the distribution in the description should have terms such as prebranchial ocular or supleural in the axial distribution or cranial-caudal information as for upper, mid-lung, or lower zone predominance of the device. It is important to create a specific description, and your radiologist should include these terms and descriptions, as imaging findings have different implications. And as we saw, the presence or absence of these findings is what will help us define the type of ILA that we're dealing with, and at the same time, create a prognostic significance of these findings. Terminology that you will hear, and we will see in these reports, non-existent cysts, ground glass or reticular abnormalities, architectural distortion, traction bronchiectasis, and honeycombing, the last three being of specific importance as their definition, they imply interstitial fibrosis. So, some examples of these ILA findings, the non-existent cysts, these are examples of cystic spaces with a supleural distribution. These are not findings that are suggestive of emphysema. Emphysema has specific imaging findings that are not present here. We do not see a central dot sign. These are very well-defined and kind of irregular cystic spaces that are associated with that ocular or supleural ground glass. There are no traction bronchiectasis, and this is what we can see as non-existent cysts. In some patients, specifically smokers, and that's an incidental finding. The non-supleural interstitial lung abnormality does not have fibrotic features, and usually these non-supleural ILAs do not progress. Again, most of these observations have been in the smoking population. We can see in this case some central ground glass that is very subtle. In this case, the patient also has smoking-related respiratory bronchiolitis, and we do not see features of interstitial fibrosis. Again, the supleural lung is normal. More frequently, we see a supleural distribution, and this is an example of a non-fibrotic interstitial lung abnormality with supleural distribution. Three different axial images that point towards those focal areas of supleural ground glass. Here we can see on the right, in the right upper lobe, some supleural bands, which are these horizontal bands that we see associated with these findings. Then, in the pulmonary basis lower lobe, we can see the ground glass without any feature of fibrosis, meaning we do not see traction bronchiolitis, honeycombing, or intralobular septal thickening or architectural distortion. This is an example of a supleural non-fibrotic ILA. Another example of non-fibrotic interstitial lung abnormality in a different patient, you can see those focal areas of supleural articulation without traction bronchiolitis or honeycombing. Again, with an apical component, we can see some components in the upper lobe, and as we go down to the lower lobe, we can see a basilar predominant. So that description of the chronic cartilage distribution of the findings will be relevant in this case as well. Non-fibrotic ILAs with that supleural distribution, really important to get the prone imaging. You can see how different from the dependent atelectasis in this case, those abnormalities persist during prone imaging. Again, no fibrotic features in this case, and it's defined as a supleural non-fibrotic ILA. To show you progression and moving towards the fibrotic ILA, this patient presents in 2019 with a non-fibrotic interstitial lung abnormality, some focal areas of supleural ground glass and minimal articulation, no traction bronchiectasis. You can see involvement of both upper lobes in these images. In 2020, the interstitial lung abnormality persist and is slightly more conspicuous. But again, in 2020, we still don't see any traction bronchielectasis, although in 2022, the abnormality progresses. It's much more conspicuous. As we go down in the basis, we start seeing some architectural distortion, mainly left upper lobe in the image on the right, this 2022, and then persistent ground glass and focal areas, so very subtle traction bronchielectasis that are much more evident and conspicuous in the pulmonary basis. In this case, the supleural distribution of the abnormality also has a basal predominance in that coronal-caudal description, and it's an example of a fibrotic interstitial lung abnormality that first started as a non-fibrotic IOS. Another example of a supleural fibrotic interstitial lung abnormality in a different patient, you can see those focal areas of supleural ground glass in both the lower lobes. This is in 2021, minimal traction bronchielectasis, and then in 2022, after the patient is worked up, this is a fibrotic NSIP, and you can see a sample of the progression of coronal-caudal, ground glass, supleural ground glass and architectural distortion with extensive traction bronchielectasis. So the risk factors and follow-up of these interstitial lung abnormalities is something that is under development. As we mentioned before, most of these ILAs are detected in an incomplete study, so the indication is to complete a high-resolution chest CT to further characterize the interstitial lung abnormality, as well as create a clinical assessment. An important thing to remember are the risk factors for progression of interstitial lung abnormalities. So we have clinical risk factors that you are very aware of and the radiological risk factors, and that combination in the clinical assessment and imaging assessment is what will lead the imaging and clinical work. Specific radiologic risk factors, as we gave some examples before, in the non-fibrotic interstitial lung abnormalities with a basal and peripheral or supleural distribution is one of those risk factors for progression. Fibrotic interstitial lung abnormalities with basal and peripheral predominance that we have without honeycombing, which is what we find as a probable usual interstitial pneumonia pattern and the usual interstitial pneumonia pattern, which is the same findings associated with honeycombing. So these three, as you can see, the common factors, the basal and supleural distribution, is what will be relevant in the description of the imaging type. So the appropriate timing for follow-up is actually unknown. There are some guidelines based on the position paper from the Fleischner Society. I'm going to magnify this portion of the algorithm. Once we identify the interstitial lung abnormality, we assess if there are any radiological, clinical features that increase risk of progression based on those factors that I just mentioned. If there are some of those identified, then the active monitoring will be indicated. This implies risk factor reduction and we assess both clinically and radiologically with radiological follow-up. The repeat CT is recommended at 12 to 24 months or sooner, either it's clinical or physiological progression of the end. So the real question here, when we see patients actually pretty frequently is, is this a real interstitial lung abnormality and what else could it be? So this is just an example. I think the topic which will be better understood once Jonathan gives his portion of what is not considered an ILA and you can have that comparison of imaging price. So this is what I have for you all today. Thank you so much for your time and thank you for being forgiving and seeing this lecture recorded. I hope you enjoyed the session. Thank you. Hi. My name is Jonathan Chung. I'm a radiologist. And so I'll be giving the last session here. And so this one will be less technical. This will be kind of easier on your brain. Because the first two talks, there's a lot of data there. It's a lot to get through. This is more just kind of a wrap things up, I think. Because I think that you kind of, you guys now know what interstitial lung abnormality is. You know what the definition is. There are some other definitions out there. But you get the gist. Bottom line, it's like schmutz in the lungs. We don't know if it's stable scarring or if it's a progressive pulmonary fibrosis. That's really the gist of what interstitial lung abnormality is. And I'll say, it's the big question in interstitial lung disease and fibrotic interstitial lung disease. Whoever can answer that question and figure out which of these ILAs are going to be stable over time and which ones are going to actually progress, that person is going to be given many slaps on the back. Because we need to figure that out. We have treatment now. From University of Chicago, I'm a chest radiologist. But I like to call myself an ILD radiologist. I went into chest imaging not for the lung cancer or metastatic disease workup. Truly was for pulmonary fibrosis. That's the reason why I get up in the morning. These are my disclosures, none of which are directly pertinent to this talk. So what are we going to do? I'm going to show you some examples. I'm going to show you just some cases. And we're going to see, and I've blinded myself to these as well. So I purposely put this together quite some time ago now. So I can't remember which of these interstitial lung abnormality cases are ones that progress over time and which ones don't. So I probably will embarrass myself in front of you guys by trying to predict. But we'll embarrass ourselves together. We'll make it fun. But so it's not going to be quite audience response. But I'm going to show you an image. And then I want you guys, in your mind, to try to figure out, do you think this will progress or this will stay stable? And all the follow-ups are around two years. Admittedly, some of them are like a little less than two years. Some are a little more than two years. But I tried to, so we can compare apples to apples, get them all to have a baseline CT and then a follow-up about two years later. All right, before we do that, we do want to talk about what interstitial lung abnormality is not. So you guys may see this area of some ground glass opacity and reticulation in the right lower lobe adjacent to this osteophyte here. You may see, well, that seems to fit the definition of interstitial lung abnormality. There's reticulation, there's ground glass opacity there. Maybe it's non-subplural, that kind of thing. But this is specifically called out as not being interstitial lung abnormality because we know that these cases, these areas of fibrosis, these really actually are areas of fibrosis. If you biopsy these areas, these come back UIP. It's probably why you don't want to biopsy these areas adjacent to osteophytes because now you're stuck, right? You're like, oh crud, now what do I do, right? So this is a classic location for osteophyte fibrosis. And so almost always occurs in the right lower lobe. And why is that? It's because it's kind of hard to see here, but most people have a left-sided aortic arch, right? Like 99% of us have a left-sided aortic arch. And so the pulsations of the aorta keep osteophytes from forming on the left side of the vertebral body, believe it or not. That's what's happening. And so there's no osteophytes on the left side of the body. So the left lower lobe is essentially spared. But in the right lower lobe, we oftentimes will get these big osteophytes, especially older patients. And as the lung hinges over that area of osteophytosis repeatedly on inspiration and expiration, it actually causes fibrosis and can actually progress over time locally in that right lower lobe. But it's not a progressive pulmonary fibrosis you would think of like UIP or hypersensitive pneumonitis or non-specific other fibrotic diseases. These are ones that are almost always clinically... I'm going to say always. I don't know if you've ever heard a radiologist say always rather than hedging, but I'm going to say always. It's always clinically insignificant, okay? So these are ones that are not included in interstitial lung abnormalities. So if you see something there, ignore it. Do not work it up. Definitely don't biopsy. It always comes back to UIP. I've been involved with two cases now where they did a surgical lung biopsy, came back to UIP, came to our ILD clinic, and it's a huge headache because the clinician at the outside center, like now they think the patient's IPF. They've already told the patient they have IPF, right? And then everyone Googles IPF and they're like, oh my gosh, three years survival, this kind of thing. And now, you know, so you go down this, unfortunately, this vortex of badness and they never had to go down that vortex, right? So anyways, ignore these, okay? Osteofibrosis. I wish they would call it something else because it makes it sound like it's actually like real progressive pulmonary fibrosis. These are two other cases where just, I know this is a little more than we would call interstitial lung abnormality, and this looks like frank pulmonary fibrosis to me, but just bear with me. Pretend like there's just a little bit less reticulation there. This case of asbestosis is nice as well. We see just a little bit of reticulation here. If the patients didn't have any history of anything, if they just came in and didn't have risk factors for pulmonary fibrosis, we probably would call both these cases interstitial lung abnormality, okay? But remember, if they are at risk for pulmonary fibrosis, like someone with systemic sclerosis or someone who has a clear asbestos exposure, I don't know if you can make out the plaques there. No, it's not projecting that well, but there are some plural plaques here as well, bilaterally. You see those plural plaques, bilaterally, this patient clearly has asbestos exposure and any pulmonary fibrosis or reticulation at all, you have to make the assumption that it is asbestosis, okay? So patients who are at risk, occupational, connective tissue disease, you see reticulation, you see something that looks like interstitial lung abnormality, especially the lung bases, you don't call that ILA. You just say it's very, very early pulmonary fibrosis or interstitial lung disease, okay? And then this is the last thing. So technically this is, so if you, if you, eight years ago, or maybe five years ago, you would have called this interstitial lung abnormality. Any radiologist would see this and say, okay, well that's interstitial, right? It's like, get these little lumpy, bumpy central lobular nodules, this is going to interstitial ground glass, central lobular ground glass nodularity, and they would have just labeled it like that. But now that we're using that term much more specifically, that ILA, I think the community, including radiologists, we have to be more careful in how we sort of use that terminology. And so we are not calling this interstitial lung abnormality, because when someone says interstitial lung abnormality in the ILD community, what are we sort of implying now? We're saying this is potentially pulmonary fibrosis, progressive pulmonary fibrosis, but we don't know. We don't know what this is. We know that this central lobular ground glass, schmutzy stuff within the lungs, this doesn't turn into pulmonary fibrosis. Almost always this is going to be, this is going to be something like smoking related. We know that respiratory bronchiolitis is extremely, extremely common in patients who smoke. I think the data says that if you have more than 10 pack years smoking history, and you were at a biopsy, all patients who have more than 10 pack years smoking history, the rate of respiratory bronchiolitis in lung is close to 100%. And so on CT, we see this all the time. And it's really just, we have started to undercall it, because we see it so often, unless it's floored like this. But why am I showing this case? It's because if you see this, you shouldn't really be thinking about this as a quote-unquote interstitial lung abnormality as we're defining it now. Okay, very important to use that terminology the way it should be used. Anytime we're, anybody in the pulmonary community uses interstitial lung abnormality, we should really be using it in that setting. This is something that might be progressive pulmonary fibrosis, but we don't know. That is the gist of what ILA is. Okay, all right, here are the cases. I have talked way too much, even before I've shown you the cases, but I will leave time for questions, okay? We're good? Okay, excellent. I can, obviously, I editorialized too much, as you guys probably have already figured out. So this patient has something that looks like an adenocarcinoma in the left upper lobe, okay? But in addition, they had this interstitial lung abnormality of the lung basis. A little bit of ground glass opacity, a little bit of reticulation, and a little bit of traction bronchiolitis is there in the right costophrenic angle. I don't know if you guys can make that out, but there is a little bit of traction changes there. So if you ask me, right out of training, if I could figure out which patients had just mild scarring at the lung basis or had early pulmonary fibrosis, I would have said there's no way. But if you asked me five years ago, I was probably overconfident, probably that whole sophomore thing. Five years ago, I would have said, almost all cases, I could figure out which cases are progressive pulmonary fibrosis and which cases are scarring. If you ask me now, I'm just going to throw up my hands and be honest with you, I have no idea. Like a lot of these cases, I feel like I'm just flipping a coin. So I would guess this would be a progressive case because it has a little bit of traction bronchiolitis at the lung basis. And indeed, we do see a little bit of progression here two years later. You see how this is a little worse, a little more reticulation than here. But to be honest, I was guessing. I really don't know. Okay. I don't know. Anyone, anyone, anyone? Actually, you know what, I should, I should, next one, I'm going to actually have people raise hands. Okay. So here's another case. Okay. So in the, in the posterior aspect of lungs, we see some linear opacities, a little bit of reticulation, maybe a little bit of traction bronchiolitis at the lung basis back in here. So in your mind, try to figure out, do you think that this is a case of that's going to progress after two years or one that's going to be stable? And I'm not going to make you guys raise hands. That's so, I don't know, 1996. I should have used the audience response like Sean did. Yeah. But, but in your mind, in your mind, just so, and so this is what I'm thinking. I'm thinking it's got the bronchiolitis. So whenever you see things that are like, frankly, fibrotic, like things like traction, bronchiectasis or bronchiolitis, supplemental honeycombing, architectural distortion, hard to define architectural distortion, right? It's like, well, what is architectural distortion? It's kind of like, you know, when you see it, it's like one of those things where things are kind of jagged, it's kind of pulled. There's a little bit of architectural distortion there. I thought that this was going to be a case of progressive disease. And two years later, it actually gets better. Right? Yeah. So, so it's not even stable. It just gets better. So a lot of that probably was atelectasis. But as you know, a lot of these ILAs are going to be detected on non-HRCT images. And I, so as a purist, as a purist, you know, when you, when you call interstitial lung abnormality, you're supposed to do on a really high quality study, HRCT, but come on, that's not, that's not reality, right? Most of these can be detected on CTs that are just of like kind of middling quality. And even if they're good quality, oftentimes not going to have prones. And most centers that I talked to, they don't even do prones routinely for their HRCTs, right? So that is a problem. But you know, I would have gotten this one wrong. It's hard. Another case here. This I think is, yeah, these are, this is a little thicker cut. So that's a nice segue, nice transition. So this is just practicality. This is just real life. Not all CTs are going to be beautiful, but there's clearly some interstitial lung abnormality here. I would have thought this would be stable. This just looks like, you know, it's a little ground glass, a little reticulation in the posterior lateral aspect of the lungs. Two years later, it gets worse, right? So I'm just, I don't know. I'm, when I'm batting 333, I think that's pretty good for MLB, but not good for radiology, unfortunately. And on, on Corona, it's even more floored, the amount of worsening over the course of two years. Okay. Another one here. This is very mild reticulation. Okay. I'm pretty sure there's going to be stable. Pretty sure. I don't know what you guys think. I don't see any evidence of architectural distortion, no traction bronchiectasis, no honeycombing. It's worse. Turns into UIP for two years later. Gosh, darn it. Right. So like, yeah, fooled me too. Another one here. I won't belabor the point here. This one actually is a five millimeter thick cut reconstruction. Anyone doing five millimeter thick reconstructions on CT? No one want to fess up. Yeah. So I, I still see five millimeter thick reconstruction down the, some community places. We should stop doing that, right? Three millimeters is probably where we want to be for like normal chest CT. And obviously for HRCT, we want to go thinner, one millimeter or less. Okay. So it's five millimeters, but even on five millimeters, we see a little bit of schmutz-y stuff here and then, huh. Stone cold stable. Okay. So this one, I, I really wouldn't have been able to figure that out. Another case here, more of the same. This one, I'm sure this one's going to get worse. Look, there's a little bit traction bronchielectasis here. Looks coarse. Feels like maybe there's some architectural distortion. In this case, I'm right. Why not? Am I right? Yeah, no, I'm right. Sorry. I'm blinded myself, right? Blinded myself so well, I can't even tell what's going on in the follow-up, but clearly there is worsening on the second scan. So there, there I'm doing pretty well. Another case here, more of the same. Oh, so has this gotten worse? Let's raise the hands. Who thinks this is worse? Where do you think it's worse? Anyone want to venture a guess? Which lung? Just yell that out. So, so the, the reason why I include this case, so let's say the, let's say the left lung is stable. Okay. These are, these are reconstructed slightly differently. This is a low dose. That's why it looks so motley compared to this one. Let's say the left lung is stable. The right lung, you, you agree there's more stuff here, more reticulation, some traction bronchielectasis as well. That looks bad, right? Looks fibrotic. But where is it? Osteofibrosis, right? Yeah. So you, so if it gets worse that you got to just exclude, I know it looks bad. I know your gut wants to say it's worse, but you got to say, you know what? I'm going to literally not look at that portion of lung when, when it comes to pulmonary fibrosis. I mean, there might be cancer in there. So you want to look there for cancer, right? I mean, you know, don't, don't be silly. Right. But in terms of pulmonary fibrosis, I progressed pulmonary fibrosis in the ILA. Just ignore it. Okay. It looks worse. I mean, it is worse. Clearly it's worse there, but you know what? It's osteofibrosis. We know osteofibrosis can get worse. Okay. And definitely, please don't buy us data. Okay. Please do not. So what's the point of this exercise? So it's like, Jonathan, why'd you spend 10 minutes showing me these cases, showing that you are essentially inept at trying to figure out which of these ILAs are going to progress and which ones are going to stay stable. It's because the old qualitative assessment that we've used for CT, try to figure these things out. It doesn't work. Right. And I, I, so I've been out in practice 12 years now and I love ILD. Like I think about ILD almost all the time. I'm not, you don't even believe this. It's the, it's the truth. My wife hates me. Right. Right. I'll talk about UIP, NSIPH. My kids know UIP, NSIPHB. They know, they hear this all the time. Right. But it doesn't work. Right. So what do we do? You know, I think the, maybe someone in this audience is going to figure it out, but we have to figure this out. And the old methodology of just radiologists or pulmonologists, just qualitatively assessing CT scans, it's not going to give us the answer. What's going to give us the answer? Probably, I think probably radiomics and artificial intelligence. And so, you know, this is just like, we throw a bunch of scans, a bunch of CT scans that are, that have answers into some sort of algorithm. It creates a model and then it analyzes CT for us. Right. Because the problem with the radiologists and the pulmonary community is that we're biased. We know we know, and we think we know the answer already. And so it's hard for us to kind of open up our minds and look at everything else. Because maybe the answer is actually not in the lungs. Maybe it's in the mediastinum. Maybe it's in the pericardial fat. Right. Maybe it's in the shape of the lung. Maybe it's in differences, inspiratory and expiratory. We don't know. So we need some sort of artificial intelligence tool, which can answer that for us in an unbiased manner. Right. Because there, you shove the data in and then it gives you an answer. Other things that we can try to do are maybe PET-CT, not FDG PET-CT, but other types of tracers and molecular imaging. Something exciting actually right in the forefront I think is xenon imaging. I don't know are there, Joe Mamarapolo from Duke is here. Right. He's another radiologist. He's actually doing pretty amazing work with xenon imaging. Right. Xenon ventilation imaging, which seems to be like early marker for, can be a marker for pulmonary fibrosis. And so maybe that's what we have to do. We have to actually think outside the box to try to identify other biomarkers for pulmonary fibrosis. And then I mentioned the non-pulmonary biomarker imaging. I'm going to, I'm going to skip this for now because I do want to leave time for questions. So what do I do with these ILA type cases? Again, I think we've gone through all this. So I'm not going to actually go through all these. The one thing I didn't want to talk about is actually CT scans. So right now the Fleischner guidelines suggest you do a CT follow-up every 12 to 24 months. I would do them every 12 months. I think 24 months is a bit much. Right. I really do think so because you don't know what sort of the, the trend of these, these worsenings of fibrotic cases. So I think waiting 24 months might be too much, especially if guys with maybe a higher cardiorespiratory reserve, maybe their fibrosis get much worse before they're symptomatic, especially guys who maybe aren't like athletes, that kind of thing. And so I've gotten pushback on that though, because people say, well, what about the radiation dose? Okay. There is no evidence in adults, you know, kids are different, right? You know, kids are not just little adults, right? In kids, you want to avoid radiation, right? There, you know, there are, there are a future, you know, we should treat them like the treasures they are, right? And there are evidence to suggest that in kids, you can actually induce hematological malignancies on a population level. So kids, completely set them aside, please, right? Do not irradiate them. MRI ultrasound as much as possible, but adults, no evidence, no evidence to suggest that medically indicated radiation leads to increased malignancy. Anyone here from Denver? Denver? Oh, there we go. I used to work at National Jewish for, for five years. It was, it was awesome. Denver is great. So Denver, the people there, everyone has a population, they exposed to 30 to 40% more radiation than everyone else in the U.S. because they live on this big slab of granite, okay? And apparently there's all this radiation that comes out from it. And it sounds bad, but on a population level, if you look at the cancer incidents in Denver versus other parts of the country, it's actually lower on population level, right? So it shows you that just because radiation is more doesn't mean it's a bad thing. Also, but you know, that is confounded data, right? So everyone in Denver has a six pack and runs a marathon, right? Everyone, right? Except me, right? It wasn't me. I wasn't one of those people, but yeah, she looks actually very fit, right? So she probably is one of the marathon runners, right? So that, so it is a confounded data. But bottom line is radiation should not be of concern. If the patient needs an annual CT scan, go ahead and get it. The radiation dose for average chest CT is now down to two or three millisieverts. I don't want to say it's almost nothing, but you should consider it almost nothing, right? Even if you, you told me, you know, the patient's like 45 years old, she's going to need 40 CT scans in her life, I'd say, okay, do it, right? If it's medically indicated, there's no data to suggest that that patient is at risk for increased malignancy. All the data that we have for increased malignancy risk is based on atomic bomb survivor data, which is crazy, right? Because they have exposed like so much radiation dose and we're trying to extrapolate that linearly back, which obviously is wrong, right? You know that that's just in your gut. That's not the way to do it. So anyways, thank you very much. This is, these are my kids.

interstitial lung abnormalities

ILA

CT scans

disease progression

mortality rates

Fleischner Society

guidelines

radiomics

prognosis